Preferred Citation: Zelin, Madeleine. The Magistrate's Tael: Rationalizing Fiscal Reform in Eighteenth Century Ch'ing China. Berkeley:  University of California Press,  c1984 1984. http://ark.cdlib.org/ark:/13030/ft4k4005k7/

The Magistrate's Tael Rationalizing Fiscal Reform in Eighteenth-Century Ch'ing China Madeleine Zelin UNIVERSITY OF CALIFORNIA PRESS Berkeley · Los Angeles · Oxford © 1984 The Regents of the University of California

Preferred Citation: Zelin, Madeleine. The Magistrate's Tael: Rationalizing Fiscal Reform in Eighteenth Century Ch'ing China. Berkeley:  University of California Press,  c1984 1984. http://ark.cdlib.org/ark:/13030/ft4k4005k7/

To my parents,
Murray and Rita Zelin

Preface

Not long ago, all courses in modern Chinese history in the United States began with the firing of the first cannon of the Opium War. Even for Chinese scholars, the year 1840 has become the dividing line between ancient and modern in the evaluation of their country's past. The indigenous evolution of the Chinese polity has been obscured by a single-minded focus on its failure to respond to the challenge of Western imperialism. Recent work on population growth, commercialization, and the development of handicraft industries and agricultural specialization has shattered an image of a static late-imperial economy. But in the political realm, the spectre of the decaying Ch'ing empire, so deeply imprinted on the minds of nineteenth-century chroniclers, still survives. Although we no longer speak of "unchanging China," the paradigm for understanding China's early-modern experience remains a weak, corrupt imperial state, barred from innovation by an archaic philosophical heritage. As a result, we often overlook the complexity of that experience and deny ourselves the opportunity to explore the real roots of China's nineteenth- and twentieth-century decline.

This book is about China before the "coming of the West." It is also about corruption, but more important, it is about change. The first century of Manchu rule in China was more than just a page in the history of barbarian adaptation of Chinese political forms. It was a period of vigorous reform that had important im-

plications for the potential development of a strong, modern Chinese state.

The main arena of reform in the early Ch'ing was fiscal policy. Building on China's preexisting tradition of bureaucratic rule, the new Manchu dynasty endeavored to use fiscal reform to enhance the power of the monarchy and strengthen the institutions of centralized government. They drew a clear distinction between the inner and the outer court that was reflected in the separation of the imperial purse and the state treasury. Efforts to forge a direct relationship between subject and ruler were accompanied by policies designed to diminish the power of the gentry and local strongmen whose mediating role under former dynasties had weakened the authority of the state. In order to improve tax collection, tax assessment was simplified and the land and head taxes were merged. At the same time, the Ch'ing clearly delineated which revenues were to be the province of local and of central administration, and they established a system of annual accounting to monitor revenue collection and utilization and ensure central-government supervision of all state income.

Taken together, these measures constituted a major step forward in the evolution of China's administrative apparatus. However, they could not contend with the underlying weakness of late-imperial finance. Corruption and tax evasion were a continuous threat to the fiscal stability of the early Ch'ing state. Chinese rulers in the seventeenth and eighteenth centuries faced many of the same problems as their contemporary counterparts in Europe: a weak agricultural fiscal base, poorly articulated distinctions between public and private revenues, and intense competition, both within and without the bureaucracy, for the limited surplus product of the realm. The existence in China of a mature bureaucratic system of rule, as well as the innovations introduced by the Manchu dynastic founders, exacerbated these problems by demanding morality in fiscal administration without providing sufficient funding at the local level to accomplish this goal. The great achievement of China's early-eighteenth-century reformers was the creation of a fiscal system which not only answered the needs of traditional political economy, but had the potential to develop into a system of government finance suited to the requirements of a strong centralized state.

The culmination of early Ch'ing fiscal reform was the "return of the meltage fee to the public coffers" (huo-hao kuei-kung ), imple-

mented during the reign of the third Ch'ing emperor, Yung-cheng. Officials in each province were authorized to collect a fixed percentage surcharge (huo-hao ) on all regular land and head taxes remitted to the central government. This huo-hao was then retained in the province of origin to provide officials with substantially increased salaries (yang-lien ) as well as "public-expense funds" (kung-fei ) with which to carry out certain administrative responsibilities and projects of benefit to the local area.

Although the concept of huo-hao kuei-kung was simple, its rationalizing impact on the structure of Chinese fiscal administration was enormous. At the level of intrabureaucratic relations, the provision of adequate funds for local officials eliminated the need for institutionalized government corruption. The existence of a reliable source of funds for public expenses made it possible for officials to undertake the budgeting of local expenditure and engage in the long-range planning of local construction projects. Moreover, with income guaranteed, local governments were able to take on as their own responsibility many of the services and public works either neglected or relegated to the private sector during the last decades of imperial rule.

Many of the problems addressed by this new fiscal system were the legacy of the preceding dynasty. The Ming had seen several efforts at cleaning up corruption, but, except in the case of the "Single Whip" reforms, no attempt was made to restructure the fiscal apparatus of the state. The monetization of taxation, which was a major goal of the Single Whip reforms, became an important factor in the crisis that made reform necessary in the early Ch'ing. However, differences in the operation of government during the two dynasties largely account for the latter's ability to introduce such dramatic changes in fiscal institutions.

Although the bureaucratic structures of the two regimes were similar, ideological appeals seem to have played a far greater role as the basis for decision making as well as in the assessment of official behavior during the Ming. Even more important was the role of the emperor in the two regimes. The Ming dynasty is famous for its elimination of the post of prime minister, an act that is generally seen as vital to the rise of autocracy in the late imperial period. Nevertheless, most late Ming rulers were little more than puppets of those members of the literati who manned the upper echelon of the central bureaucracy.^[1] On the other hand, the early Ch'ing rulers, as aliens, were far less fettered by traditional modes

of interaction between emperor and court. Moreover, the Manchus brought to China new notions of imperial authority which in practice placed the emperor in a much more central position than he was accorded by traditional Confucian notions emphasizing the ritual centrality of the monarch. The direct role played by the early Ch'ing emperors in the formation of policy often contributed to the introduction of new ideas and new techniques in government. By the early eighteenth century, that role was reinforced by new institutions such as the secret-palace-memorial system, which both improved the emperor's access to information and his control over administrative affairs. Finally, the Ch'ing bureaucratic apparatus itself was far more centralized than that of the Ming, a fact that facilitated empire-wide reform and imperial direction of its implementation.

These conditions made possible the rational approach to solving China's fiscal crisis that was adopted during the 1720s and 1730s, but they did not guarantee that such bold measures would be undertaken. Each of the emperors during the so-called High Ch'ing was an able and dedicated ruler. However, differences in their personalities and in the political climate during their reigns had a profound effect upon the way in which they dealt with the problems of fiscal administration. The Ch'ien-lung emperor, last of the High Ch'ing triumvirate, deliberately sought a balance between the overly lenient policies of his grandfather, the K'ang-hsi emperor, and the excessive harshness that marked the reign of his father, the Yung-cheng emperor.^[2] It was not merely style that distinguished these two great rulers. K'ang-hsi reigned in an environment in which Manchu legitimacy was not yet established. His notorious refusal to attack corruption may have grown out of the necessity to placate a predominantly Chinese bureaucracy not yet reconciled to Manchu rule.^[3] His son no longer had to fear recriminations from officialdom or the literati. By the time the Yung-cheng emperor took the throne, China was united and at peace, and almost eighty years of Manchu rule had left no other realistic focus of loyalty for an elite whose status depended as much on imperially conferred ranks and degrees as on local wealth and influence. The Yung-cheng emperor was a tough and pragmatic man, but it is unlikely that he could have taken the steps that he did to rationalize administration and combat corruption and gentry privilege had he ruled at an earlier time.

In part, the peace and prosperity of the High Ch'ing must be attributed to the huo-hao kuei-kung reforms. Unfortunately, despite its contribution to China's early modern development, huo-hao kuei-kung ultimately failed. By the nineteenth century, China was indeed a country ravaged by corruption and torn by centrifugal forces which the government was increasingly powerless to resist. Nevertheless, the nature of our inquiry into modern Chinese history is changed dramatically when we realize that the "traditional" Chinese monarchy, which appears so resistant to change in the century of Western ascendency, was not always so. We must learn to look at China in the late-imperial period as a dynamic state, struggling to devise its own formula for rational and efficient bureaucratic rule. If the huo-hao kuei-kung reforms were a failure, then we must ask why.

Nineteenth-century Chinese intellectuals sought the roots of bureaucratic corruption in the declining moral qualities of China's official class, and modern critics have tended to follow suit. Faced with the evidence in government statutes of strong sanctions against officials found guilty of fiscal malfeasance, scholars have postulated that the government was too weak or too lazy to enforce its own regulations, or that the force of custom was too great for any agent of the throne to overcome. In the end we are left with the unsatisfying notion that China was corrupt because China was corrupt. Such an explanation, or lack thereof, provides us with no basis upon which to judge the actual capacities of the late imperial state. Even more important, it allows us neither the data nor the analytical framework within which to compare the problems faced by Chinese dynasties with those of other premodern societies, or with those of Chinese governments in the twentieth century.

Personalities did, of course, play a role in the demise of huo-hao kuei-kung , as they did in its implementation. The Ch'ien-lung emperor, more than any of his predecessors, was a product of a Confucian education and a prisoner of Confucian myths of kingship.^[4] His desire to be judged a benevolent ruler led him to approach Yung-cheng's reforms with hesitation. On the one hand, he feared that support of huo-hao kuei-kung might be seen as contrary to the way of the ancient sages. On the other hand, he dared not reverse the policy so painstakingly implemented by his own father. In the end, Ch'ien-lung's efforts to satisfy both imperatives were indeed damaging to huo-hao kuei-kung , but they were not solely

responsible for the reemergence of institutionalized corruption in the late Ch'ing. More important were the constraints imposed on rational fiscal administration by the structure of the late-imperial socioeconomic system. If we are to understand properly the decay of the Chinese empire, then we must examine the impact of that system on the breakdown of imperial control of corruption. The gradual demise of huo-hao kuei-kung provides an excellent example of the limits of reform in the late-imperial period. Its analysis will alert us to the ways in which the government evaluated the task of maintaining adequate levels of revenues and the techniques that it used to do so. If the government did try to eliminate corruption and failed, then a study of the reform process should enable us to comprehend more fully the tensions within the late-imperial polity that led to China's nineteenth- and twentieth-century decline.

Acknowledgments

Many people have given generously of their time and energy to help with this study. I am most indebted to Professor Frederic Wakeman for his constant encouragement, criticism, and insights during the writing of this work and throughout my career. Beatrice Bartlett, Philip Kuhn, Alan Ware, Jonathan Spence, David Keightley, James Polachek, and Wolfram Eberhard have read all or part of the manuscript and I hope the version before them now succeeds in reflecting their suggestions for its improvement.

Funding for this project was provided by the Fulbright Foundation, the Social Science Research Council, the National Endowment for the Humanities, and the Columbia University Council for Research in the Humanities. Most of the materials used were situated in Chinese archives. I wish to express my special gratitude to Dr. Chiang Fu-tsung, Mr. Peter Chang, Mr. Chuang Chi-fa, Mr. Li T'ien-ming, and Ms. Wang Ching-hung of the National Palace Museum in Taiwan and to Mr. Huang Hsiao-ts'eng, Mr. Chu Teyuan, and Mr. Liu Kuei-lin of the Number One Historical Archives in Peking for their patience and assistance. I also wish to thank Professors Wei Ch'ing-yuan, Chiang Kuei-ke, and Ch'en Chieh-hsien for their advice and aid with textual problems.

Finally, I would like to thank my family for their unceasing support of and interest in my work. My parents have never failed in their faith in me, even when I abandoned biochemistry to study

something so exotic as Chinese. My husband, Alan Ware, has shared all my burdens with me, and has been my most perceptive and friendliest critic. And my son, lain, whose birth coincided with the completion of this book, has taught me how precious one's working time can be.

1—
The Roots of Fiscal Instability

In the annals of late-imperial Chinese history, few emperors have achieved the fame of the second Ch'ing ruler, K'ang-hsi (1661–1722). His legendary humanity and benevolence as a ruler were given ample expression during a reign that lasted sixty years. Although the final decades of K'ang-hsi's rule were tainted by one of the most celebrated succession crises in China's long imperial past, he is best remembered as the emperor who brought peace to the land and restored conditions in which good government and the pursuit of learning could flourish. Little known or discussed is the negative legacy that K'ang-hsi bestowed on the empire. Yet the way in which this legacy was handled was to have a tremendous impact on the future of the Ch'ing dynasty and the development of the Chinese state.

When the K'ang-hsi emperor died in 1722 he left a state treasury containing only 8 million taels of silver.^[1] This sum, equal to about 27 percent of the central government's total annual tax revenues,^[2] could in no way be compared with the empty treasuries and royal indebtedness of K'ang-hsi's eighteenth-century European counterparts. Nevertheless, within the context of late-imperial Chinese economic thought such a figure was to be deplored. This was a political system that worshipped the balance of expenditure against

income and considered central-government reserves essential lest man be unprepared for the ravages that nature could bring.

Depleted treasuries were certainly not new to China. In fact, their existence was so common that they played a central role in the traditional "dynastic cycle" explanation of the collapse of imperial regimes. When the talents and virtue of the imperial line were declining, when officialdom was becoming corrupt, and when the pressure of population growth was beginning to strain the resources of both the land and the state, it was expected that the imperial coffers would shrink. The previous Ming dynasty had presented a classic case of fiscal strain, resulting in increased surtaxes, popular resistance, growing military expenditure, and finally, inability to cope with foreign invasion. However, in 1722 the Ch'ing dynasty was not dying. On the contrary, there had been no major internal disorder since the defeat of the Three Feudatories and the pacification of Taiwan in the 1680s. Under the devoted sponsorship of the late emperor, scholarship and the arts were once again approaching the high standards of previous dynasties. For the most part, almost eighty years of Manchu rule had won for the alien conquerors the confidence and loyalty of China's native population, especially the important scholar-gentry elite. To stimulate the economy and rehabilitate the areas devastated by nearly a century of rebellion and warfare, both the dynasty itself and private entrepreneurs were engaged in large-scale land reclamation and resettlement. Reforms were also being undertaken in administration and fiscal policy. The vigor of the dynasty was demonstrated even in the succession crisis, for the empire survived unscathed and even strengthened. The factionalism that surrounded the various princely contenders remained limited to small coteries of supporters and did not spread to engulf the entire bureaucracy. The emperor who emerged from the contest, Yung-cheng, was an independent, energetic, and able ruler whose achievements as monarch far outweighed the questionable means he may have used to accede to the throne. Clearly, this was not an era of imperial degeneration and declining official morale.

The strains on imperial finances had, in fact, been reduced since the early years of Ch'ing rule. At that time, even in areas firmly under imperial control, popular sentiment against the newly imposed barbarian authority continued to be reflected in tax resistance and arrears.^[3] Forty years passed before the whole of China was pacified and could be relied on for regular tax payments to the

imperial treasury. In the areas most affected by warfare and rebellion, notably Szechuan, it was not until the mid-eighteenth century that productivity, and consequently tax remittances, reached the levels realized during the Ming dynasty.^[4] However, none of these problems are comparable with the burdens imposed on early Ch'ing finances by the initial commitments of their conquest. Military expenditures used up most of the state's revenues during the early decades of Manchu rule. According to one estimate, of the 15.7 million taels spent during the first year of the Shun-chih reign (1644–61), over 13 million taels went toward military supplies and soldiers' pay. Furthermore, total expenditures exceeded the annual income of the court by almost a million taels.^[5] Under these conditions, little was spent by government on public works or the general welfare.

State revenues did increase by the end of the Shun-chih period, to almost 30 million taels by Shun-chih 17 (1660),^[6] though part of this new income had to provide for the increased expense of governing an expanded territory. One important element in these expenses was the payments made by the state to support the Three Feudatories which grew out of Manchu efforts to subdue parts of the empire's southern provinces. The Chinese generals sent to accomplish this task established instead quasi-independent satrapies which nevertheless continued to require provisioning and revenue support from the central government. During the Shun-chih reign that grant grew to 5 or 6 million taels per year. By the time of K'ang-hsi this sum had doubled and came to encompass about one-third of the annual income of the Ch'ing state.^[7] Thus, although the dynasty recognized the need to counteract official corruption and tax evasion, the main obstacle to fiscal stability during the years of Manchu dynastic consolidation was excessive expenditure, especially military expenditure. Once peace was restored in the realm, the expectation was that fiscal balance would be restored and that the size of the state's coffers would match the glory of the regime in other spheres.

When the Yung-cheng emperor came to the throne in 1723, that dream of a vigorous and prosperous empire seemed on the verge of realization. More than a generation had passed since a major military expedition had added its burden to the state treasury^[8] and the Manchu court itself was notoriously frugal, being loath to make the same mistakes as its profligate predecessor.^[9] When the new emperor looked at his father's treasury he did not revel in the accu-

mulation of 8 million taels, but was concerned about revenues several times that amount that should have been there but were not. Their absence was not the result of a dynasty spending beyond its means, but of a fiscal administration in which taxes that ought to have been sent to the central government were being dissipated elsewhere. Moreover, the Board of Revenue treasury was not the only one exhibiting shortages. The fiscal independence of the other government ministries in Peking, an inheritance from the Ming, had also left its mark. Even a cursory examination brought to light deficits in the stores of almost every treasury in the capital, in addition to serious discrepancies in central-government accounts.^[10]

In some respects the problem of central-government finances was no more than a reflection of the developing crisis in provincial and local fiscal administration. If the central government was not receiving a large portion of its annual tax quota, it was because these lower agencies of the Ch'ing bureaucracy were not sending it on. Investigations into the state of provincial finances disclosed widespread arrears, some dating back a decade or more. Deficits in provincial and local treasuries were enormous. In Shantung the chou and hsien showed shortages exceeding 600,000 taels.^[11] Shansi equaled that sum merely by taking account of those deficits so old that those responsible were no longer living or so surrounded by corruption that the guilty parties could not be determined.^[12] In Kiangsi, the debts of individual officials ranged from ten thousand taels to as much as forty thousand taels each, and the unassigned deficit raised that province's liabilities by an additional one hundred twenty thousand taels. In Chekiang, the financial commissioner's treasury alone showed a shortage of three hundred thousand taels.^[13] Grain reserves also reflected the disorder in local finances. Granaries which according to the records were stocked to overflowing to guard against famine and to stabilize rice prices were found to be half empty. Some granaries did not exist at all except in the accounts of those officials charged with their maintenance.^[14]

Behind the façade of Ch'ing power and prosperity was a fiscal system whose foundations were extremely fragile. Within a structure that was one of the most rational of its time, a fiscal tug-of-war was being waged between the central government and local governments, and between both and the representatives of the various interests among the local population on whom they depended. That the system had survived at all was a function of the

adaptability of those who participated in it. How much longer Chinese finances could have persisted in this way we shall never know. For in 1723 one of the first acts of the new emperor was to declare total war on deficits and those who caused them. How Chinese finances reached such a state we can begin to discern by looking back at the development of fiscal administration in the late Ming and early Ch'ing.

The Legacy of Late-Ming Fiscal Reform

Like most preindustrial societies, China relied on levies on the produce of the land for the major portion of its tax revenues. Therefore, it is not surprising to discover that China faced many of the same problems encountered by other states attempting to assess and collect a reliable income from a tax base composed largely of small agricultural units. Where China differed from both its Asian and its European counterparts was in the degree of bureaucratization of the tax system.

In England, China's chief nemesis in the nineteenth century, the crown had, by the eighteenth century, come to rely primarily on indirect duties on the commercial sector of the economy. More comparable to China are France and Japan. In Japan, direct taxes on the land continued to be the main source of daimyo income, but collection was based largely on community self-assessment of tax quotas set in the han capitals. In France, not only the land tax, but commercial taxes as well, were collected primarily by tax farmers, who had a legitimate role within the French fiscal system. In both instances, the state's main interest was receipt of its quota. In the case of Japan, the solidarity of the village unit and the fact that the rural elite had been largely removed to castle towns seem to have prevented gross inequalities in assessment. These factors, as well as the already high rate of Japanese land taxes and the comparatively small size of Japanese administrative units, can probably account for the low level of corruption evident in the tax system. On the other hand, in France the chronic royal shortage of funds due to the twin demons, waste and warfare, made the French government willing to mortgage its taxes to tax farmers who would advance funds to the court on the strength of future receipts. In none of these countries was a large bureaucracy employed to determine the productivity of the land and who owned

it, and to collect from each individual taxpayer the government's share. Likewise, in each of these three states those who governed below the level of the central government did so largely with their own resources and formed a power base separate from the throne.

In China, the state was also concerned with collecting its quotas, but here the early growth of the bureaucratic state, the demise of a feudal-style aristocracy, and the relative weakness of rural communal institutions were combined with increasing rationalization of the tax-collection process, culminating in the highly centralized fiscal structure of the early Ch'ing. Tensions certainly existed between local, provincial, and metropolitan interests, but at least in theory China's vast territory was administered by a unified civil service, funded by taxes collected by and for the government itself. To a large extent it was the very sophistication of the Chinese bureaucratic and fiscal system that was its downfall.

By the eighteenth century, the main source of funds supporting the Chinese bureaucracy came from the land and head tax (ti-ting ch'ien-liang ). For the most part, this and other direct taxes were collected by county-level magistrates with the aid of clerks and yamen runners who, with the exception of a small number of the latter, were unsalaried, though regular members of the subbureaucracy. Technically, all such revenues were the property of the central government. However, as we shall see, a portion of these taxes was left in the provinces to pay official salaries as well as certain local- and provincial-government expenses. Hence, on paper at least, China had one of the most rational and centralized fiscal administrations in the world.

This had not always been the case. During the Sung dynasty (960–1278) provincial-government expenses were derived largely from the income of official estates granted to the incumbent for the duration of his term in office. By the Ming dynasty this system had been eliminated, but the distinction between public and private in fiscal matters was still unclear. Only slightly less ambiguous was the distinction between central- and local-government revenues. Although taxes were classified as either ch'i-yün (those to be sent to the central government) or ts'un-liu (those to be retained by the locale), these terms were often applied arbitrarily and items covered by each varied.^[15] Even retained income could be earmarked for the purchase of local products for shipment to the capital,^[16] and inasmuch as all deposits in granaries and treasuries were considered ultimately to be the property of the emperor, no funds could

be disbursed without an imperial decree.^[17] At the same time, within the central bureaucracy each ministry received its own income directly from a multiplicity of tax-distribution points throughout the provinces. What seems to have saved this system from collapse was the predominance of taxation in kind and the existence of corvée obligations among the local populace, which provided local government with ample material and manpower resources.

Despite the government's claims to centralized control of tax utilization, the Ming fiscal system as a whole was very decentralized. This contradiction can be chiefly attributed to the founder of the dynasty, who saw the Chinese economy as eternally undifferentiated and nonexpanding. This vision, combined with his distrust of bureaucrats, found its expression in the main vehicles for Ming tax collection and assessment, the li-chia and liang-cbang .^[18] Both of these innovations were designed to protect the rural population from illegal claims on their resources and guarantee the delivery of taxes to the central government. This was to be achieved by reducing the influence of local government in the tax-collection process and by increasing local self-regulation. In the end they did neither, but though they gradually disappeared as formal taxation devices, many of the problems in Ch'ing tax collection evolved from them.

Under the liang-chang system, collection of the grain tax and its transportation were completely removed from the magistrate's control and relegated to local magnates whose personal wealth was sufficient to ensure that taxes would be remitted in full. However, largely as a result of its independence from the local administration, this system provided rich opportunities for the development of tax farming. The establishment of a separate tax-collection structure with its own staff could have provided the foundation for a specialized tax administration. However, the fact that the post of liang-chang was considered a kind of corvée, and was never integrated into the regular administrative hierarchy, deprived it of its rationalizing potential.^[19] By 1421, with the increasing monetization of the economy and the transfer of the capital away from China's riceproducing heartland to Peking, the liang-chang system began to decline in importance. Its tax-collection functions were gradually taken over by the li-chia , and the ultimate receipt and transportation of the taxes to the capital once again became the duty of the magistrates.^[20]

Though the li-chia originally played an important part in the al-

location of corvée, by the late Ming it was primarily a system of rotating responsibility designed to prompt tax payments (ts'uik'o ). Responsibility for the tax quota of an entire artificial "neighborhood" fell to the incumbent holding the rotating post of li-chia headman. Few commoners wanted this job, because failure of one's "neighbors" to pay their taxes in full often resulted in bankruptcy for the headman's household. However, by interposing an informal tax-collection structure between the individual and the state, the li-chia made possible the arrogation of this tax-prompting power by unscrupulous gentry and commoners who saw in it a chance for profit. This was usually accomplished with the tacit consent of the magistrate, whose sole concern was to see that the tax quota was remitted in full. The result was frequent falsification of the receipts given to taxpayers and the collection of illegal surcharges on a large scale.^[21]

During the last few decades of the sixteenth century a series of measures were undertaken independently by a number of officials throughout China, aimed at alleviating some of the weakness of the Ming system of tax collection. Known collectively as the Single-Whip reforms, these measures were designed to simplify and consolidate the diverse taxes levied on the people and to convert payments in kind and in labor service to payments in silver. Unfortunately, a lack of centrally directed coordination of the reforms meant that commutations were often confused and arbitrary and rates varied from place to place. Although tax payments were commuted to silver, the myriad categories into which they had been divided when paid in kind remained. As the censor, Liu Yintung, pointed out in the first year of the new Ch'ing dynasty, "There are hsien with a regular quota of around three thousand taels, but it is divided into over forty designations. One item may require payment of only 1.6 or 1.7 taels, but surcharges of twenty or thirty taels are added on."^[22] Nor was any effort made to consolidate the large number of agencies to which the taxes from a particular place were due.^[23] Moreover, the failure to deal with inequalities inherent in the head tax led to a series of localized efforts at equal corvée which extended from the sixteenth century well into the period of Ch'ing rule.^[24]

What the Single-Whip reforms did accomplish by converting taxes to cash payments, was to subject local finances to a degree of scrutiny that was inconceivable when taxes consisted largely of goods and services. If Ch'ing peacetime deficits were more shock-

ing or attracted more attention than those in the past, it was in part because they were more visible. Still worse was what this visibility meant for the operation and financing of local government. Prior to the Single-Whip reforms, the staffing of local yamen and the provision of stationery, firewood, and so on was carried out through corvée labor and levies in kind. So long as these goods and services were provided in kind, there was considerable flexibility in local fiscal management and it was very difficult for agencies outside the locale to ascertain the extent of such resources or try to take a share. Once they were commuted to silver, the logistical obstacles to commandeering local revenues were overcome. As a result, the central government, the army, and even members of the imperial household began to appropriate the resources once used to operate local administrative units.^[25] The effect was to leave local government with a bare minimum of legal sources of funds at its disposal. Expected now to hire yamen staff and purchase yamen supplies, local and provincial officials gradually worked out a new system of funding, one that was responsible for much of the shortages and corruption that plagued the bureaucracy in 1723.

Bureaucratic Solutions to Structural Problems

Until the eighteenth century, few men in a position to influence imperial policy recognized or were willing to acknowledge that it was genuine need among local- and provincial-level officials that prompted irregularities in the Chinese fiscal system. For many years, temporary but far more immediate causes of shortages, such as war and rebellion, tended to mask the defects in the system itself. Moreover, the success of the local bureaucracy in compensating for its difficulties, at least for the time being, made the paucity of funding available for local use seem less acute than it was. However, most important was the political economy of traditional China. Formulated at a time when taxes were paid in kind, when officials still came from a privileged class with independent means, when informal or customary claims on the labor and productivity of the masses were generally recognized, and when little was expected of government, it now perpetuated myths that no longer accorded with Chinese realities.

Kuo-chi min-sheng were the watchwords of Chinese political economy in imperial times. Translated in modern terminology as

fiscal administration and people's livelihood, they are better understood as a dual exhortation to provide enough taxes for the government to function without depriving the people of the means of subsistence. Because government was seen as existing to transform not the physical landscape but rather the moral landscape of the kingdom, it was not unreasonable that low taxation would be seen as a basic principle of good government.^[26] Rule by moral example was an inexpensive form of government. If the ruler, and by extension his representatives in the field, were virtuous, the populace would follow suit. Small outlays would be necessary for waterworks and for armies directed at external enemies, but large internal peacekeeping forces would normally be unnecessary, the people would be hardworking and content in their occupations, and harmony would reign over the land.

Of course, such ideal conditions rarely existed, but they did produce a formula for evaluating fiscal matters that was accepted by most officials in late-imperial times. Given the low level of taxation in China, it was expected that the people would pay their taxes in full. If they did so, officials need only manage the proceeds frugally for both the state and the people to prosper. If, instead of prosperity, the government was faced by mounting deficits, it could be due only to two factors: either taxes were not being received from the people or they were being received and subsequently dissipated through mismanagement or official peculation.

Beliefs such as these were not conducive to a fundamental restructuring of the imperial system of taxation and finance. Rather, reformers tended to attack each symptom individually in a manner that kept the basic framework of government finance intact. This is not to belittle the importance of such changes, as we shall see in the case of the reforms undertaken in the early Ch'ing.

Officials could and often did use the pretext of taxpayer default to cover a myriad of their own fiscal misadventures. In late-imperial China, however, claims that taxes had never reached the magistrate's yamen were to be taken seriously. In an economy chiefly composed of small landholders, where even the very wealthy generally owned no more than a few hundred acres, taxpayers were particularly vulnerable to the vagaries of nature. Flood or drought could easily deprive whole counties of the ability to pay their full quota. Most dynasties dealt with temporary disabilities of this kind by granting tax remissions and by allowing the less fortunate to spread their liability out over a number of years. This sort of

taxation on the installment plan became extremely widespread during the Ch'ing for dealing with arrears of all kinds.^[27] The frequency of natural disaster and the remissions and relief that they necessitated constituted one of the reasons that Chinese governments would prefer to keep a net surplus and not simply a balance of expenditure and income in their treasuries.

Less easily controlled were instances of deliberate nonpayment. Grouped under the category of tax resistance (k'ang-liang ) in the tax codes, these could range from refusal to pay taxes as an open political protest to secretly bribing yamen personnel to lower obligations or expunge arrears from the yamen records. Once these moves were discovered, however, punishment coupled with retrieval of the arreared sum was the usual prescription. Real inability to pay could be wrongly assigned to this category and dealt with in the same way, especially when an area failed in its petition for disaster-zone status, when tax records were tampered with so that an inordinately large portion of the quota was levied on a locale's weaker inhabitants, or when surtaxes and illegal levies made it impossible for the peasantry to pay its regular tax obligations.

Both of the above instances concern the failure to pay taxes. Far more insidious were cases where the taxpayers had fulfilled their obligations to the state, but the state received little or none of the funds collected. Such cases usually involved embezzlement by intermediaries and were grouped under the heading of pao-lan or tax farming. Tax farmers were local yamen functionaries, members of the local elite, or simply local gangsters who contracted with the peasants to collect and pay their taxes for them. In its purest form this meant that, in return for a small fee, the peasant was saved the time and expense of carrying his taxes to the county seat. However, peasants often were coerced into entrusting their taxes to tax farmers or were prevented from turning in their own taxes by strongmen who, in cooperation with local yamen functionaries, controlled the county's tax-collection centers.^[28] In either case, the tax farmer could keep all or a portion of the taxes collected, depriving the government of its quota and making the taxpayer appear to be in arrears.

Even when the government did receive the people's taxes, this did not guarantee that the full amount would be applied to the functions of government for which they were intended. Between collection and allocation or remittance to the central government there were numerous opportunities for the dissipation of such

revenues through official mismanagement, extravagance, or outright theft by the official or members of his staff. It was to overcome the depredations of these fiscal parasites that many of the early Ch'ing reforms in tax collection and assessment were introduced.

Centralization of Tax Accounting

At first glance there seems to be little to distinguish Ch'ing fiscal administration from that of its Chinese predecessors. The same taxes levied by the Ming and delineated in the Comprehensive Books of Taxation and Services (Fu-i ch'üan-shu ) were adopted by the new government and made the basis of its own finances. The division of taxes into ch'i-yün and ts'un-liu and the relative shares allocated to central and local government were also a readily accepted legacy of the Ming. After nearly a century of war and rebellion, the most important measures undertaken by the Manchu conquerors were those aimed at promoting productivity and agricultural recovery. None of them involved any fundamental restructuring of the fiscal apparatus, but rather they employed traditional methods of expanding the tax base while relieving the tax burdens on the people. Land reclamation was encouraged by means of both tax exemptions and rewards of official rank to those who succeeded in resettling large numbers of people on barren land.^[29] The court itself was careful to exercise austerity in its own expenditure. Imperial benevolence was manifested in frequent tax remissions. And, as the supreme sign of the new dynasty's good will and adherence to the dictates of traditional Chinese political economy, in the last decade of his reign the K'ang-hsi emperor paid tribute to the infinite fecundity of man and the finite area and productivity of land by freezing the head tax at the 1711 level.^[30] Yet behind this façade of continuity there were taking place several important changes that set the stage for the fiscal crisis and the reforms of the Yung-cheng period.

If the fiscal system established by the Ming founder was characterized by decentralization and the substitution of informal for formal mechanisms of control, the system that evolved during the early Ch'ing was the opposite. The first generation of Chinese under Manchu rule witnessed the gradual development of a highly centralized system of tax administration and the evolution of a new, direct relationship between the taxpayer and the government.

As a first step, the Ch'ing founders attempted to resolve much of the chaos inherent in the Ming tax structure by placing all taxes delineated in the Fu-i ch'üan-shu under the direct control of the central government. Instead of having each ministry of the central government collect and manage its own revenues from the provinces, all central government funds were now to be supervised by the Board of Revenue.^[31] Moreover, under the first Ch'ing emperor the Ming proliferation of tax categories was eased and large numbers of taxes were consolidated, with the result that the opportunities for corruption and the confusion they entailed were eliminated.^[32] Further refinement was undertaken by K'ang-hsi, who ordered the simplification of Fu-i ch'üan-shu registration^[33] and the elimination of tax categories for which there was no longer any basis in the Ch'ing economic structure.^[34] All of these measures simplified the collection of taxes and helped eliminate some of the local idiosyncracies in the tax administration that had made central government supervision almost impossible.

Probably the most far-reaching innovation of the early Ch'ing fiscal system was the institution of "annual accounting" (tsou-hsiao ). It is also the best example of the Manchus' more centralizing approach to government. Even before their entry into China, the Manchus had begun to investigate their own finances at the end of every year. However, the comprehensive system of fiscal auditing that eventually bore the name tsou-hsiao chih-tu was equally a product of the changes in fiscal administration carried out during the last years of the Ming and the first years of Ch'ing rule. Given the technology of the eighteenth century, only a tax system based mainly on cash payments, divided into relatively standardized and simple categories and collected and distributed through a clear and direct hierarchy of officials, could sustain detailed supervision of this type.

The basic parameters for annual accounting were established during the early years of the Shun-chih reign. In SC 3 (1646), the emperor ordered the Board of Revenue to investigate each category of taxes received by each yamen in the capital, with special attention to be given to how much the original Ming quotas had been and how this income was now received and allocated. At the same time, the Board was told to supervise similar investigations in the provinces. Their purpose was to find out the extent to which the people were still being burdened by surtaxes instituted to cover military costs in the late Ming, and, in a more general vein, to find

out how much land was currently under cultivation, and how taxes on that land should be collected, remitted, and retained by local government.^[35]

More than anything, the 1646 edict indicates how little the government knew about the practical working of its own fiscal administration. It was not until five years later that Metropolitan Censor Wei Hsiang-shu memorialized the emperor to request that the supervisory functions of the Board of Revenue be regularized through annual reports on finances from the provinces.^[36]

The expenditure of state taxes is controlled by the Board officials and income is controlled by the provincial financial commissioner. The amount of income is not clear. Therefore, the amount of expenditure is not clear. I request that starting in SC 8 (1651), the financial commissioner of each province, at the end of the year, calculate the taxes of the entire province and compile an itemized account to be presented to the governor, governor-general, and judicial commissioner for auditing. On the one hand the governor shall compile a routine memorial with the totals and send it along to the Emperor with a "yellow account" (huangts'e ) for his perusal. Another "clear account" (ch'ingts'e ) shall be compiled and sent to the yamen in the capital for auditing. In this way we can prevent duplicity on the part of the financial commissioners and also check on the irregularities of Board officials.

In SC 9 (1652), the practice of sending annual accounts of paid and unpaid taxes to the Board was approved.^[37] Once the principle of annual Board audits was established, further legislation concentrated on refining the rules governing deadlines for reporting, items to be reported, responsibility for verification of accounts, and punishments for falsification and delays.

Some sense of the nature of these refinements can be obtained by examining several orders promulgated under the K'ang-hsi emperor. In KH 11 (1672), the emperor became concerned about the failure of officials to investigate discrepancies in accounts. He ordered that thenceforth, in the reports of the annual accounts (tsou-hsiao ts'e ), the total taxes reported collected by the chou and hsien must correspond to the amounts found in the lists sent to the people to inform them of their tax obligations. Furthermore, the actual amounts collected, the original quotas, the financial commissioner's totals, and the detailed figures submitted by the chou

and hsien had to tally.^[38] Such data would seem to be a minimum requirement for accurate audit of provincial finances. That the emperor saw fit to issue an edict specifying their inclusion indicates that considerable confusion existed as to what information was required by the Board. Moreover, by requiring submission of accounts at each level of the administration, an attempt was clearly made to institute a system of checks to avoid embezzlement or fraud at any level.

This system of surveillance was further clarified in KH 28 (1689). At that time, regulations were established whereby a governor or governor-general would be punished for failure to memorialize deficits in the provincial treasury and to impeach the officials responsible. An edict was also issued requiring that the governor of each province personally inspect the provincial treasury each year at the time of the annual accounting. When there were no deficits, this fact was to be memorialized and commendations requested for the officials who had filled their quota. In addition, at the end of the year, each prefect was to compile an account of the items collected and approved for delayed collection in the chou and hsien under his jurisdiction, and investigate them personally. When cases of "collecting more and reporting less" (cheng-to pao-shao ) were discovered, it was the duty of the prefect to report this immediately to his superior.^[39] Responsibility for investigating other revenue stores was also codified. Taxes kept in the treasuries and granaries of the grain intendant and post intendant were to be inspected by the financial commissioner. Those stored in the various prefectural treasuries and granaries were the responsibility of their respective circuit intendants (tao ). Where no deficit or arrears were found, the prefects were to be recommended to the governor, who would issue rewards to express encouragement.^[40]

This allocation of supervisory responsibility was to take on particular importance when large-scale inquiries of deficits and arrears (ch'ing-ch'a ) were undertaken later in the dynasty. Responsibility for making good shortages was placed on the official himself if the shortage had been reported by his superior at the time it occurred. However, in cases where shortages had not been reported, whether due to negligence or cover-ups, the supervisory official was also required to share in repayment (fen-p'ei ) and was occasionally charged an additional fine at the rate of as much as ten times the original amount due.^[41]

The primary focus of the annual accounting system was the ti-

ting ch'ien-liang , those taxes derived directly from levies on the population by head and from private land under cultivation. Not only was this the largest single source of central-government revenue, but its volume was fixed by quota and was therefore more easily controlled.^[42] In addition, it was the one source of government revenue that was shared statutorily with the provinces in the form of "retained taxes." Inasmuch as the purpose of the tsou-hsiao system was to control both income and expenditure, the ti-ting ch'ien-liang was obviously given special attention. In response to a memorial in KH 7 (1668), the special nature of the land and head taxes in relation to the tsou-hsiao reports was clearly outlined.^[43] All taxes collected by the chou and hsien but not emanating from the ti-ting were to be reported in the same manner used prior to the reform. These included the salt gabelle, duties on tea, customs duties, the reed-land tax, tribute grain, tribute presented by local aboriginal chiefs, miscellaneous duties, the pawnshop tax, contract tax, brokerage tax, taxes per head of livestock and mules, fines and payments to atone for crime, and so on.^[44]

At the end of the year, the magistrate compiled detailed accounts of the taxes collected and those still outstanding in his chou or hsien. These accounts were then turned over to the financial commissioner, sometimes through an intermediary such as the prefect or circuit intendant. The financial commissioner himself drew up the four-column accounts for the whole province, listing the balance to be carried forward, new receipts, expenditures, and the present balance. In turn, he submitted a draft to the governor or governor-general, whose staff transcribed it in the form of a "yellow account" for the emperor and a "clear account" for the Board. These reports included a summary of how much ti-ting ch'ien-liang was collected, how much was paid and in arrears, how much was remitted to the capital and retained in the province, how much was transferred to the military, how much was used to buy local products for the capital, and how much was left over.^[45]

This new system of auditing not only meant that provincial officials had to justify their accounts at the end of the year, but also gave the Board of Revenue authority over the way in which local revenues were used on a day-to-day basis. Although enforcement must have been difficult, the law now required that no local retained funds could be allocated without prior Board authorization. Even funds that had been so budgeted could not be disbursed until another report had been made indicating the exact amount

that was required at the moment of allocation.^[46] At the end of the year, the Board compared the local official's annual accounts against these "receipts" in the capital to make certain that all items of income and expenditure balanced, and that all expenditures had been approved in advance.^[47] After an item-by-item audit, the Board would either accept the accounts (chun ) or reject them (po ) and return them to the governor for revision or clarification. Although the work of verification and compilation was carried out at lower levels of the provincial administration, culminating in the financial commissioner's report, the ultimate responsibility for promptness and accuracy again lay with the governor and governor-general. If a magistrate was found guilty of delaying his tsou-hsiao report, or of compiling confusing accounts designed to conceal fraud, he was to be impeached immediately by the governor or governor-general. If the latter failed to report irregularities perpetrated by their subordinates, or were lax in their supervision and failed to discover malfeasance that was later revealed, they were held jointly responsible.

In view of the large number of deficit cases uncovered in the first years of the Yung-cheng reign, the efficacy of this system of centralized control may be doubted. As we shall see, it was possible to evade careful scrutiny of accounts by bribing the Board officials and clerks. Moreover, the penalties for delays were minor and were prorated according to the length of the delay.^[48] Sanctions for discrepancies were also light, ranging from demotion of one grade and transfer for a magistrate or prefect to a fine of one year's salary (feng-yin ) for a commissioner or circuit intendant and six months' salary for a governor or governor-general.^[49] This was not considered a great hardship, inasmuch as feng-yin was itself a token sum.^[50] Moreover, during part of the K'ang-hsi period, official salaries were suspended to help pay the costs of pacification campaigns.^[51] Even when they were granted, the salaries of officials and yamen staff were often arrogated in full or in part by the governor to pay for expenditures not covered by ts'un-liu funds, a practice euphemistically called chüan-feng or "contributions of salary."^[52]

However ineffective this system of scrutiny may have been in the early years of the dynasty, it did establish a standard for local and provincial accountability in the area of public finance that was strengthened under later emperors. In particular, it marked a further refinement of the Ming system by placing supervision of both collection and expenditure under one agency. Not only did this

contribute greatly to the centralization of fiscal authority, but it also enhanced the central government's interest in improving tax-collection and local-accounting procedures. The form of the tsou-hsiao ts'e itself can be seen as contributing to the development of local budgets. Moreover, without such a system, the investigation of deficits undertaken in the Yung-cheng period would have been impossible. Finally, by placing all legitimate income clearly under the control of the central government, the tsou-hsiao system of annual accounting also contributed to the crisis in local funding that led to the reorganization of provincial finances in the early eighteenth century.

Centralization of Tax Collection

The aim of the tsou-hsiao system of annual accounting was to prevent revenue loss resulting from official waste and corruption. However, the early Ch'ing government also devised methods to rectify the defects in the Ming fiscal apparatus that had prevented taxes from ever reaching the local and provincial coffers. Whereas the Ming had attempted to utilize local elites as instruments of government policy and had failed, the Ch'ing remained committed, until the nineteenth century, to the principle of a direct relationship between the government and the people. In the realm of tax administration this meant waging an unending campaign against both official and nonofficial intermediaries who took on the role of tax farmer, and creating an informed public that could both do without such middlemen and resist their efforts at control of tax collection.

The first step in this direction was taken as soon as the Manchus entered China. In part because the dynasty had not yet been able to compile its own editions of the Fu-i ch'üan-shu , the new rulers turned to the use of "easy-to-read lists" (i-chih yu-tan ) to inform taxpayers of their annual rates. Yu-tan were not unknown in the Ming dynasty. However, the memorialists requesting the use of "easy-to-read lists" in 1644 placed a new emphasis on their effectiveness in cutting through the confusion of regulations, quotas, and tax rates as they appeared in the Fu-i ch'üan-shu . This was not intended as a mere stopgap, but as a way to improve tax collection in the long term by providing both officials and taxpayers with an easily understood enumeration of the actual amount of taxes each household owed.^[53]

Whatever the hopes of its original formulators, the system of ichih yu-tan failed as a panacea for China's fiscal ills. By KH 24 (1685) they were discontinued because officials and yamen runners were found to be using the costs of their production as a pretext to levy surcharges on the people.^[54] In place of the defunct "easy-to-read lists," the emperor ordered all county-level administrative units to set up stone tablets in front of the yamen informing the people of the tax rates in the vicinity according to the Fu-i ch'üan-shu .^[55] Unfortunately, this system once more left the peasantry prey to the machinations of middlemen, because few villagers made regular trips to the county seat where they could see these tablets. Even those who did undertake such a journey would still have to estimate their own personal liability and manage to insist on that amount in the face of the conflicting and often much larger claims of yamen runners and tax-collection clerks. Thus, the task facing government was twofold—to find a new way to inform the peasants of their individual tax liabilities and to free the taxpayer from those institutions that perpetrated the hold of middlemen over them.

The most culpable of such institutions was the old Ming li-chia . By the early Ch'ing this system of rotating responsibility for tax remittance had become little more than a vehicle through which local strongmen and bullies (hao-lieh chien-kun ) took control of an area's taxes. Domination of the collection process by third parties fostered extortion. In some places the practice of levying surcharges through the li-chia was so well established that the methods used had even been given names. If all the li in a hsien shared in the annual payment of these fees it was called juan-t'ai . If, on the other hand, each li and chia took turns paying, it was called ying-t'ai .^[56] Whatever the name, the result was the same: added burdens for the people and diminished revenues for the government.

The means devised to overcome these malpractices was the system of "rolling lists" (kun-tan ). Unlike the "easy-to-read lists," "rolling lists" were not simply grafted on to the li-chia , but were meant to replace them. Under this new system, households were still grouped together in units called li . However, the Ming arrangement of households into artificial decimal units, regardless of physical proximity or social relationship, was abandoned. "In a single hsien, a li might have five households or ten households."^[57] The basis for establishing these new units was that they be natural

groupings of households whose relationships with each other went beyond tax collection. The kind of information provided on the "rolling lists" was also more extensive than the system they replaced. Under the name of each household in the li was clearly delineated the amount of land its members owned, how much tax was due on that land in silver and in kind, how much should be paid in the spring collection period, and how much should be paid in the fall. These payments were then further divided into ten installments and the amount due in each installment was also set out in detail.

The most important feature of the "rolling lists" was not what it was, but what it was not. It was not a system of collecting taxes. It was only a means of notifying the taxpayers of the taxes due. In the edict promulgating the use of "rolling lists," special emphasis was placed on informing the populace that each taxpayer was to wrap his own taxes and deposit them personally in the county tax chest. No one was permitted to act for the people in this regard, whether it be the old li headman, silversmiths, yamen personnel, or any other third party. The "rolling lists" would guarantee tax payments not by making one man responsible for delivery of everyone's taxes, but by making rotation of the list itself contingent upon each household's fulfilling its obligations to the government. The list containing all of the above-mentioned information would be issued to the first name and passed on in turn to the others. As the list came to each household it was to pay its individual tax. When one installment period was completed, the list was passed around again for the second period. If a taxpayer destroyed the list, did not pay in full, or failed to pass the list on to his neighbor, he as an individual was to be investigated, arrested, and severely punished.^[58]

If taxpayers were to be responsible for wrapping and depositing their own taxes, in turn they had to be protected against false accusations of arrears. Under the system inherited from the Ming, chou and hsien magistrates entered the amount of tax received from each taxpayer on special receipts bearing the magistrate's official seal (yin-p'iao ). Two copies were made for each payment. One was kept by the magistrate and one was given to the taxpayer as proof of payment. Abuses were common. Most often, on the pretext of needing the second copy to provide yamen runners with evidence when investigating and prompting payment of arrears, taxpayers were deprived of their own copies. Unscrupulous offi-

cials and clerks would then record the paid taxes as unpaid or would record payment at less than the amount actually collected.^[59] The helpless taxpayer was left with no recourse but to pay his taxes a second time or produce a bribe in order to avoid being harassed by tax-prompting runners.

To solve this problem, "three-stub receipts" (san-lien yin-p'iao ) were introduced. Each receipt had three sections. One was retained by the magistrate, one was turned over to the taxpayer as proof of payment, and one was kept by the yamen runners to compare with the other two in prompting payment of arrears (ying-pi ).^[60] It was hoped that the addition of one piece of paper would not only deprive the magistrate and his underlings of any pretext for withholding taxpayers' receipts, but would also ensure the magistrate's superiors of a record against which to compare the former's remittances to the capital.

Imperial Benevolence and Fiscal Administration

"Rolling lists" and "three-stub receipts" represented an important advance over the Ming system of tax assessment and collection. By placing the emphasis in tax collection on the individual household in a direct relationship with the government, these early Ch'ing reforms served as a local counterpart to the centralizing efforts within the bureaucracy manifested in the tsou-hsiao system of annual accounting. Yet, just as the latter was plagued by evasion and manipulation, so did "rolling lists" and "three-stub receipts" fail to wipe out tax farming and the oppression of the masses by rapacious yamen functionaries.

Seeing the problems facing them strictly in terms of tax evasion and bureaucratic corruption, late Ming and early Ch'ing reformers naturally sought to root out the sources of government shortages through refinements in the existing system of imperial finances. As important as these refinements were, they were all bureaucratic solutions to what were perceived as fundamentally bureaucratic problems. The weakness of such measures is epitomized in the case of the three-stub receipt. When, in the Yung-cheng period, this system was found inadequate in preventing official and clerical corruption, it was replaced by the four-stub receipt. But even a ten-stub receipt would not have solved the problems that gave rise to this corruption in the first place.

One source of difficulty was the laxity with which the govern-

ment handled offenders. We have already seen that the sanctions accompanying the implementation of annual accounting were insufficient to dissuade the most determined officials from exceeding deadlines and tampering with their fiscal records. As will be seen, the incomplete state of most tax registers after the long period of dynastic transition also impeded the government's efforts at fiscal control.^[61] The existence of such obstacles could only have contributed to the K'ang-hsi emperor's personal leniency as a ruler, a trait for which he was famous. For K'ang-hsi, gaining the confidence and support of the local elite and the overwhelmingly native Chinese bureaucracy was as pressing a task as increasing government revenues. Leniency, as a personal quality, may have suited the emperor's temperament, but in the early years of the dynasty its importance was clearly political.

Leniency was common in the handling of both of the traditionally recognized causes of government shortages. When the level of unpaid taxes rose too high, the K'ang-hsi emperor would simply declare a tax amnesty. For example, in KH 43 (1704) the emperor cancelled all the tax debts of the entire populations of Shansi and Shensi. In KH 50 (1711), he excused Chihli, Fengtien, Chekiang, Fukien, Kwangtung, Kwangsi, Szechuan, Yunnan, and Kweichow from 8,377,100 taels in land and head taxes. At the same time, in order to lessen the burdens on the people, the emperor cancelled accumulated arrears of more than 1,185,400 taels.^[62]

Even in cases of obvious official corruption, K'ang-hsi seems to have been exceedingly hesitant to apply strict disciplinary sanctions. In 1685, the Director of Grain Transport memorialized on corrupt practices in the shipment of tribute grain to the capital. The Council of Ministers deliberated the case and recommended the impeachment and removal of the officials involved, but the emperor's edict on the affair advised instead that the guilty parties simply be given a stern warning against any future repetition of such activities. As malfeasance of the kind they were accused of had long since become common practice, it was, K'ang-hsi felt, unfair to single these men out for punishment. The emperor was equally lenient in the face of mounting official deficits. Despite regulations requiring officials to repay all deficits incurred during their terms in office, he ruled in a case in 1702 that such repayment was too difficult and asked that the Grand Secretariat find another solution to the problem.^[63]

An excellent example of the more "benevolent" approach taken toward tax offenders during the K'ang-hsi period can be seen in the handling of so-called "hidden lands" (yin-ti ). Hidden lands were the product of the widespread concealment of newly reclaimed land from the tax registers. Some of this concealment was perpetrated by individual peasants who reclaimed small plots and either deliberately or through ignorance of the law never reported them to the authorities. However, according to later investigators, much of the land was reported to the officials, who themselves collected the taxes and did not report these additions to the tax rolls to the central government.^[64] Thus, the problem of hidden land can be seen as encompassing all of the defects in early Ch'ing fiscal administration: incomplete records, tax evasion, and official peculation.

In KH 51 (1712), an exchange of views took place that highlighted the contrasting approaches advocated on the issue of hidden land.^[65] The debate developed over the poor progress being made by Szechuan province in restoring its tax income to that collected during the late Ming.^[66] The original quota was 1,616,600 taels. However, by 1710 the most that could be wrung out of the province was 202,300 taels. Szechuan Governor Nien Keng-yao was appalled that after nearly seventy years of Ch'ing rule, revenues in the province had been raised to only a little over 10 percent of what they had once been. In part, he blamed officials who used the occasion of reporting new land for taxation (shou-liang ) as a pretext to extort fees from the people. Consequently, the people often did not report reclaimed land.

Nien took a hard line on the problem. The only way to ensure increasing quotas was a strict policy of rewards and punishments. He suggested, therefore, that promotions be granted to all chou and hsien magistrates who could increase tax collection to 40 or 50 percent of the Ming quota. On the other hand, promotion would be blocked for those who could not reach at least 20 percent of the quota. Any official failing to collect even 10 percent of the taxes collected in the Ming would be demoted and transferred, and those showing no increase over present levels would be deprived of their rank and office.

A more moderate approach was advanced by the censor Tuan Hsi. Tuan pointed out that the main cause of Szechuan's low level of tax revenues was the depopulation that had occurred in the province during the uprisings of the late Ming. Even a diligent gov-

ernor like Nien himself had been able to add only 26,000 taels to the tax rolls. Whereas Nien would claim that the wealth, in population and cultivated land, was there to be tapped if only the law gave him the weapons to do so, Tuan felt that such a policy would benefit only the corrupt. Virtuous officials would find these figures impossible to achieve and would be impeached. Unscrupulous officials seeking promotion would force the innocent to confess to hiding land they did not have in order to reach the magic numbers they needed. Even though Tuan's plan supported a thorough investigation of hidden land and omitted taxes (yin-lou )^[67] and punishment of any official who used the process of reporting new land to engage in extortion, its main emphasis was moral exhortation. Raising Szechuan's quotas could be accomplished only by ordering officials to encourage the people wholeheartedly to report their land and by punishing those who evaded taxation. Nothing was said of those officials who succeeded in getting the people to declare the fruits of reclamation but who never passed this added income on to the state.

Needless to say, it was Tuan's method of gentle persuasion that was endorsed by K'ang-hsi. Benevolent rule was upheld, but hidden land, along with other forms of embezzlement and tax evasion, continued to be a problem long afterwards. Nevertheless, it was not simply laxity that led to the deficits discovered by Yungcheng when he inherited the empire from his father. Had that been the case, the strengthening of sanctions and the tightening of official discipline imposed by Yung-cheng from his first day as emperor would have been enough to solve China's fiscal problems and the reforms of the 1720s would never have taken place. It was certainly with no more ambitious prescription in mind that the new emperor initiated his attack on the fiscal chaos that was his birthright. Once the campaign was launched, it disclosed weaknesses in the Ch'ing fiscal system that went far deeper than anything admitted by the formulae of the traditional political economy and would require far more fundamental changes than the patchwork refinements and exhortations to virtue so popular in the past.

2—
Informal Networks of Funding

Since the days when free traders and missionaries first insinuated themselves into Chinese society, Westerners have clung to an image of a China ruled by a corrupt bureaucracy in which the "squeeze" or bribe was the usual means of achieving objectives. When the Yung-cheng emperor first took up the reins of power, he also saw corruption as the root of his government's fiscal difficulties. Such an analysis made possible the Westerners' sweeping denigration of the Chinese system of rule. Ironically, for the Chinese the all too visible evidence of corruption in their ranks facilitated uncritical acceptance of fiscal arrangements that had long since been removed from the administrative structure for which they were designed. If government deficits and arrears were the result of evil deeds by evil men, all that was needed was the intensification of moral exhortations and the punishment of the worst offenders. It was by these means that generations of Chinese reformers sought to restore the health and purity of the system as a whole.

No one would deny that fiscal malpractices abounded in late-imperial China. Nor would anyone deny that the chief victims of these evil deeds were the common people. However, if we view corruption strictly in terms of the venality of the ruling strata of Chinese society, we miss the opportunity to understand why it

played such an important part in Chinese public life. Men who steal for personal gain exist in any social system. Far more pernicious were the conditions that made manipulation of the legal tax administration necessary for the very survival of the Chinese bureaucracy. Once such practices became established, it is not surprising that their more self-aggrandizing manifestations also multiplied.

Revenue Sharing in the Early Ch'ing

The early Ch'ing rulers were centralizers. Having entered China from outside its borders, they had little stake in local customs and little knowledge of local needs. When implementing policy they generally aimed at uniform application of the law (hua-i ). When they adapted Ming institutions, they did so in such a way as to enhance the authority of the state. This is most striking in the area of fiscal administration. We have already seen the extent to which the Single Whip reforms of the late Ming contributed to the weakening of the fiscal bases of local government. The Manchus took over the framework of postreform finance and added to it refinements designed to strengthen further the control of the central government over all the resources of the empire. As a result, provincial and local administrations were left without adequate sources of income of their own. We can begin to understand the dilemma facing the official in the field by examining how tax revenues were distributed during the early Ch'ing.

Retained and Remitted Taxes

China's early-eighteenth-century fiscal crisis was not primarily a result of an inadequate appreciation of the taxable resources of the late-imperial economy. Taxation was moderate, but included a wide variety of direct and indirect levies. Besides the predominant land and head tax (ti-ting ch'ien-liang ), paid largely in cash, several minor taxes in kind also persisted.^[1] In addition, several provinces owed a quota of grain tribute (ts'ao-liang ) in kind, shipped to the capital along the Grand Canal.^[2] Miscellaneous indirect taxes (tsa-shui ) without fixed quotas brought the government a share of local commercial profits. Taxes were levied on the sale of specific commodities such as tobacco, cotton, wine, pigs and other live-

stock, and goods being unloaded at transit points or for sale at local and regional markets. Real-estate transactions were recorded and taxed through the levy on title deeds, and retail establishments, pawnshops, and brokers were also subject to imposts and licensing.^[3]

The most important sources of commercial revenues were the salt gabelle, the tea tax, and customs duties. Next to the land and head taxes, these constituted the largest portion of the central government's annual income.^[4] Moreover, unlike the land and head tax, miscellaneous duties, and so on, which were collected by the regular county-level bureaucracy, each of these duties was generally managed by its own special administrative apparatus. The existence of special agencies to collect the salt, tea, and customs revenues strengthened the central government's control over these duties, though according to dynastic law all the above-mentioned taxes were the property of the central government. In documents pertaining to fiscal matters these taxes were often referred to collectively as the "primary revenues" or "regular taxes" (chenghsiang ) to distinguish them from taxes or tribute of a more regional or sporadic nature and from surcharges levied on the regular tax quota itself.

Of all these taxes, only one was statutorily shared with the provincial and local governments. This was the land and head tax. In the Comprehensive Books of Taxation and Services for each province, the land-and-head-tax quota was divided into the amount to be retained for provincial use (ts'un-liu ) and the amount to be remitted to the imperial capital (ch'i-yün ). It is therefore on the basis of the retained land- and head-tax quotas that we must judge the adequacy of the Ch'ing system of revenue sharing. Table 2.1 presents the distribution of retained and remitted taxes in 1685. The retained tax quotas for Honan, Shansi, Shantung, Chihli, Kiangsi, Hupei, Hunan, and Chekiang are high because they include revenues sent directly to assist nearby provinces with small gross-tax receipts. Likewise, the high percentage of local revenues retained by provinces such as Szechuan and Yunnan was a reflection of their unusually low level of land-tax income and not any greater munificence in providing for their local administrative needs. Because these retained taxes underwent extensive redistribution, the most we can conclude from these figures is that an average of 21 percent of all land and head taxes was left in the provinces for local use.

This does not mean that ts'un-liu quotas were unalterable. As the total income from direct taxes increased, so did the amount left in the provinces.^[5] In fact, the quotas of retained taxes were set annually, on the basis of estimates of salary and military-supply requirements submitted by the governor and governor-general of each province.^[6] Thus, adjustments could be made to meet specific increases in anticipated expenses, such as wages for additional troops stationed in a province, or salary raises resulting from the upgrading of subordinate administrative units. There was also provision for emergencies. Additional assistance funds could be requested from neighboring provinces, and if these proved insufficient, salt-gabelle revenues or direct aid from the central government might also be obtained.^[7]

Nevertheless, the flexibility of ts'un-liu funding was extremely limited. The definition of an emergency allowed only for such events as major natural disasters and military campaigns. Nothing was provided for the local authority confronted with pressing shortages in the course of day-to-day administration. Such con-

frontations were frequent, in part because the share of revenues left for local government was small. Equally important was the fact that these funds were available for specified uses only, and even then were strictly monitored by the newly created tsou-hsiao system of accounting.

The Poverty of Local Administration

The calculation that 21 percent of all land and head taxes were allocated at the local level masks a startling fact about ts'un-liu revenues. In reality, an overwhelming portion of these funds was not available for local expenses at all.

The uses to which retained taxes in the Ch'ing could be applied fell into three broad categories: military supply, the imperial post, and local expenses. Until recently it was possible only to attempt rough estimates of the relative weight given to each of these types of expenditure. However, with the opening of the Ming-Ch'ing Archives in Peking to foreign scholars we can now examine existing tsou-hsiao accounts to determine the precise manner in which local revenues were distributed during the late K'ang-hsi period.

The only complete tsou-hsiao ts'e containing the information we seek is a report on Shansi land and head taxes for the last year of the K'ang-hsi reign. This report is particularly valuable because it delineates, hsien by hsien, the amount of retained ti-ting taxes that was allocated to each of the categories for which local revenues were allocated. Most revealing in the Shansi statistics is the unexpectedly high proportion of retained revenues destined for military expenses. If we add to this the portion reserved for the upkeep of the imperial post stations, a responsibility exercised at both the county and provincial levels,^[8] we find that over 84 percent of all land and head taxes left for local use were earmarked for expenses primarily of concern to the central government.^[9]

The constraints on local finances become more apparent when we examine the kinds of expenditures covered by "local expenses." These were quite specific. Besides the government stipend for all ranking officials (kuan-feng ), retained funds included wages for yamen runners (i-shih ), ceremonial expenses (ch'i-ssu ), student stipends (ling-shan ), and charity for the orphaned, widowed, and poor (ku-p'in ).^[10] That these were fixed allocations based on the type of administrative unit is obvious from the narrow range of variation in the figures in the sixth column of table 2.3. If the size

and complexity of an administrative unit was reflected in its total land- and head-tax revenues, the percentages of retained taxes allowed, as listed in the third and seventh columns, show that there was little relationship between funds allocated and the administrative needs of a chou or hsien. Larger county-level units, with greater population and productive capacity, kept a much smaller proportion of their gross output in taxes than did their more remote and backward neighbors.

These figures appear small to have served as the total administrative budget legally available to local governmental divisions. However, not even these sums were destined entirely for local use. For accounting purposes most of the retained taxes of the whole province were deducted at the county level.^[11] Thus, the average of about 1,500 taels listed under local expenses in each chou and hsien also included its share of the expenses of all higher administrative units. After deducting their own allotment, most counties seem to have sent their portion of the prefecture's expenses directly to the prefectural seat.^[12] The remaining taxes were then transmitted to the provincial financial commissioner, who completed the distribution of ts'un-liu revenues to higher levels of government.

Included in the third category of retained funds was also a vague item usually called miscellaneous expenses (ching-fei tsa-chih ). Most of these funds were stored in the provincial treasury and were utilized for emergencies and minor outlays. In the event of a serious natural disaster, retained funds were used to provide famine relief. At times they were even used by one province to buy grain for relief of a neighboring province.^[13] Early in the dynasty a portion of the regular taxes was allowed for the transportation of taxes to the capital (chiao-chia ). In KH 2 (1663), transportation expenses were also extended to shipments of military-supply funds outside the province of origin.^[14] On occasion, regular tax revenues were also made available for repairing granaries, opening mines, and other extraordinary expenses.^[15]

The allocation of regular tax revenues for the above expenditures was clearly of benefit to the province as a whole. Except in the case of famine relief, local governments were affected less directly. The costs of provincial projects not paid for by central-government grants of regular taxes were often met by assessing a contribution from each chou and hsien magistrate. They in turn

usually raised the necessary amount by levying surcharges on the local population. Thus, at the very least, the existence of ts'un-liu relieved the local administration of some of the burdens of added taxation. However, it still left provincial officials at all levels of the bureaucracy with considerable expenses for the everyday tasks of government.

Of course, Shansi cannot be taken as a microcosm of the whole empire. However, besides being the only province for which complete ts'un-liu data of this kind have been preserved, it does have the advantage of having been in the middle range of land- and head-tax quotas. Regional variations in retained revenue allocation did exist. Kiangsu province stands out in table 2.1 as retaining an unusually large portion of its regular taxes, and an examination of a tsou-hsiao ts'e for Soochow prefecture in KH 32 (1693) confirms this impression. (See table 2.4.) Not only did chou and hsien in Kiangsu receive special ts'un-liu funds for water conservancy, but the sums allocated for local expenses were also the highest in the empire. There are a number of possible explanations for this. Miyazaki Ichisada has shown that the number of statutory yamen runners in Soochow was extremely large.^[16] This is not surprising, considering Kiangsu's role as the economic and tax-producing center of early Ch'ing China. Moreover, it is likely that the figures given for official stipends, runners' wages, ceremonial costs, charity, and miscellaneous expenses in Soochow included the contribution its subordinate units made to the various specialized agencies in the region, notably the imperial salt and silk administrations. Nevertheless, it does seem that officials in Soochow, and Kiangsu as a whole, may have been less pressed financially than were officials in other parts of China.^[17]

If we were to seek a southern example to compare with that of Shansi in the north, Kiangsi is more representative. Here the problem lies in the failure of the annual accounts to separate out military supply in the enumeration of retained taxes. However, if we view "local expenses" as a percentage of total land- and head-tax revenues, they come to 6.8 percent, only slightly higher than Shansi's 4.8 percent. The similarity is even greater when we look at the breakdown of ts'un-liu by hsien. Taking as an example Chiuchiang fu, a prosperous prefecture where we might expect retained revenues to be higher than average, we can see that the funds available for local expenses were only about 100 taels greater than in

Shansi. Moreover, here, too, the amount of a unit's retained revenues does not seem to rise as a function of the size or complexity of the hsien itself.

The Fiscal Demands On Local Government

Local-Government Expenses

Even the most virtuous and frugal of local officials would have found it difficult to operate within the limits set by this system of revenue sharing. Although all ts'un-liu allocations included a category of official salaries, the amounts granted for this purpose were minuscule, comprising neither a living wage nor an administrative budget. Yet, as we have seen, along with runners' wages, these were almost the only funds legally available to civil officials for the purposes of local government.

One's salary as an official depended solely on the rank of one's post. Posts were divided into nine main grades, each containing two subgrades. (See table 2.7.) Runners' wages were considerably lower, averaging around six taels annually, although they could be as much as twelve taels for runners with highly specialized skills.^[18] Inasmuch as the number of officials and statutory runners at any

level of government was fairly constant, it is no wonder that throughout the empire, figures for local expenses in the ts'un-liu records were almost the same.

Because of the low level of local funding, all officials were urged to practice economy. However, the need to manifest one's rank in sumptuary display, as well as the need to support large entourages of relatives, servants, and disciples who accompanied officials to their posts, made real austerity impossible. Oertai, a prominent governor-general noted for his frugality, estimated that he required 6,000 taels annually for his living expenses alone.^[19] Kansu Governor Hsü Jung calculated his living expenses at 5,000 taels of silver plus 405 piculs of grain, the latter being sent to him from his own family property.^[20] The costs incurred by lower-ranking officials were undoubtedly less, but they could not have been met by the meager salaries provided by their feng-yin .

In addition to his own living expenses, every provincial official had a considerable staff to maintain. Because the quotas for yamen runners were set at a time when many administrative tasks were still carried out by the local population as corvée labor, the number of yamen runners for whom the government provided wages was never sufficient to meet the requirements of local administration. As the population grew and administrative responsibilities multiplied, most chou and hsien found that they needed extra staff. By the eighteenth century, An-yang hsien in Honan was spending over 1,700 taels per year on runners' wages alone, only a portion of which was allocated out of ts'un-liu revenues.^[21]

Although at least some yamen runners received wages from the province's retained land and head taxes, no funds were provided to pay the large number of clerks employed by yamen at every level of government. A hsien in the early Ch'ing could have anywhere from several hundred to several thousand clerks.^[22] Clerks were generally grouped into two alternating shifts, each confined to the yamen for ten days at a time, in order to insulate them from the public and lessen the chances for corruption while they conducted official business. During their confinement they had to be provided not only with the stationery and ink that were the tools of their trade, but also with food, water, firewood, lamp oil, and other necessities of life.^[23] Most officials also retained a number of private secretaries (mu-yu ) who were specialists in law, finance, and administration. Their high status and classical education merited salaries that were often greater than that which the government paid the official him-

self. A private secretary employed by a chou or hsien magistrate probably received around 100 taels a year. Salaries for private secretaries serving higher officials were considerably larger.^[24] Despite the strain this clearly put on an official's finances, private secretaries were considered an indispensable part of one's administrative staff. Every magistrate could be expected to have at least two, and it was not unusual for a high official to have ten such men in his employ.

Wages, stationery, ink, fuel, food, and so on can be seen as constituting the "inner" expenses of the yamen, those raw materials that kept the administration functioning. However, the responsibilities and expenses of a provincial official did not stop at the yamen gate. Even before his arrival at a new post, an official could be expected to go into debt to pay the enormous costs of moving his family and staff to a distant assignment.^[25] To this were added the costs of transportation, food, and lodging whenever he or his deputies left the yamen to tour the area under his jurisdiction.^[26] The burden of transporting taxes within the province was shouldered by the unit collecting them, that is, the chou and hsien and the various agencies specializing in commercial taxes. Likewise, the costs of most official communications within the province were borne by the officials who sent them.^[27] Government post stations maintained and manned with provincial ts'un-liu revenues did not exist to facilitate contact among local officials.

All of these expenses pale before what may be called the costs of physical maintenance of a province or county. Imperial gifts flowed into the provinces for the erection of Confucian temples, memorial arches, and plaques bearing facsimiles of the imperial calligraphy, but the emperor left largely to fate those structures upon which the practical welfare depended. The central government funded major flood-control projects, but the smaller dams, dikes, and irrigation projects of the seventeenth and early eighteenth centuries were buih through local initiative, with local funds and local labor.^[28] Similar means had to be found to repair yamen buildings, city walls, granaries, and bridges and roads that did not lie along major post routes. Even famine relief was frequently left to the devices of local officials whose legally available revenues provided them with scant means to aid the victims of disaster or to prevent such disasters in the future.^[29]

The records of the Board of Revenue abound with accounts of officials who could not make ends meet and who were unlucky

enough to get caught trying. Sometimes it was natural disaster that upset the delicate arrangements worked out at the local level to cover the costs of being a "parent official" to the people. The case of Ho Shih, a forty-one-year-old chü-jen from Chungking, was not unusual.^[30] Ho was appointed magistrate of Yin hsien in Ningpo prefecture, Chekiang, in 1715, and for almost nine years had served in that post, seemingly without blemish. Then, in 1723, he was impeached for embezzling 7,882.6 taels in regular land and head taxes. According to Ho's own testimony, his troubles were the result of the poverty of the area he governed. Because this was a coastal hsien, agriculture was poor and many of the inhabitants made their living from fishing. To require them to pay their taxes in silver would have been a hardship, and so he allowed them to pay in the common local medium of exchange, copper cash. Yin hsien was so economically backward that in order to remit these taxes to the provincial capital, Ho had to send the copper cash to a nearby marketing town in another county to be converted to silver. His first impeachment occurred when, unexpectedly, the Ningpo prefect arrived to inspect his treasury and found that the 3,224.7 taels being converted were missing from the hsien.

Ho's inquisitors accepted his explanation regarding these funds, because they were restored as soon as his messenger returned. The magistrate's real problems began when he tried to deal with natural disasters affecting the area at the same time. In 1721, Yin hsien suffered bad harvests due to severe drought, and the next year was also hot and dry. The poor soil of this coastal region had never supported an early rice crop, and in 1722 the late rice crop was lost. Although he had requested a tax remission for the hsien, the shortage of food was so urgent that Ho took 2,217.9 taels of taxes due to be sent to the central government and bought rice from wealthy households in the hsien to try to stabilize rice prices. Unfortunately, many of the families under his jurisdiction were too impoverished to buy rice even at reduced prices. Therefore, he had each affected locality draw up a register of the very needy and offered direct loans of this rice at a rate of one peck (tou ) per adult and five pints (sheng ) per child. The cost of the rice was to be repaid by the recipients after the harvest the following year.

Later in the year, Ho became concerned about further rice shortages because the drying up of waterways had discouraged outside merchants from transporting rice into the hsien. He asked his superiors to urge merchants from Soochow to bring in cheaper

rice and himself allocated 2,380 taels in land and head taxes to send a messenger to Soochow to make a direct purchase of 1,700 piculs of rice for sale in the hsien. Once again an untimely visit by the Ningpo prefect's representative led to impeachment for embezzlement, but because this rice was eventually sold and the money returned to the treasury, Ho was also excused for this indiscretion. In fact, Ho might have escaped prosecution altogether had these two years of drought not been followed by tidal waves in 1723 that made it impossible for the people to pay back the 2,217.9 taels worth of relief grain borrowed the previous year. As a result, the magistrate was forced to sell his personal property to reimburse the central government for relief services provided out of taxes that should have been remitted to Peking.^[31]

The story of Magistrate Ho is only one of many examples of central-government failure to provide for even basic relief measures necessitated by natural disaster. Nevertheless, had Ho's manipulation of regular tax funds worked as originally planned, and had the Ningpo prefect not turned up just when Yin hsien funds were temporarily missing, he might have escaped censure altogether. More pressing was the expense for which a magistrate could expect no reimbursement, as was discovered by the Chüning hsien, Szechuan, magistrate Chuang Ch'eng-tsu.

Magistrate Chuang was probably not the most savory of characters to be found in the early Ch'ing bureaucracy. Among the numerous charges of official malfeasance leveled against him was that of collecting illegal surcharges (chia-p'ai ).^[32] Chuang was able to clear himself of part of the charge by showing that the silver in circulation in Chü-ning was only 70 to 80 percent pure, so that an additional weight had to be added to every tael of tax paid. However, he was still held responsible for levying an illegal surcharge of 10 percent on all taxes remitted in the hsien between the last month of KH 59 and the time of his impeachment in mid-YC 1. Chuang's explanation for his actions reads like a litany of expenditures left to badly underfunded local governments:

The total surcharges I collected came to 2,167.1 taels. During these three years I allocated 320.4 taels for yamen runners' wages. To repair the tower over the city gate and the city wall l spent 238.43 taels. To repair the hsien prison l spent 126.035 taels. To build a charity school (i-hsüeh ) I spent 148.78 taels. To repair the Confucian temple I spent 482.49 taels. To repair

the temple to the city god I spent 149.09 taels. To build an alms house I spent 82.47 taels. To build an evernormal granary I spent 52.54 taels.^[33]

In addition, Chuang disbursed 210.7 taels for construction of a new stone bridge at Hsi-mei chen, 63.27 taels for repairs on the large stone bridge outside the south gate of the hsien city, 39.31 taels for repairs on the temple to loyal officials, 24.6 taels for construction of a ferry at the city's lower pier, and 37.8 taels for the wages of ferrymen. The total expenditure came to 2,168.66 taels, a little over one tael having been "contributed" by the magistrate himself. When the investigators pressed Chuang again, accusing him of using these repairs as a pretext to levy surcharges, Chuang pointed out that all of the above undertakings were the responsibility of a local official, even if funds were not provided for them:

If you expected me to contribute my own money to repair them, I didn't have the means. If I had let them topple, this too would have meant shirking my duties. This is why I added a surcharge to the regular tax. This is really a case of using public funds to pay for public expenses (i-kung wan-kung ). All the gentry and commoners in the hsien know I haven't taken a bit for myself.^[34]

The failure of the regular system of revenue sharing to provide for such basic expenses as small-scale relief efforts, construction and repairs of city edifices, and the hiring of sufficient runners to man the hsien yamen left local officials little choice but to devise other means for their provision. Of course, once such methods were in force, it was easy to extend their use to personal gain. However, whatever happened, the local official was unlikely to emerge from his term in office completely unscathed. If he limited his expenditures to those revenues allowed him in the provincial ts'un-liu budget, the hsien would almost surely suffer and the magistrate would be impeached for dereliction of duties. If he did not remain within his budget, he was open to impeachment for corruption and embezzlement, a charge that carried a penalty of banishment to the outer reaches of the empire. Even if the government recognized that he had used unauthorized funds for public purposes, the responsible official would be forced to sell his worldly goods to repay the amount he had improperly used.^[35] The central government made the task of local officials no less difficult by itself

placing irregular demands on the meager retained revenues of the provinces.

Higher-Level Demands

The central government was entitled to almost 80 percent of all land and head taxes collected in the provinces. However, this did not prevent Peking from frequently denying local officials even their full share of ts'un-liu quotas. Sometimes this took the form of outright confiscation. During the war to suppress the Three Feudatories, military expenses were so high that most of the regular taxes designated for the provinces, including for a time the salaries and wages of officials and yamen runners, were appropriated by the court.^[36] More often, deductions from ts'un-liu were the result of a regulation that required that the central government be sent its full ch'i-yün quota before local officials began to put aside their own retained share of tax revenues.^[37] Since all chou and hsien experienced arrears because of poor harvests, ineffective administration, tax farming, or simple tax evasion, few officials escaped shortages in their operating funds. An official struggling to rejuvenate an administration plagued with arrears was doubly handicapped by loss of all or part of the funds allocated to support himself and the very yamen employees he used to collect taxes. One practice even made officials responsible for compensating the central government for portions of the tax quota that could not be collected because the land on which they were levied had become unproductive and the inhabitants had fled.^[38] Thus, just at a time when additional funds might be needed for relief work, reclamation, or to deal with popular unrest, officials were expected to make deductions from their own salaries and that of their staff to fill the central government's coffers.

Direct encroachment on provincial retained revenues was not the central government's only contribution to local fiscal problems. The rigorous accounting procedures developed by the Ch'ing to ensure strict central-government supervision of local expenditures caused as much corruption as it prevented. The tsou-hsiao system of annual accounting required exact allocations for specific items of expenditure. Because many administrative expenses were budgeted years in advance, a certain amount of rigidity was built into the system. When budgeted funds proved insufficient, an official often had to advance funds from other sources.^[39] Inasmuch as such

advances were illegal, a chain reaction was often set in motion whereby an official borrowed from one category of expenditure after another to cover up his mounting shortages. A post could change hands many times before the discrepancy was discovered, by which time it might be impossible to trace the official originally responsible.

Another aspect of the annual accounting system that could cause difficulties for local officials was the Board price (pu-chia ). The Board of Revenue established rates at which local officials could use regular tax revenues to purchase goods for shipment to the capital and pay for materials and labor for the military, water and road works, and post-station maintenance. When the market price was lower than the Board price, officials could net a profit from such purchases, but where the Board price was underrated or prices fluctuated greatly, the burden on officials could be enormous. In the imperial post service, the costs of horse feed and wages for the laborers who tended the horses varied considerably from place to place. It fell to the officials in each area to make up any discrepancies between the real costs and the funds budgeted for these expenses.^[40] The same was true of many construction projects. Once a budget was set, the costs of materials could change, weather conditions could slow up work, and unforeseen engineering problems could emerge, but the original allocation usually remained unaltered. Cost overruns were so common that an official would arguably have been better off raising the necessary funds by contributions and illegal surcharges, rather than expose himself to charges of mismanagement of ts'un-liu appropriations.^[41]

The provincial financial commissioner had access to greater ts'un-liu revenues than did individual local officials, but these were rarely granted to supplement the costs of local endeavors. Chou and hsien magistrates could borrow funds from the provincial treasury in the event that tax collection was postponed because of natural disaster. However, this procedure, known as chieh-kei , applied only to circumstances in which the magistrate was left without enough money to cover official salaries, runners' wages, and other fixed ts'un-liu expenses, especially those necessary to perform the seasonal sacrifices. Only rarely were emergency funds extended to local government for more tangible purposes, such as bridge and road construction. Even then, the financial commissioner was bound by the same regulations that governed all ts'unliu expenditures. Estimates of costs had to be submitted and veri-

fied by the financial commissioner, and all allocations had to be approved by the Board of Revenue prior to disbursement.^[42] The Board could, and often did, reject such proposals or reduce the allowed costs. The financial commissioner had little discretionary control over the retained revenues in his treasury, and along with the governor, was often held responsible for repayment of any funds the Board deemed to have been wrongly spent.

The Board of Revenue placed one further claim on local tax resources. A portion of the retained taxes of every jurisdiction in the empire was illegally shunted back to the Board in the process of reporting the annual accounts. Ostensibly this "Board fee" (pufei ) was a voluntary contribution to cover the Board's own stationery and clerical costs. In reality no official could hope for approval of his accounts without it, and much could be overlooked if it was sent.^[43] Inadequate funding of central-government administration thus combined with the weaknesses of the local fiscal base to undermine official discipline and the elaborate precautions taken by the court to prevent corruption and shortages in the taxes remitted to Peking.

The emergence of the practice of salary and wage contributions by officials (chüan-feng ) is evidence of the dilemma facing provincial officials trying to balance the demands of provincial administration against the requirements of the fiscal system. At least by the late K'ang-hsi period, lower officials and yamen runners were routinely being forced by their governors and financial commissioners to remit to the provincial capital all or part of their salaries and wages granted by the central government.^[44] A portion of these funds was used by the provincial leadership to pay for specific projects whose costs could not be met with regular ts'un-liu revenues. Honan Governor T'ien Wen-ching reported that when dredging of the Chia-lu River began in 1706, 29,000 taels were raised for the enterprise from the salaries and wages of local officials and runners.^[45] A somewhat different case involved the distribution of rice gruel to famine victims in three prefectures of Shantung province during 1719–20. Over 117,000 piculs were drawn from the province's relief granaries at the time, all of which were replenished through contributions from the salaries and wages of the local bureaucracy.^[46]

The most common justification for appropriating salaries and wages was to make up provincial deficits. The prevalence of this practice may have resulted from the apparent sanction it received

from the throne. Because the chou and hsien were required to remit the central government's tax quota before deducting their own ts'un-liu allowances, the contribution of salaries and wages that came from the ts'un-liu was in a sense built into the fiscal system. This was reinforced in 1706 when the K'ang-hsi emperor reprimanded the Honan bureaucracy for reporting tax arrears that year amounting to over 400,000 taels. Clearly, the emperor did not view these shortages as arrears, and accused the Honan officials of falsifying their accounts in the hope of receiving a large tax remission. If the shortages in Honan were, on the contrary, a product of bureaucratic incompetence, K'ang-hsi felt it only right that the officials bear the burden of repayment from their own salaries and their underlings' wages.^[47]

For provincial governors, chüan-feng was an invaluable tool for balancing provincial accounts, and the Board of Revenue was liberal in approving its use. Governor Chang Po-hsing received permission to use salaries and wages for the years 1710–12 to repay 150,000 taels of deficits in Kiangsu.^[48] During his term as Yunnan financial commissioner, Li Wei reported that sizeable military expenditures which had not received Board approval were now being made up through chüan-feng .^[49] Contributions of wages and salaries released high provincial officials from personal responsibility for repayment of certain deficits and as such overcame some of the rigidity built into the tsou-hsiao system of annual accounting. It did so, however, at the expense of local officials, who could ill afford yet another encroachment on their already meager legitimate resources.

The Informal System of Funding in the Early Ch'ing

Shortages were part of the life of every provincial official in late-imperial China. Yet, throughout the seventeenth and early-eighteenth centuries, rivers were dredged, city walls repaired, bridges built, and roads improved. Justice was administered and order maintained, the poor were given aid, and the greatest consumer of public revenues, the bureaucracy itself, continued to function unimpeded. By no means was China under the Manchus a welfare state, nor was it ruled by a particularly activist regime. Nevertheless, the government of early Ch'ing China operated at a level of achievement far beyond the capacity of the regular system of taxation and finance.

With all regular taxes the property of the central government, and revenue-sharing provisions grossly inadequate to meet local administrative needs, additional funds had to be found elsewhere. Donations from influential and wealthy families were of some assistance, as were the independent undertakings of prominent individuals and lineages. The total contribution from these sources, however, was small when compared with the myriad expenses local governments incurred each year. Nor did China have the benefit of great banking houses, which could have themselves developed as the underwriters of public finance. In the end, provincial officials in China had only two options: either siphon off funds allocated by the central government for other purposes or destined for remittance to the central government, or squeeze the necessary funds from the people in the tax-collection process or in the marketplace.

By the eighteenth century, there had evolved in China a complex informal system of funding local government that utilized both options. Operating parallel to, and in coordination with, the regular fiscal administration, it was informal only in that it was not part of the statutory system of revenue sharing. As such, the money and goods that entered this system were not subject to the control of any higher authority. Most of the methods used to acquire them were illegal, but the informal funding network was not simply corruption institutionalized on a national scale. Although many of those who participated in the network undoubtedly took advantage of it for personal gain, its existence was primarily a response to the failure of the late-imperial fiscal administration to provide officials with the means to carry out their duties. For at least a century the informal funding network filled this gap so well that many in government refused to acknowledge the existence of the problems that had brought it about in the first place.

The Magistrate

As the official closest to the people, the magistrate bore the main responsibility for funneling the wealth of the countryside into the coffers of local government. It was he who collected most of the empire's taxes, including the land and head tax. The magistrate was also the issuing agent for local merchant licenses and contract and deed certificates and was the collection agent for a number of commercial taxes. He compiled the tax and census registers, collected contributions from local wealthy households, and in some

cases supervised the sale of salt, tea, and other government monopolies. As such, the official assigned to govern a chou or hsien was in a unique position to manipulate the legal tax structure in order to generate the extralegal revenues that allowed the massive Chinese bureaucracy to function.

Few magistrates engaged in overt extortion of large sums of money from the people. The methods used were more subtle, and the amount taken from any one source was usually small. The two most commonly employed means to acquire extra funds were the levying of surcharges (chia-p'ai, ssu-p'ai, k'o-p'ai ) and "collecting more and reporting less" (cheng-to pao-shao ).

Surcharges were not always levied surreptitiously. In the event of a beneficial government undertaking, especially an irrigation or dredging project, local people were frequently assessed a part of the costs according to the amount of land they owned.^[50] Because the revenue from land and head taxes in Yunnan was low, the governor ordered all chou and hsien to collect a fixed surcharge on the regular tax to pay for government operating expenses. The rates for this surcharge, called kung-chien , were engraved in stone to prevent their collection from giving rise to further exactions.^[51] Most chou and hsien collected a general surcharge in the form of wastage allowances. These allowances, called hao-mi in the case of tribute grain and huo-hao or hao-hsien in the case of taxes paid in silver, were originally instituted to cover any shortages resulting from rotting grain or losses in the value of silver when it was melted down into larger ingots for transport to the capital.^[52] As we shall see, by the end of the seventeenth century, in some provinces, these surcharges had grown to as much as 50 percent or more of the original tax quota.

Several other means were used to levy surcharges without appearing to do so. One of the best-known methods was the manipulation of conversion rates.^[53] Magistrates would allow the people the convenience of paying their taxes in copper coins instead of silver. However, the ratio of copper to silver would be set higher than the prevailing market rate. The official would then make a small profit when he converted the taxes back to silver. The same process was applied to the collection of tribute grain, allowing the people to pay in cash at a rate higher than the official would have to pay to buy the equivalent amount of grain. Similar results were obtained by employing weighted scales or oversized measures at the tax-collection points set up throughout the county. An inge-

nious combination of devices was used in the collection of tribute grain in parts of Kiangnan. During the K'ang-hsi period, magistrates engaged in a practice known as chiu-k'ou . For every picul of rice paid by the taxpayer, the magistrate issued a receipt for only nine pecks. In addition, the use of weighted scales meant that in the end the taxpayer was credited with only 80 percent of his actual remittance. When this practice was discovered and outlawed, a weighing fee (tui-fei ) of 0.06 tael per picul was permitted. The chou and hsien promptly converted this payment to copper coins at a rate of 1,000 per tael. The price of copper coins in Kiangnan was quite high at the time, 1.2 to 1.3 taels per thousand. Moreover, as harvests had been good and the price of rice was low, the people continued, in effect, to pay the chiu-k'ou .^[54]

In the process of collecting the land and head tax, several methods of exacting surcharges were also widespread. At the beginning of the tax-collection period, when the magistrate compiled the tax-collection registers listing the quota due from each taxpayer, he would secretly add a hundredth or a thousandth of a tael to the liabilities of each taxpayer. This practice was known as fei-sa . If the people were ignorant of the tax rates, as was often the case, it could be done with virtually no chance of detection. Although the amount added to the tax on each mou of land was minuscule, for an entire hsien, particularly if it was located in an area of high taxes and productivity, the yield from fei-sa alone could be as much as several thousand taels.^[55]

When submitting his taxes in silver, a practice that was increasing during the late Ming and early Ch'ing, the taxpayer had to rely on the services of a silversmith. The latter weighed the silver and melted it down into ingots, applying his stamp as proof of the metal's purity. Some magistrates took advantage of this by requiring the people to go to "official silversmiths" (kuan-chiang ) who set up shop in the vicinity of the chou or hsien yamen. If the taxpayer's silver did not bear the stamp of the official silversmith, he was not permitted to deposit it in the tax chest. The silversmiths usually took a fee of several hundredths of a tael for every tael of silver weighed and wrapped. In return for the magistrate's patronage, the silversmith would then send a portion of his fee to the yamen.^[56]

Even after the taxpayer had returned home, secure in the knowledge that he had settled his accounts with the government, he was still not free from the magistrate's demands. On the pretext that

there were deficiencies in the weight of the silver (tuan-p'ing ) found when the taxpayer's tax envelope was opened, the magistrate would occasionally send a trusted runner or clerk to the countryside to force the taxpayer to make up the additional amount. The actual silver would be removed and a red tag attached to the empty envelope. On the tag a shortage of 0.03 or 0.04 tael might be recorded for every tael of taxes due. The silver having already been combined with the other taxes received by the yamen, the taxpayer could not demand that his own remittances be reweighed in his presence. He had no choice but to pay the increase. This practice, called "making up the weight of an empty envelope," could also result in hundreds or thousands of taels of additional income for the magistrate's yamen.^[57]

The above techniques for acquiring additional revenues were carried out mainly at the expense of the taxpayer. When the method of "collecting more and reporting less" was utilized, the magistrate siphoned off silver and grain that should have gone into the regular tax network and funneled them instead into the informal system of funding. Although the common people might also suffer from this practice, the main victim was the central government.

When paying the land tax, it should be recalled, each individual taxpayer was required by law to deposit his wrapped tax silver in locked chests set up either at the hsien yamen or at other locations in the chou or hsien established for the convenience of the people. These chests were locked to prevent embezzlement prior to remittance to the capital. The taxes collected each day were entered in registers known as liu-shui hung-pu , and the taxpayer was given a receipt for the amount paid. Two other copies of the receipt were kept at the yamen for comparing with the taxes recorded in the liu-shui hung-pu and with the taxes actually found in the chests when they were opened. That these methods of cross-checking were taken seriously by provincial officials is testified to by the care they exercised to cover their tracks when taxes were "stolen." It was common for magistrates to take a portion of the taxes paid and report it to his superiors as unpaid (wei-wan ). This was accomplished either by fei-sa , by the use of weighted scales, or by lifting whole packets of taxes from the chests by means of specially designed hooks.^[58] The evidence of such malpractices was hidden by issuing receipts to the people which did not bear the official seal^[59] or by writing one amount on the official receipt given the taxpayer and a smaller figure in the records kept by the yamen.^[60]

These taxes were then reported as people's arrears (min-ch'ien ). Because the sanctions against an official reporting arrears were less severe than those incurred for embezzlement or for having deficits, it was usually to the magistrate's advantage to appropriate regular taxes illegally rather than be found lacking in the funds with which to carry out his official duties.

Another closely related method used by magistrates to conceal deficit spending was the practice of "shifting funds" (no-i ) and using newly collected taxes to cover up old shortages (i-hsin yenchiu ). When an official found himself short of funds he might delay reporting his annual accounts. Because the annual accounting (tsou-hsiao ) by the governor was sent to the Board of Revenue no earlier than the fourth month of the following year, by the time the documents from the province had reached the Board and a Board response had been sent back to the province it was usually winter. By then, the magistrate would have begun collecting the taxes for the next year and could use a portion to pay the shortages amassed the year before.^[61] This process was repeated year after year, setting in motion an endless cycle of deficits and concealment. The central government rarely received its full quota, and the magistrate could avoid being implicated in malfeasance before he was transferred.

When the magistrate did report people's arrears, whether real or fabricated, he could easily profit from the way in which arrears were collected. The central government being concerned only with recovering the total taxes not yet paid, it left the matter of who owed what to the individual magistrate. In order not to burden the taxpayers so heavily with back payments that they could not pay the new taxes due, arrears were generally collected in installments (tai-cheng ) over a period of from eight to ten years. A fixed total was set for each year, but magistrates often pressed the people to pay more than their installment for a particular year. The magistrate could then keep the difference and would long be out of office before the people began to complain that they had already paid in full.^[62]

In addition to the land and head tax, magistrates were also responsible for the collection of a number of commercial taxes called tsa-shui .^[63] Although they were supposed to be remitted in full to the central government, most magistrates sent a fixed amount established by custom and retained the rest.^[64] This was in part a hedge against fluctuations in the volume of market transac-

tions. Magistrates feared that if they remitted a large quota for several years they would be accused of embezzlement at times when business was slow and tsa-shui revenues decreased. However, the amount retained by magistrates was often as great as, or greater than, the amount they remitted. For example, Yun-kuei Governor-general Kao Ch'i-cho reported that the regularly remitted quota of miscellaneous duties in Yunnan was about 14,700 taels. However, a prefect might have a surplus beyond the quota of between 2,000 and 6,000 taels. Likewise, the surplus collected by a chou or hsien usually ranged from 40 to as much as 600 taels.^[65]

Land reclamation provided an excellent opportunity to manipulate the tax rolls in order to divert taxes from the regular tax network into the hands of chou and hsien magistrates. Hiding reclaimed land in order to keep it from inclusion in the Fu-i chüanshu was not the only means available to profit from economic recovery and expansion.^[66] In some instances, magistrates did report reclaimed land. According to the law, a taxpayer had from six to ten years, depending on whether wet or dry cultivation was being used, from the time his land was entered in the tax rolls to the time he actually began paying taxes (sheng-k'o ). This was in recognition of the fact that land recently opened to cultivation required a number of years before it could reach full productivity. Magistrates took advantage of this regulation to begin tax collection immediately, and added these taxes to their other illegal revenues until the stipulated time to remit them to the central government was reached.^[67]

Magistrates were also in a position to use imperial tax postponements and cancellations to their own benefit. When a suspension of taxation was declared, the chou and hsien would occasionally delay making the Board of Revenue's notification public and continue collecting taxes as before.^[68] Some magistrates would even falsify reports of natural disasters in the area under their jurisdiction in the hope of receiving a tax remission for that year. Because productivity had not actually been affected, they would then be able to collect the exempted quota and keep it for other expenses.

In regions where the reed-land tax (lu-k'o ) was levied, it was particularly easy to manipulate the tax quota.^[69] In provinces such as Kiangsu, Kiangsi, and Hukwang, with sandy land along the banks of rivers that would not support the cultivation of food crops, the people planted reeds and paid taxes at a lower rate than on other types of land. Because this land directly abutted river

beds, it was constantly being built up and washed away. In order to balance out losses to the reed-land tax quota when the land eroded, magistrates generally did not report new additions to such land. Taxes collected on the new land could then be added to the other sources of informal income at the magistrate's disposal. Some officials used this as an opportunity to exact bribes from those whose land had eroded before granting them a tax exemption and from those opening new reed land in return for keeping their land off the tax rolls.

At all levels of the official hierarchy, whenever funds were allocated or purchases made, various methods were used to "save" money from the officially available funds. No real revenues were generated in this way, but it enabled officials to circumvent the rigid system of accounting established in conjunction with the regular revenue-sharing system. This process was most important in providing informal funds for higher provincial officials, especially the financial commissioner, but magistrates were also able to exploit the local population by saving money in the procurement of supplies for the yamen and post stations. Some magistrates went so far as to commandeer supplies, granting no payment at all to the merchant, craftsman, or farmer who provided them.^[70] In other cases, the people were paid an official price (kuan-chia ) below the market price (shih-chia ) of the goods or services.^[71] In situations where magistrates made purchases of local products for the central government, they were allocated funds based on rates set by the Board of Revenue (pu-chia ). Unlike the kuan-chia set by the local officials themselves, during the early Ch'ing this Board price was often higher than the rate at which magistrates could procure the required item. Moreover, because market scales usually weighed lighter than the scales issued by the Board for use in the official yamen, an additional discount was possible.^[72] The funds saved in this manner were also added to the informal network.

A number of other irregular sources of funds were available to magistrates. All chou and hsien maintained one or more granaries to provide the people with relief in times of famine and in order to inject grain into the market to stabilize prices. The stores contained in these granaries were less easily measured by inspecting superior officials than were the contents of government treasuries. For this reason, some magistrates kept the silver obtained from sales of government grain instead of replenishing stocks at harvest time. Others accepted cash payment of the taxes and contribu-

tions by which the granaries were to be filled and never bought grain to fill them.^[73]

Contributions from wealthy households to help finance government projects were a regular part of the legal system of government finance under the Ch'ing. Sometimes, however, officials oversubscribed donations for a particular project, saving the remainder of the funds for unrelated expenses. Magistrates were even known to force wealthy households to make contributions, assessing the amount they should provide in accordance with the amount of land recorded under their names in the tax registers.^[74] Commoners making grain contributions in order to purchase an official degree were often charged an extra fee before they could receive the certification of their new rank.^[75]

Like all officials, magistrates were also the recipients of numerous customary fees and gifts (lou-kuei, kuei-li ), some of them from their clerks and runners. It is obvious that without the aid of clerks and runners the magistrate would not have been able to engage in many of the above practices. The necessity to acquire funds by illegal means lessened the magistrate's control over his own staff. If clerks and runners were instructed to tamper with the taxcollection process for the benefit of the local administration, they could equally do so to benefit themselves. Just as we cannot determine what portion of funds so acquired was pocketed by the magistrate and what portion was used for government business, it is impossible to determine the magnitude of clerk and runner engrossment. At least some of the funds they amassed by illegal means were used for their own legitimate stationery and travel expenses. However, the measures they employed to extract money from the people often worked against the interests of the magistrate and the government as a whole, and the fruits of their manipulation of the tax system were less likely to enter the nexus of funds used to operate the provincial administration. For this reason, the role of clerks and runners in the informal funding network more properly forms a part of our examination of those factors that obstructed the collection of taxes at the local level.^[76]

High-ranking Provincial Officials

Few high-level provincial officials were involved directly in the process of collecting taxes from the general population. In order to operate their yamen, support their families, and perform the nu-

merous public services for which they were responsible, they had to rely on transfers of funds from subordinate administrative units through the informal funding network. These transfers were of four main types: (1) revenues submitted directly to a superior official by a subordinate in the form of gifts and fees; (2) "surplus" from major specialized revenue-collecting agencies, namely the customs and salt administrations; (3) percentage deductions from taxes and goods remitted along the regular taxation network; and (4) funds skimmed off in the process of purchase and allocation.

Customary Fees

In the broad sense, the term customary fees (lou-kuei ) may be applied to all illegally obtained revenues. In official correspondence denouncing corruption, this usage was fairly common. However, lou-kuei had a more limited meaning of fees and gifts sent on a regular basis by one member of the bureaucracy to another. The distinction is an important one, because it was primarily lou-kuei in its narrow sense that was ardently attacked by the Yung-cheng reformers.

All officials and functionaries, from the lowest assistant magistrate to the governor-general, periodically sent gifts of customarily established sums of silver to their superiors. These gifts were usually sent on the occasion of an official's birthday (sheng-ch'en kuei-li ), or his arrival at a new post (tao-jen ho-li ), when paying a visit to the official (piao-li ), and at the time of the four major festivals of the lunar year (chieh-li ).^[77] These gifts could amount to a considerable sum, especially when the recipient was a high provincial official such as the governor or financial commissioner. In YC 2 (1724), the Liang-kuang governor-general confessed to having received 47,110 taels in chieh-li from his subordinates.^[78] The total annual chieh-li sent to the Kwangsi governor was 12,400 taels. In addition, he received gifts upon arriving in office equivalent to the chieh-li for one season, and surplus from the unloading duty (lo-ti shui ) collected in Kuei-lin, P'ing-le, Hsün-chou, and Wu-chou prefectures of 7,000 taels per year.^[79] Shantung Governor Sai-leng-e reported that the former financial commissioner of that province had accepted 9,784 taels in gifts from the chou and hsien, and that the former educational commissioner had been the beneficiary of 3,204 taels from the same source.^[80]

Even when the immediate source of chieh-li was an official

above the chou and hsien level in the official hierarchy, it must be noted that the original source of all customary fees of this kind was the magistrate, and ultimately, the people. The magistrate sent customary fees of varying small amounts to every official in whose jurisdiction he was located. He sent ts'ao-kuei to the grain intendant^[81] and i-kuei to the post intendant,^[82]chieh-li to his prefect and circuit intendant, as well as the customary fees he remitted directly to the financial and judicial commissioners, the governor, and the governor-general. Likewise, all those officials above the magistrate consolidated the fees they accumulated from their subordinates and sent a portion to each of their superiors.

In addition to the fees sent to higher officials, subordinates usually contributed to the operating costs of the yamen above them in the form of "food silver" (fan-shih yin ) for the superior yamen's clerks. This practice probably grew out of the custom of sending a small sum whenever business was transacted with a superior yamen, to cover the food and stationery costs of the clerks who were confined to their offices while processing the documents or taxes sent by lower officials. Eventually these contributions became regularized, and commonly accompanied tax remittances as a fixed percentage of the quota.

The names applied to such fees varied from province to province. In Kwangtung the chou and hsien sent an extra 3 taels for every 100 taels remitted in taxes to the provincial treasury. Of this sum, 10 percent was deducted as hsiao-fei yin and used to pay the expenses of the treasury keepers (k'u-ta-shih ), treasury clerks (k'uli-shu ), boatmen, cooks, messengers and runners, guards, and scalekeepers associated with the financial commissioner's yamen.^[83] In Chihli, a customary fee called hsiao-fan yin of 4 taels per thousand was sent to the treasurers and treasury clerks when local taxes were transported to the provincial seat.^[84] In the customs administration, food silver was also collected in addition to the regular tax quota. For example, at the Huai customs an extra 16,404 taels was received in the form of fan-shih yin to pay the daily living expenses and stationery costs of the customs clerks and patrols. Part of this sum was also remitted to the superintendent of customs to cover his own living expenses and administrative costs.^[85]

Besides the food silver for their own yamen, the governors and financial commissioners had to set aside customarily agreed-upon amounts of fan-shih for the clerks in the central-government ofrices with which they regularly transacted business. The judicial

commissioner sent stationery and food money (chih-pi fan-shih ) to the Board of Punishments. Governor Yang Wen-ch'ien reported that prior to the Yung-cheng reforms, the chou and hsien in Kwangtung annually remitted 6,042 taels for this purpose. These funds were used to cover the costs of remitting fines and confiscated property to the Board, and to provide food silver for the Board clerks and travel expenses for officials and runners on government business. Contributions from the province also paid the costs of compiling the records for the autumn assize of capital offenses and provided gifts of ink and stationery for prisoners.^[86]

At the time of the annual accounting, the financial commissioners of every province remitted fan-shih yin to the Board of Revenue and the Office of Scrutiny of the Board of Revenue. Szechuan sent 1,300 taels for this purpose.^[87] Honan-Shantung Governor-general Wang Shih-chün claimed that there was no fixed quota prior to 1728, but in that year it was established that 3,600 taels should be sent to the Board of Revenue and 600 taels to the Office of Scrutiny.^[88] Moreover, when purchases in kind were made for the central-government storehouses, additional sums were allocated for the travel expenses of the officials escorting the shipment and for Board fan-shih . This so-called pang-fei could be quite large. For Honan's annual shipment of paints, fabrics, cattle tendons, and wax, the pang-fei averaged around 5,300 taels.^[89]

Although lower officials contributed to the expenses of the clerical staff in their superiors' yamen, it was not uncommon for high officials to play a part in the selection of clerks in their own and subordinate yamen. In return they received "gifts" (kuei-li ) from the clerks appointed. In effect, the clerks paid a fee for the purchase of their posts. Most of the existing evidence of this practice relates to Kwangtung province, where an intensive investigation was conducted as part of a high-level impeachment case. Other provinces, however, probably had similar customs enabling high officials to take advantage of the competition for government jobs to add to their own informal income.

In Kwangtung it was the local officials who posted the notices inviting local people to apply for vacancies in the ranks of lowlevel functionaries and clerks. Although they did the actual hiring, every appointment had to be approved by the financial commissioner. In his letter of authorization, the financial commissioner specified the amount of kuei-li he expected from the lucky candidate in accordance with the level of the post being filled. The fees

demanded varied widely and the way in which this practice was carried out may have depended on who was financial commissioner at the time. In one report, dated YC 5 (1729), Liang-kuang Governor-general K'ung Yü-hsün testified that this fee could be as little as a few tenths or hundredths of a tael, but was known to rise as high as seventy or eighty taels for some positions. K'ung called the fee hsin-hung and said it served to purchase the red ink used to apply official seals to documents.^[90] During an investigation of official corruption in Kwangtung, Yin Chi-shan reported that the amount of kuei-li generated in this manner really ranged from several tens to several hundreds of taels per applicant. The combined fees collected by the financial commissioner were divided monthly with the governor and governor-general and, according to Yin, were used to defray the costs of operating the yamen of all three officials.^[91]

Kuei-li was particularly steep when appointments were made in the customs administration. These posts were highly coveted as they provided the clerk with access to a large amount of fees and graft in the collection of customs duties. Fu T'ai, when governor of Kwangtung, received 300 taels from each of the five clerks appointed to the Canton customs during his term.^[92] The Huangchiang customs clerk customarily paid zoo taels each to the financial commissioner, governor, and governor-general to purchase his post, because it was considered very lucrative and the term of office was long. Under investigation during the mid-1720s, Financial Commissioner Kuan Ta was accused of extorting 1,250 taels from a clerk newly appointed to this position. Of this, only 240 taels each, slightly more than the customary amount, was sent by Kuan to the governor and governor-general.^[93]

Contributions from the Customs and Salt Administrations

Besides the various categories of lou-kuei sent by subordinate officials, a large portion of high provincial yamen expenses was offset by customary contributions from the customs and salt administrations operating within their territory. It was chiefly in this way that the provinces were able to tap the growing wealth of the commercial sector in the early Ch'ing period.

The income generated from customs duties was transferred to high provincial officials in two forms, customary fees (shui-kuei ) and surplus (ying-yü ). Shui-kuei was not unlike other customary

fees. It was sent by a subordinate to a superior, often on special occasions, the annual remittance commonly being agreed upon in advance.^[94] Sometimes these fees were sent by officials other than customs officials who were responsible for collecting specific duties. For example, the Shantung financial commissioner received 1,500 taels each year from the incense tax (hsiang-shui ) collected by the magistrate of T'ai-an chou.^[95] However, the largest portion was sent by officials in charge of customs-collection stations. In Fukien, after deductions were made to cover the costs of running the customs administration, the total income from customary fees was divided into four shares. Two shares were sent to the governor's yamen, one share was divided among the deputy customs inspectors, and one share was apportioned among the messengers and personal servants working in the customs administration.^[96] The Canton customs was especially prosperous and produced an average of sixty thousand to seventy thousand taels annually in customary fees, which were also divided among those working in the customs stations and high government officials.^[97] In Shantung the customs stations handled a far smaller volume of trade. Besides its regular tax quota of 7,540 taels, the Chiao-chou customs sent the governor 3,676 taels in kuei-li yin and Lai-yang sent 497 taels along with its 770-tael quota.^[98] Sometimes the lou-kuei from customs was sent to a lower-ranking official with supervisory responsibilities for particular duties and he sent a portion to the governor and governor-general. Such was the case in Szechuan, where the judicial commissioner sent the governor-general 2,500 taels in salt and tea fees (yen-ch'a kuei ) and the financial commissioner sent him 4,040 taels from the fees on exported rice (liang-kuei ) and customs (shui-kuei ).^[99]

Where did the funds to pay these customary fees come from? For the most part they were derived from surcharges levied on merchants in addition to the regular tax. During an investigation of the customs station at the T'ung pass it was discovered that officials had long been in the habit of increasing the rates for duties established by the Board of Revenue. The Board had set the tax for a mule-load of thread or satin at 0.6 tael, but the customary practice was to charge 1.2 taels per load of thread and 1.4 taels per load of satin. On cotton goods the Board authorized a duty of 0.11 tael per load of finished cloth and 0.06 tael per load of raw cotton, but the actual rates of collection were 0.18 tael and 0.08 tael respectively.^[100]

Sometimes lou-kuei was collected as an across-the-board percentage on regular duties. Special names were often applied to such fees, as in the case of the registration fee (kua-hao ) and tankuei collected in Kwangtung^[101] and that of the tan-yin tsa-hsiang found in the records of customs stations in Fukien.^[102] In other instances, these fees were combined under the rubric of wastage allowances (huo-hao, hao-hsien ), the same term used to describe land- and head-tax surcharges.^[103]

Unauthorized fees might also be charged for vehicles and livestock passing through customs barriers. In Fengtien, customs inspectors exacted 0.16 tael for every small cart that arrived at their stations. Of this, 0.08 tael was sent to the metropolitan prefect, 0.02 tael went to the registrar, and the rest was divided among the prefecture's clerks as food money and to cover stationery costs. For a large cart the fee was doubled. Every year an average of 1,600 to 1,700 taels was collected from cart fees alone. A similar fee, 0.03 tael per head, was levied on pigs and sheep passing through the customs barriers.^[104]

The Huai-an customs administration was particularly tenacious in exacting extra duties from merchants. Originally a 2-percent huo-hao was collected when merchants paid their regular duties. However, in 1725 it was reported that when the new inspector, Ching Yuan, came to office he eliminated the wastage allowance and set up five or six small customs barriers (hsiao k'ou-tzu ) along the main transportation route. Whenever a merchant passed through one of the barriers he was required to obtain an inspection certificate (yen-p'iao ), for which he paid a number of fees. If these surcharges, including inspection certificate cash (yen-p'iao ch'ien ), stationery fees (chih-pi ch'ien ), and food money (fanch'ien ) were not paid, the merchant's boats were not allowed to proceed. The fees at each barrier were at least 0.13 or 0.14 tael per tael of regular duties, so that a merchant might end up paying almost double the legal customs tax before he reached his destination.^[105]

Profits were also generated by levying duties on items not designated for taxation in the Board's regulations. By precedent, common people transporting small quantities of rice to large provincial markets were not taxed. However, at the Huai-an customs, a lone peasant carrying rice on his back would be detained until several others came along and their combined load equalled a whole picul. They would then be treated as a merchant convoy and or-

dered to pay the tax.^[106] It was also the custom that silks that had been cut and made into clothes, even when carried by traveling merchants, were to be regarded as personal clothing and not subject to taxation. On the pretext that female clothing so transported had to be used clothing destined for resale, the Huai-an customs taxed all clothing, male or female, old or new. None of these revenues were reported to the central government, but were sent directly to the provincial coffers.^[107]

The Canton customs was in a unique position to exploit profit-making opportunities by levying special taxes on foreign trade.^[108] Besides levies on the value of cargo (shang-huo fen-t'ou ), customary fees were collected for opening the hatches, inspecting the hatches, and unloading cargo.^[109] Compradors and interpreters also paid a public-expense fee (kung-fei yin ) in return for monopoly privileges in supplying services to the foreigners. Those working for boats anchored at Whampoa paid between 40 and 100 taels and those residing in the provincial capital paid from 60 to 160 taels, depending on the size of the foreign vessel by which they were employed.^[110] During the Yung-cheng period it was estimated that almost 70,000 taels were collected in this way, though by that time not all of the revenues were retained by the province.

An even more unusual levy was the so-called "people tax" (jenshui ). During this period, western traders were restricted to residence in Macao. In order to discourage them from traveling inland, the Kwangtung authorities, on their own initiative, imposed a charge of eighteen Mexican silver dollars on every westerner coming to Canton from Macao and an additional two Mexican silver dollars on those traveling from the provincial capital back to their base.^[111]

Several other methods were used by customs officials to accumulate the revenues they shared with the provincial government as shui-kuei . These were quite similar to the practices of magistrates and other officials directly involved in tax collection. Customs stations in Kiangsi were found to be using scales to weigh tax silver that weighed heavier than those used in the financial commissioner's yamen. These heavy scales produced a "surplus from weighing" (p'ing-yü ) of 3 taels for every 100 taels of duties collected.^[112] At the Kan pass, the taxes collected on single articles, ranging from 0.05 to 1.5 taels, were simply not entered in the accounts.^[113] Small bits of silver collected were also consolidated and retained for provincial use.^[114]

By far the greatest volume of unreported customs duties fell into the category of surplus (ying-yü ). Unlike lou-kuei, ying-yü was not the product of official and staff peculation, but of the growth of trade beyond the volume calculated when the government's customs quotas were originally established. As in the case of the surplus commercial taxes collected by magistrates, once the official quota was filled, the officials could amass large sums simply by continuing to collect the duties at the legal rate. Until the reforms of the Yung-cheng period, most of these surplus duties remained in the provinces, although it is difficult to determine what portion was transferred along the informal funding network and what portion was pocketed by officials directly engaged in customs administration.

We can get some idea of the amount of funds that were potentially available to assist in paying provincial expenses by examining the reports concerning ying-yü submitted during the reform period. In YC 11 (1733), the Lung-chiang and Hsi-hsin customs in Kiangnan reported a gross surplus of 78,900 taels beyond the regular quota of 152,000 taels.^[115] Two years earlier the Fukien customs administration boasted a total surplus of 54,516 taels.^[116] Kwangsi was granted 15,000 taels of customs surplus in 1727, but it was later discovered that the Wu-chou and Hsün-chou customs stations annually generated another 31,400 taels.^[117]

Because these surplus funds were carefully hidden from the central government, we do not have accurate reports of their distribution through the informal funding network. We do know that a portion was utilized to cover the costs of wages and supplies, travel expenses, and stationery costs of the customs personnel. Some funds were undoubtedly pocketed by customs officials, as in the case of the official who reported 600 taels of surplus, only to have it revealed that there was an additional 2,100 taels, which he was ordered to return.^[118] However, because the existence of large sums of ying-yü was known to financial commissioners and governors and they were later legally incorporated into those officials' administrative budgets, we can assume that at least a part of this provincial windfall reached these higher officials during the early Ch'ing, either as a direct remittance or as part of the lou-kuei sent to the provincial capital.

One of the most bountiful sources of revenues flowing into the informal funding network was the salt gabelle. Opportunities for profit were numerous, because officials were often intimately in-

volved not only in marketing salt but also in the process of manufacturing it. Because they owed their strong economic position to a government-sponsored monopoly and relied on the government to protect them against bandits and smugglers, salt merchants were also a constant source of "voluntary contributions" (le-chüan ) to the provincial coffers.^[119] Moreover, salt manufacture and marketing brought the merchants into contact with officials at all levels of the bureaucracy. In part, this accounts for the fact that customary fees from the salt administration, unlike customs fees, were not confined to high-level officials.

There were three major sources of informal income from the salt administration: (1) surplus income from the sale of salt, (2) loukuei collected by officials and clerks employed in the salt administration, and (3) contributions from salt merchants on an annual basis and for specific projects.

In China, salt was sold by specifically designated salt merchants. Merchants were required to obtain salt certificates (yen-yin ) by payment of the gabelle (yen-k'o ). Each certificate entitled the bearer to sell a certain amount of salt in a specified area.^[120] The income from the sale of salt certificates was fixed by quota and remitted to the central government. However, the income from salt sales was often not limited to the legal quota. Either through percentage fees on the sale of certificates or by allowing additional sales of salt beyond the quota, the provinces were able to accumulate a salt-revenue surplus which was retained locally for public expenses. In Ho-tung, the regular salt-gabelle quota was around 171,700 taels. Income from the sale of excess certificates was as high as 104,300 taels.^[121] The annual salt quota for Chen-chou prefecture in Hunan was 1,375 taels, but it was discovered that the actual amount collected was about twice that figure.^[122] In Yunnan and Kwangsi, official participation in the actual marketing of salt resulted in sales far beyond the limits set by the Board of Revenue. Soon after the Yung-cheng emperor assumed the throne, a scandal erupted in the Yunnan salt administration. According to Metropolitan Censor Chao Tien-tsui, the salt gabelle from the nine salt wells in Yunnan yielded an annual quota of 168,145 taels. The amount of salt to be sold in each chou, hsien and prefecture was also fixed by quota. However, because the management of the salt wells was in the hands of the provincial administration, the governor, governor-general, and salt intendant had authorized the production of extra salt and increased the quantity of salt sold in the

province's retail shops. This had resulted in sales of "private" salt far in excess of the official salt, the profits from which went to the central government. Even if the official salt went unsold, there was pressure on the shopkeepers to sell the governor and governor-general's "private" salt. In Yunnan this was possible because salt was not marketed by salt merchants but by runners and clerks deputed by the salt intendant.^[123] In Kwangsi the salt was also transported and wholesaled by deputies of high provincial officials. For every sack (pao ) of salt authorized for sale by the quota, an extra eighteen chin was added. This extra salt was called p'ing-t'ou and the proceeds from its sale, like the proceeds from the sale of private salt in Yunnan, were used for provincial expenses.^[124]

It is difficult to determine the exact nature of the lou-kuei exacted by officials and runners employed in the salt administration. Nevertheless, references to the elimination of lou-kuei after the Yung-cheng reforms are fairly common.^[125] Both lou-kuei and part of the surplus described above were used to defray the operating expenses of salt-administration personnel. During his term as superintendent of the Liang-huai salt administration, I-la-chi reported disbursing 12,485 taels in salt-customary fees for officials' salaries, runners' wages, banquets, awards to merchants, and for the relief of salt workers who were the victims of floods and other disasters. He also set aside 3,500 taels for his own daily living expenses and traveling costs, as well as 19,284 taels for the food money and stationery costs of his yamen's registrar, archivist, clerks, and runners.^[126] The Shantung salt controller's yamen received 37,000 taels in kuei-li , from which similar expenditures were made.^[127]

Besides being the conduit for income generated by the sale of salt, the salt merchants themselves sent customary fees called yenkuei to provincial officials. The officials to whom yen-kuei was sent and the amount of the fees varied from province to province, but it would not be an exaggeration to say that almost every official from the hsien magistrate to the governor-general benefited from salt fees. In Kiangsi, for example, the governor's yamen was sent an annual yen-kuei quota of 16,00 taels. The financial commissioner received 3,000 taels, as did the judicial commissioner. The salt and post intendant had special responsibility for salt affairs and so received a total of 4,729 taels in gifts from salt merchants. The combined yen-kuei of the grain intendant, the Jaochiu and Nan-kan circuit intendants, and the prefects and assistant prefects of five of Kiangsi's prefectures was 7,453 taels. In

addition, the remaining prefectures and the chou and hsien also received an unspecified amount of yen-kuei .^[128]

The customary fees sent by salt merchants to high provincial officials were often quite large. The total yen-kuei in Hukwang was 160,000 taels, most of which was remitted to its two provincial capitals.^[129] The annual quota of the Chekiang governor and governor-general was over 25,000 taels.^[130] In Yunnan, the governor-general was sent 17,000 taels annually, and the governor's share of yen-kuei was 60,000 taels. To this was added another 6,000 taels sent directly to the provincial seat from two of the province's salt wells.^[131]

In addition to the customary fees paid on a regular basis, merchants were often called upon or volunteered to make contributions to supplement public expenses. Their wealth and their special debt to the government placed salt merchants in the forefront of commercially based philanthropy. This was especially true when their home areas were struck by natural disaster or required funds for a major construction project, or when the merchants themselves were the beneficiaries of extraordinary imperial favors. In Shensi, after the emperor increased the number of salt certificates that could be issued in that province, the salt merchants offered to contribute 5,000 taels annually to assist in defraying local administrative expenses.^[132] When a similar increase in the amount of salt that could be sold for each certificate was granted in Kiangnan, the Liang-huai salt merchants petitioned to contribute 240,000 taels to the charity granary maintained by the salt administration in their region, as well as 80,000 taels to Governor Ko-ert'ai to offset yamen expenses.^[133] Several years later, these same merchants also contributed to the river-conservancy administration in Kiangsu to aid in the repair of embankments.^[134]

Thus, the business of salt production and distribution played an extremely important role in supplying revenues for the informal funding network. As we shall see, even when other forms of lou-kuei were under violent attack during the period of the Yung-cheng reforms, it was very difficult for provincial officials to conceive of operating their yamen without yen-kuei .^[135]

Percentage Deductions from Remitted Taxes

Just as magistrates levied surcharges when collecting taxes from the people, higher provincial officials who received local taxes collected a portion of their own operating expenses in the form of

remittance fees from subordinates. The main recipient of these fees was, of course, the financial commissioner.

Remittance fees appear to have existed in every province, although once again the names given to such fees and the amounts collected varied greatly. For every 100 taels of taxes remitted by the chou and hsien in Kwangtung, an extra 3 taels was added. This three-percent surcharge was called ta-p'ing yin .^[136] In Shantung, when the chou and hsien delivered their land and head taxes they sent chieh-fei to the financial commissioner. The amount ranged from 1.1 to 1.2 percent of the total tax quota, depending on the tax potential of the area. Along with this, a shipping fee (yün-fei ) of five- or six-tenths of a tael was remitted to cover the costs of transporting the taxes to the Board in Peking. Any funds left over were funneled into the informal funding network.^[137] Remittance fees were also charged when miscellaneous taxes were sent to the financial commissioner and when higher officials such as prefects and circuit intendants sent in fines and confiscated property.^[138]

Both tax revenues and administrative expenses in Kiangsu were among the highest in the empire. The Suchow financial commissioner not only received an item called "food money accompanying the regular tax" (sui-cheng fan-shih ), but also collected a 1.3-percent surcharge known as tso-p'ing yin , which was used to supplement the costs of remitting taxes to the capital and to offset expenses within the province itself.^[139] Chang Tan-lin reported that during his predecessor's one and one-half years as financial commissioner these two fees alone produced a total of 76,143 taels that were made available for provincial public expenses.

In addition to the revenue accumulated through direct levies, most provinces were able to amass a small surplus when the taxes received from all the chou and hsien were melted down into large ingots for shipment to the imperial capital. This surplus was generally known as p'ing-yü or ping-feng yin . The silver added to the provincial coffers in this way could add up to as much as several thousand taels each year.^[140]

Besides surcharges on the actual remittance of taxes, there was a wide range of possibilities for levying fees connected with the accounting and record-keeping functions of the provincial yamen. We have seen that all chou and hsien were expected to pay fees at the time of the annual accounting (tsou-hsiao fei ). The methods used by financial commissioners in Kwangtung were not unlike those devised by officials in other provinces to extract additional

revenues from their subordinates. When the grain collected from military-colony land was allocated to the military as rations (ping-mi ) an inspection fee (yen-pi yin ) of 0.03 tael per picul was sent to the financial commissioner.^[141] When prefects and magistrates collected miscellaneous duties they had to obtain official "two-stub receipts" (shuang-lien shui-tan ) from the financial commissioner on which to record income and issue proof of payment. For every 100 sheets he distributed to his subordinates, the financial commissioner collected a customary fee of three taels. A similar fee of five taels was exacted for every 100 sheets of contract paper (ch'i-wei ) issued by the commissioner for use by the chou and hsien in recording land and property sales. Moreover, assessments totaling 6,000 taels were levied on subordinates at the time of the quin-quennial census.^[142]

Other high officials with revenue-collecting responsibilities were also privy to percentage deductions. For example, in Honan, the grain intendant received 14,568 taels per year in "public-expense silver" (kung-yung yin ) from the chou and hsien when they sent in their tribute rice. When grain shipments were sent to the intendant for storage in provincial granaries an additional 1,456 taels was remitted as sui-feng yin , along with 1,720 taels of kung-yung yin .^[143]

Once revenues transmitted by the chou and hsien and other subordinate administrative units reached officials such as the financial commissioner and grain intendant, they did not become their exclusive property. Rather, these tax-receiving yamen served as conduits in the flow of funds to even higher yamen. In Shantung, most of the funds received in the form of transportation fees (chieh-fei, yün-fei ) and food money (fan-yin ) were consolidated under the name of fen-kuei because they were shared (fen ) by the financial commissioner, governor, and governor-general.^[144] Likewise, the ta-p'ing yin deposited in the provincial treasury in Kwangtung was divided among the governor-general, governor, financial commissioner, and judicial commissioner.^[145] Kiangning Financial Commissioner Chüeh-lo Shih-lin reported that he had received 12,900 taels in sui-p'ing yin from the chou and hsien in Kiangsu province, from which a fixed fee (p'ing-kuei ) of 1,500 taels was sent to the governor-general and 2,700 taels was sent to the governor annually.^[146]

These fees were small in comparison with the revenues derived from customary fees from subordinate officials and merchants.

Nevertheless, they were a source of much-needed funds during a period of transition when corvée labor was no longer available to undertake the transportation of taxes. The growth of these fees to provide a surplus to help defray the other administrative expenses of high provincial officials is just one instance of how the informal funding network adjusted to and utilized the existing fiscal system to supplement statutory appropriations.

Funds Skimmed Off in the Process of Purchase and Allocation

During the late-imperial period, tax collection in China became increasingly monetized. However, there were a number of items which, while collected in cash at the local level, had to be remitted in kind to the military or the central government. During periods of high productivity and low commodity prices, the reconversion of these taxes to staple produce afforded the provincial government an excellent opportunity to make a profit. This profit was commonly called chieh-sheng yin or "saved silver," and was viewed as a legitimate addition to the informal funds accruing to the financial commissioner's treasury.

Several examples of this practice may be cited for the Kiangnan region. In both Kiangsu and Anhui, many chou and hsien paid their tribute grain in cash instead of rice. In Anhui, a conversion rate was set at 1.3 taels per picul of rice. However, in a year of good harvests as little as 0.94 tael could buy a picul of high-quality white rice on the open market. At a saving of 0.36 tael per picul, Anhui Governor Ch'eng Yuan-chang reported a profit of over 7,000 taels.^[147] A similar saving of 7,213 taels was reported by the Kiangning financial commissioner when tribute rice was purchased in Kiangsu.^[148] Savings were also possible when purchases of beans were made for feed for the horses of banner forces stationed in the province. A particular strain of bean was required. Inasmuch as it was not produced locally, the taxpayers were permitted to pay at a rate of 1.2 taels per picul. When the market price was low, the purchase of these beans yielded a profit of almost 3,000 taels.^[149]

Another source of savings from the regular tax quota closely related to chieh-sheng yin was a category of income called p'ing-yü . The use of the term p'ing-yü in this instance probably derived from the fact that this revenue was the result of differences between the weights used on yamen scales (k'u-p'ing ) and those used in the marketplace (shih-p'ing ). When supplies and taxes were dis-

bursed to other agencies of government the financial commissioner normally used the scales he was issued by the Board of Revenue. Thus, at least in theory, all payments between yamen were calculated in standard weights and measures. However, when funds were disbursed to make purchases from the people, payment was made according to the scales in the markets.^[150] The market scales in Kiangning were about 2 taels per hundred lighter than the treasury scales. Whenever purchases of goods for shipment to the capital or of materials for construction and repairs were made, this differential produced a saving from the Board price that was deposited in the financial commissioner's treasury.^[151] During his term as financial commissioner, Chüeh-lo Shih-lin accumulated 2,675 taels in this way.^[152]

P'ing-yü was a particularly common source of funds in the yamen of river-conservancy officials, grain intendants, and other medium-level provincial officials who made large purchases of supplies and materials from the province's share of regular taxes.^[153] The Kiangnan superintendent of river conservancy reported that the old scales used in his yamen had been 1.8 taels heavier than the local market scales. These scales had in fact been below the Board standard weight, and the new weights sent by the Board of Revenue during his term in office yielded 2.5 taels more than those commonly in use among the people. Consequently, when the people paid money to the government, it was by yamen scales heavier than those in use in the marketplace. When they received payments from the government, it was invariably according to the lighter market scales. The provincial government profited at the expense of both the people and the central government, with neither really feeling the pinch.^[154]

A remarkable characteristic of the informal funding network was the extent of the regularity with which it functioned. Both the paths of transmission and the types and amounts of fees and payments were standardized, and they were followed fairly consistently over decades despite numerous changes in official personnel. This was by no means the "catch as catch can" form of private aggrandizement that some contemporary critics and modern scholars have led us to expect. Nevertheless, however well this system functioned, it was bound to have adverse effects on government in the early Ch'ing.

Some of the methods used to raise funds for use within the provinces did not create any new income, but rather moved around

taxes that were meant for other purposes. When a magistrate shifted revenues designated for the purchase of relief grain and used them to pay laborers repairing dikes, it was with the hope that famine would not recur while he was in office. His actions were not corrupt, in that the funds were put to beneficial use, but he did violate the administrative code which set rigid guidelines as to how the funds allotted to him were to be spent. Worse, the fact that his actions had to be taken in secret meant that the shortage in granary stores would probably not be discovered until flood or drought made the discovery too late.

From the central government's point of view, methods such as "collecting more and reporting less" and "using newly collected taxes to cover up old shortages" were especially insidious. By creating spurious arrears, they directly contributed to deficiencies in central-government coffers. Furthermore, because the deficits of officials were camouflaged as evasion by the taxpayers, retrieval of missing funds was made more difficult and an undue burden was placed upon the people.

Practices such as these accounted for many of the deficits in central-government treasuries discovered by the Yung-cheng emperor in 1723. However, the other side of the informal funding structure was also damaging to government finance. Whether transmitted as lou-kuei , as p'ing-yü , or as fan-shih , a large portion of the funds that made their way up the provincial hierarchy originated in one form or another as surcharges levied on the common people. When surcharges became too heavy, regular taxes went unpaid. When surcharges became accepted as a source of funds for legitimate expenses, it was easy to take a bit more to line one's own pockets. Because the informal funding network was parallel to, but not integrated with, the regular system of taxation and allocation, it fed on the statutory tax structure but lacked any formal means to check abuses by individual officials or to distribute more equitably the revenues accumulated. Inasmuch as almost all the sources for funding this network were illegal, even when officials were scrupulous in their appropriation of these revenues, their very acquisition presented a threat to both official discipline and the health of central-government finances.

The weakest link in the system was at the level of the chou and hsien. Although most magistrates received their positions from the Board of Civil Appointment, their greatest concern, once in office, was to keep in the good graces of their superiors in the provinces.

It was they who held the power to impeach a magistrate for infractions of the disciplinary code. But their willingness to do so was impeded by their dependence on the chou and hsien as a source of informal revenue. Under conditions where high provincial officials could operate their own yamen only with the aid of funds received from their subordinates, a symbiotic relationship developed that was based on bribery from below and coverups and favoritism from above. It was this convergence of interests that formed the backbone of the informal funding system and made it work. The county magistrate's treasury thus became the trough at which all other officials fed. The magistrate in turn was dependent on his runners and clerks to fill the trough with the bits and pieces of silver they extracted from the multitude of taxpayers. Fundamental reform of the regular system of revenue sharing might reduce the need for many of the fund-raising devices associated with the informal funding network, but the old and profitable habits it fostered among yamen underlings would be much harder to break.

In 1723, most of those in power still saw only the symptoms and not the disease. They thought that corruption was causing shortages and did not see that it was actually shortages that had necessitated corruption. Only when the Yung-cheng emperor launched his vigorous attack on the symptoms was the true nature of the illness gradually revealed.

3—
The Return of the Meltage Fee to the Public Coffers

The informal funding network was one of the most telling manifestations of the contradictions inherent in the traditional Chinese political economy. In its commitment to benevolent rule, the Ch'ing dynasty had doomed the Chinese people to an uncertain existence, prey to the demands of officials, runners and clerks, and numerous local power brokers who took advantage of the government's weak financial condition to act as middlemen in the taxation process. Only by recognizing the need for a thorough reorganization of the revenue-sharing system could the government hope to combat the evils of a corrupt bureaucracy. Yet to do so would mean a rejection of some of the basic principles of good government upon which the dynasty's legitimacy depended.

It was not ignorance that prevented a more rational approach to local fiscal administration. Many of the reforms initiated in the early Ch'ing were specifically designed to eliminate the most conspicuous abuses associated with the informal funding network. Surcharges were prohibited as early as the first year of the Shunchih reign, and the K'ang-hsi emperor warned twice against their collection, permitting the people to report officials who persisted in this practice.^[1] Needless to say, the practice continued, but the aversion to high taxation was so strong that many in government preferred to close their eyes to the evidence of surcharges rather

than admit their necessity and make them legal. The K'ang-hsi emperor himself stated that "If a magistrate collects a 10-percent meltage fee (huo-hao ) and does not collect a bit more, he may be called a good official."^[2] However, when Shen-kan Governor-general Nien Keng-yao took this opportunity to request the legalization of meltage fees, the emperor demurred. Nien pointed out that the level of illegal surcharges collected in his jurisdiction had reached as much as 40 or 50 percent of the regular tax quota. Inasmuch as magistrates obviously needed additional funds, he suggested that they be allowed to keep a portion of the current meltage fee to cover their yamen expenses and make up deficits. K'ang-hsi reminded Nien of the stiff penalties for the collection of "private exactions" (ssu-p'ai ), but could offer no concrete solution to the problem.^[3]

This is basically a private matter. I once said to Ch'en Pin that a 10 percent meltage fee seemed tolerable. He felt this was a great benevolence, but I could not permit him to make my words public. Today, though Nien Keng-yao has sent a secret memorial, if I write a rescript it will have the same precedent-making effect as if I granted a request presented in a routine memorial. How can I bear the stigma of one who approves of surcharges?

K'ang-hsi could not deny the logic of Nien's request, but to say so would be to sanction the legalization of increased taxes and recognize the legitimate need for surcharges. Moreover, although his government did employ legal means to control the behavior of officials, the emperor's statements reflected a very narrow interpretation of Confucian principles of good administration. Thus, institutional change was not the ultimate solution to corruption. Only personal restraint on the part of moral officials would serve this purpose. It was for this reason that the emperor preferred to relegate the issue of surcharges to the private realm. This was also why the emperor could praise an official if he limited his "private exactions" to a minimal amount. What the emperor failed to confront was why a good official should have to collect any surcharges at all.

If K'ang-hsi's rejection of institutional reform was based on principle, there were many in the bureaucracy with a more immediate interest in concealing the magnitude of local need and local

corruption. Magistrates often welcomed the insecurity of the informal funding network rather than risk the increased control by central-government officials that rationalization would entail.^[4] For their part, the latter had little understanding of the administrative difficulties or fiscal requirements of their provincial subordinates. Anxieties over the health of their own treasuries caused them to underestimate local needs, and ignorance led them to suspect the motives of the few who did take up the cause of local fiscal reform.

Ironically, it was the informal funding network itself that was probably most responsible for the persistence of the fiscal inequities in response to which it had evolved. The real requirements of local administration were masked by the success with which informal methods provided local and provincial officials with additional revenues. Pleas for more funds for public business rang less true at court when one's predecessors seemed to fare perfectly well on what they had. More important, the techniques used within the informal funding system to raise revenues necessarily sabotaged the statutory checks on corruption and destroyed the utility of tsuo-hsiao accounts as an accurate reflection of local expenditure. Before fundamental reform could take place, the smokescreen created by the informal funding network had to be dispersed. One precondition for this was a new emperor whose approach to government was different from that of his father. A necessary catalyst was a new approach to the problem of government shortages that would leave local officials exposed, without the mechanisms of the informal funding network to hide behind.

The Attack On Government Deficits

Arrears and Deficits in the Early Ch'ing

Arrears and deficits were a chronic feature of Ch'ing fiscal administration. An important distinction existed between these two categories of government shortages. Strictly speaking, arrears (min-ch'ien ) referred to shortages in tax receipts, the result of failure of officials to collect the tax quota in full. Deficits (k'uei-k'ung ), on the other hand, applied specifically to discrepancies between the funds on hand in government treasuries or granaries and the amount that should have been there had all funds been disbursed according to government regulations. During different periods, emphasis was placed on alleviating shortages in one or the other of these two categories. In fact, deficits and arrears were inti-

mately related. Real tax evasion naturally contributed to the fiscal insolvency of local officials, and the pressure on officials to cover the costs of government expenditures not provided by the regular funding network led officials and their staff to utilize a number of devices that camouflaged their deficits as arrears. The decision to wage a campaign against deficits or against arrears was often a reflection more of the central government's perceptions of who was more guilty of abuses, the people or the bureaucracy, than of evidence that could be culled from administrative records.

This is not to say that arrears and deficits were not distinct problems or that one or the other was not more central to financial difficulties in different places or at different times. Severe natural disaster always brought on a period of economic hardship for the people in the stricken area, which was usually reflected in tax arrears for a number of years thereafter. However, the traditional fiscal apparatus was fairly well prepared to meet such crises in the form of imperial tax exemptions and deferred payments (liu-ti ). Certain areas, most notably Kiangnan, suffered from endemic arrears, in part as a result of tax resistance perpetrated by the large numbers of literati households with tax-exempt status in the region.^[5] Measures designed to restrict the privileges of degree holders and reductions in the region's excessive tax quota were combined to bring the situation under control for a time.^[6] Moreover, as we have seen, reforms in tax-collection procedures were introduced throughout the empire to reduce opportunities for interference by middlemen and peculation by yamen staff. Despite these efforts, arrears continued to exist, and when they became too great the usual solution was to declare a tax amnesty, wiping the slate clean or relieving the population of current obligations in order to allow them to catch up with debts long due.^[7]

Deficits were a more enigmatic issue in the early Ch'ing. As we have seen, given the level of regular local funding, it was almost impossible for a provincial official to spend any time in office without incurring some deficits. Although the official position of the government was to discourage shortages and punish offenders, the actual record of prosecutions and impeachments points to a good deal of tolerance toward such malfeasance.^[8] Many cases of deficits went unreported, and when missing funds were brought to the attention of the government, it was common practice to use the wages and salaries of all the officials in a province to pay them back.^[9] As a result, some officials became complacent about balancing their accounts. Moreover, the practice of feng-kung contribu-

tions to make up shortages furthered government insolvency because these same funds were needed to offset the expenses that had led to deficits in the first place.

The existence of large deficits in government treasuries during the early Ch'ing did not grow out of a failure to institute statutory controls on official behavior. Stringent regulations existed to ensure the meticulous monitoring of local-government income and expenditure. Under the tsou-hsiao system of annual accounting, officials were required to report all expenditures to the Board of Revenue. Only after approval was given by the Board could an official disburse funds and declare their expenditure (k'ai-hsiao ) in his annual accounts. One could not simply report expenditures at the end of the year. If an official had to appropriate more funds than originally requested, or failed to request permission in advance to allocate funds for a particular purpose, he was acting contrary to the regulations and could be impeached. Whether or not the additional disbursements were legitimate, they were considered deficits and the guilty official was compelled to pay them back.

Laws such as these were designed to forestall embezzlement, the only recognized cause of deficits. Similar precautions were taken when an official handed his office over to his successor (chiao-tai ). When an incumbent official left his post, he compiled an inventory of all income, arrears, and taxes disbursed and actually on hand and turned it over to the incoming official. The latter checked the records and weighed the funds on hand to be sure they agreed with the incumbent's inventory. Only after his successor had issued a bond of guarantee testifying to the accuracy of his chiao-tai accounts was the outgoing official supposed to leave his post for transfer or retirement. If discrepancies existed, the incoming official was to report them to the governor or governor-general, who would impeach the outgoing official. The latter was then held responsible for the deficit and the new official was cleared of any liability.

The operation of the informal funding network had a profound effect on these two methods of fiscal scrutiny, rendering them useless in many cases. Oertai described the situation graphically when he was Kiangsu financial commissioner.^[10]

Kiangsu's wealth is the greatest in the empire. Regarding previous items of deficits, those owed by the people are probably caused by disaster and famine, [but] those

owed by officials are in part caused by avarice. Investigating deeply into the cause, leniency (tan-ku ) is really the source of avarice and following the old habits (yin-hsün ) is the legacy of leniency. Given leniency and adherence to old habits, there are none who will not tend toward avarice. The high officials extort from their subordinates and the subordinates oppress the little people. When the income extorted [from the people] is insufficient to meet expenses, then they disburse treasury funds (k'u-t'ang , i.e., cheng-hsiang tax revenues) . . . The result is deficits. The chou and hsien are not necessarily all without scruples (wu-liang ). In reality, sometimes unscrupulous governors, governors-general, commissioners, and circuit intendants are responsible.

In return for their heavy contribution to provincial expenses, magistrates were often protected against charges of amassing deficits. Only rarely did the high provincial officials impeach lower officials as part of the tsou-hsiao process. The early Yung-cheng investigations of official deficits uncovered hundreds of thousands of taels of deficits that had never been brought to the attention of the central government by the governors of the province at the time they occurred.^[11] Moreover, to protect themselves from implication in these cover-ups, the governors had to subvert the chiao-tai process and force incoming officials to accept the accounts of their outgoing subordinates, deficits and all.^[12] Most new officials were willing to do so because they knew that in all probability they too would incur deficits and would receive similar protection from prosecution.

Officials of the central government also participated in the subversion of the scrutiny system established during the first two reigns of the dynasty. Because Board officials were no more adequately provided with administrative expenses than were their colleagues in the provinces, the Boards participated in the informal funding network by accepting "Board fees" (pu-fei ) whenever they had business with provincial officials.^[13] This practice was particularly damaging in the case of fees that accompanied provincial annual accounts.

Although the accounts are clear and in order, if Board fees are not paid, they will be rejected on the grounds that the figures therein do not match those in the routine memorials [requesting prior Board approval for said expenditures] or on the pretext that the figures

are a few taels off. As soon as the Board fees are paid, millions of taels may be wasted, but the tsou-hsiao accounts will still be approved. Sometimes [the accounts] will be rejected for an unimportant matter to preserve the pretense of rejecting accounts while the truth is covered up. Then when [the official concerned] replies, [the Board] will deliberate to approve his accounts. Officials in the capital and in the provinces conspire to steal and lie.^[14]

Thus, even if an official had deficits his accounts were approved, so long as he paid the required fee to the Board of Revenue. The façade of fiscal scrutiny was maintained while deficits in the provinces grew year after year.

The Target Shifts

The focus of reform during the K'ang-hsi period was on arrears. This was reflected in efforts to improve tax prompting and in the liberal granting of tax remissions. The tendency to view fiscal problems primarily in terms of the failure of the people to pay their taxes in full handicapped officials seeking to rectify basic weaknesses within the regular tax system itself. In fact, if arrears really were the main problem, little could be done beyond exhorting officials to exert greater energy in tax collection or issuing exemptions and remissions. The latter measure, in effect, removed the problem by decreeing it out of existence.

This focus changed dramatically with the Yung-cheng emperor's ascent to the throne. His first policy initiative was to declare a frontal attack on the problem of government deficits and official and clerical peculation. Almost immediately after K'ang-hsi's death, the Yung-cheng emperor issued two edicts laying the blame for deficits not on the people but on the corrupt activities of officials and their clerks. In an edict to the Board of Revenue and Board of Works he first called for stricter auditing of provincial cheng-hsiang expenditures.^[15]

Wealth is the means for accomplishing one's ends. The ancient rulers considered frugality the basis for abundance for the state and the people. Since taking the throne I have constantly worried about embezzlement of government treasury funds by clerks. Hereafter, these two Boards must carefully investigate all taxes in cash and in kind and the costs of materials and labor

in the annual accounts (tsou-hsiao ) and compile and memorialize them in a clear account (ch'ing-ts'e ). They may not report vague estimates. If there are those who report less as more (i.e. in disbursing funds) or cheap as expensive, or whose figures don't balance or who make false estimates [of the costs of labor and materials], when they are discovered, let the Board officials [who approved their falsified accounts] be severely punished along with them.

The emperor recognized that one way in which funds were diverted from the regular tax network to the informal network was by manipulating the accounting of those projects for which regular tax funds were provided. By holding the Board officials equally accountable, Yung-cheng was also making clear his knowledge of the Board's complicity in this process. As a check against collaboration by the Board of Revenue in circumventing the tsou-hsiao system, the emperor established a separate accounting office called the Hui-k'ao fu. At the head of the Hui-k'ao fu, Yung-cheng placed his most trusted officials in the central government—his favorite brother, Prince I; his uncle, Lungkodo; the Grand Secretary Pai Huang; and the Censor Chu Shih. These four officials were commissioned to handle all tsou-hsiao matters, no matter which Board was involved. To assist them in their investigations, a separate yamen was established with a staff consisting of a Manchu and a Chinese assistant department director (lang-chung yuan-wai lang ), three second-class board secretaries (chu-shih ), and ten clerks of the seventh to ninth rank (pi-t'ieh-shih ).^[16] By requiring that all questionable annual accounts be investigated by this independent authority the emperor hoped to halt the practice of evading scrutiny by paying a fee to the Board of Revenue. Moreover, the Hui-k'ao fu undertook to investigate the accounts of the Boards themselves to help clear up deficits on the central-government level. In YC 3 (1725), when the emperor was satisfied that the corruption in local and central finances had been brought under control, the Hui-k'ao fu was disbanded and its duties returned to the jurisdiction of the Board of Revenue.

Yung-cheng's understanding of the vicious cycle of revenue manipulation and concealment was expressed even more vividly in the second of these edicts.^[17]

Since ancient times, regular taxes have been stored up for the expenses of the military and the state. In times

of peace, [we] must cause the granaries and treasuries to be full in order to be prepared and without cause for worry. Recently there have been numerous circuit intendants, prefects, and chou and hsien magistrates in each province with tax deficits. The governors and governors-general are well aware of this abuse, but cover up for them until a time of extreme difficulty, when they usually report embezzlement as "shifting funds for public expenses." When ordered to set a time limit for recovery [of the embezzled funds] they regard them as ancient history and those who pay up in full are few. Repayment is put off for years but the pretense of recovering [deficits] remains. In the end there is no one who can be held responsible (ch'üan-wu-cho-lo ). New officials are forced by their superior officials to accept the handing over of their predecessors' accounts (chiao-tai ). Although [these accounts] contain deficits, [an incoming official's] fear of the power of high officials compels him to accept them. Moreover, in imitation of [his superior's corruption] he will use [the fact that his superior allowed the previous incumbent to amass deficits] to intimidate his superior and will of necessity get [his superior's] protection for wrongdoings so that he may embezzle as he pleases. This in turn leads to even worse deficits. I calculate that this was the cause of several hundreds of thousands of taels of deficits in the Shantung financial commissioner's treasury. How could this all be the result of shifting funds for public expenses (yin-kung no-i )? Now, although [officials in Shantung] are using feng-kung to make up the deficits, in reality they cannot but take [the needed funds] from the people, adding on surcharges beyond the tax quota. If this is the case in Shantung, [the situation] in other provinces can be imagined. When the flesh and blood of the common people is used to rectify [the deficits] of the officials, how can there not be hardship in the countryside? I am deeply concerned about these abuses.

The emperor's first impulse was to combat corruption by instituting a general audit of all provincial treasuries similar to the type used in earlier reigns to ferret out gentry tax evaders. However, his recognition of the extent to which these practices were imbedded in the fabric of local government prompted him to leniency. Rather than create conflicts between the central government and the provinces by sending imperial commissioners to pry into local affairs, he ordered the governors and governors-general

to investigate the deficits of their own subordinates. No one would be punished for having shortages in his yamen treasury or for having failed to detect such shortages in those of his underlings, but all deficits would have to be repaid within three years. Moreover, officials were warned against trying to settle their accounts with the government by extorting from the common people or further falsifying local records.

Thus began the Yung-cheng emperor's crusade against corruption. All government deficits were to be cleared within three years. Thereafter, anyone whose treasury exhibited shortages would be severely punished. This was a policy marked by stern determination, but it promised little originality in the management of the empire's fiscal affairs. What ruler in Chinese history would not have liked to wipe the slate clean, so that moral government under his own leadership could be given a fresh start? The emperor's hard line on official peculation precluded any consideration of the difficulties faced by the bureaucrat in the field. On the contrary, his equally strong commitment to the protection of the people from the excessive demands of the county tax collector magnified the problems of local officials. At least during the first few months of his reign, Yung-cheng continued to view with skepticism claims that missing funds were "shifted for public expenses," and held to the conviction that deficits could have resulted only from the embezzlement of local revenues.

Therefore, it is not surprising that on the inaugural day of his own reign period, Yung-cheng categorically refused permission to use huo-hao to make up deficits in the manner suggested by Nien Keng-yao and others during his father's reign. Instead, he took advantage of this auspicious occasion to deliver a series of traditional exhortations to officials at each level of the bureaucracy to behave virtuously in their assigned duties. In his edict to chou and hsien magistrates he linked the pursuit of moral conduct directly to the issue of deficits.^[18]

As for taxes, they are especially important. The smallest grain of rice represents the blood and sweat of the people. Increase the taxes by one fen [0.01 tael] and you increase the burden on the people by one fen . Decrease them by one fen and the people receive one fen of grace. In the past there were those who requested to add huo-hao temporarily [to the people's taxes] in order to make up deficits in the public treasury. The late emperor issued an edict in court disapproving the re-

quest and all the officials heard it. Now the huo-hao of the chou and hsien is being increased with impunity and is viewed as a fixed precedent. How can the people bear it? Henceforth it must be prohibited. If it is discovered and [the official is] impeached by his superior or impeached by a censor, he must be severely punished and certainly cannot be treated leniently. If one wants to clear up the source of deficits, there is no better [way] than frugality and upright conduct. If one is frugal there will be no lack of funds for expenses. If one is upright in conduct, one cannot be [drawn into misdoings] by one's superior.

It is impossible to determine whether the emperor's stance during these first few months was dictated by the exigencies of filial piety, hesitancy to offend the sensibilities of officials who had risen in the bureaucracy under his father, or a genuine belief that the real source of difficulty lay in official corruption and not in the fiscal system itself. What we do know is that Yung-cheng's commitment to clearing deficits remained firm, but his approach to the problem was soon to change.

Who is Responsible? Who Shall Pay?

The emperor's decision not to penalize officials for deficits incurred prior to the Yung-cheng reign was a shrewd one. Within weeks of his edict, reports of shortages began pouring into the capital as officials attempted to demonstrate their zeal and struggled to rid themselves of the burden of deceit and manipulation that deficits demanded. The missing funds uncovered during the three-year grace period totalled millions of taels. Unraveling the tangle of accounting devices used to disguise fiscal insolvency was an onerous task. However, the real challenge lay in transforming acknowledged shortages and fictitious reserves into tangible stores of silver and grain.

For those deficits that did have to be repaid, primary responsibility fell on the chou and hsien magistrates and on financial commissioners. In cases where these officials did not have the resources to compensate the central government and the governor had failed to report the debts at the time they were incurred, the latter was held jointly responsible for repayment.^[19] High provincial officials were also held responsible when expenditures had been made without prior approval or funds had been shifted from one cate-

gory to pay for expenses in another.^[20] In cases where an official was no longer in the post in which he incurred his deficit, the amount of the shortage was supposed to be retrieved by voluntary or forced sale of the official's personal property, both at his new post and at his ancestral home.^[21]

On paper, such an allocation of responsibility promised the speedy retrieval of all outstanding revenues. As the investigations proceeded, however, it became increasingly obvious that it would require more than an imperial fiat to succeed in eliminating deficits and the corruption that had brought them about. Not only was there no consensus regarding either the causes of deficits or the way to prevent them in the future, but it soon also became apparent that a large number of deficits existed for which it was impossible to assign responsibility.

The one source of deficits widely acknowledged by both provincial and central-government officials was mismanagement of granary stores.^[22] Scattered throughout the provinces, government granaries were, by nature, more difficult to inspect than treasuries and were equally prey to manipulation. It was not uncommon for officials short of silver in their treasuries to sell off grain as a source of additional revenues. In some places, public grain contributions were diverted immediately to other uses and never stored as a safeguard against famine or high grain prices. When they were set aside for disaster relief, it was often in the form of an equivalent sum of silver. Not only did this result in shortages when grain was needed promptly for relief, but the conversion rate at which the silver was stored was frequently below the current market price of grain.^[23] Even the most scrupulous magistrate could find himself without the statutory stocks of grain because loans made to the people in bad times were often never repaid.^[24]

Far more controversial was the issue of treasury deficits. Most central-government officials felt that shortages in provincial- and local-government supplies of silver were the result of embezzlement. Therefore, their contributions to the debate were usually in the form of suggestions for more stringent oversight of provincial financial transactions.^[25] In particular, they decried the shifting of funds for public expenses (yin-kung no-i ). Under this rubric were included all instances in which local officials used funds designated for one purpose to pay for something else, even if those funds were originally intended for remittance to the central government. Implicit in the condemnation of such practices was the belief that

shortages at the local level were the result of official extravagance and not real need.^[26] In no case was it accepted that either inadequate revenue allocations to local government or heavy demands for funds from superiors were responsible for the discrepancies in local accounts.^[27]

The recommendations of central-government officials tended to deflect attention away from the causes to the symptoms of the disease that infected the Chinese fiscal system. They cannot be blamed for ignoring the real problems faced by local officials trying to fulfill their administrative duties. Many central officials had no experience in local government and their present positions excused them from the practical tasks of clearing the deficits that already existed. Even censors, whose role it was to oversee the activities of officials in the field, spent little time in the provinces. However, as reports began filtering in from provincial officials themselves, a very different picture of the problem began to emerge.

A frequent claim was that deficits in treasury stores were actually the result of arrears on the part of the taxpayers. Particularly in Kiangnan, the tax quotas were often too large for a single official to handle. The likelihood that an official assigned to such a post would be dismissed for failure to collect his taxes in full was great. When a new official came to replace him, he would not only be responsible for collecting the taxes for that year, but would also have to dun the arrears accumulated under his predecessor. In order to avoid removal himself, the new official would have to shift regular taxes (cheng-hsiang ) designated for other purposes to cover the taxes he was unable to collect.^[28] Eventually the shortage would be discovered, but by that time the manipulation of tax records would have hidden all evidence that the deficit was once "people's arrears." Even in remote and tax-poor provinces such as Kwangsi, tax resistance and embezzlement by tax farmers could force magistrates to take such measures to prevent impeachment.^[29] In fact, so common was this problem that in some provinces Prince I and the Board of Revenue ordered a special investigation to determine whether any of the deficits attributed to official peculation could be traced to nonpayment by the taxpayers.^[30]

The central government was less sympathetic toward officials who reported that their deficits had resulted from legitimate expenditure. Scores of officials memorialized having allocated funds for necessary purchases not provided for under the regular revenue-sharing system. In most cases they were in support of the military,

maintenance of public security, and construction of public works. All of these areas were of vital concern to the central government, as well as the local jurisdiction. Nevertheless, once the deficits were discovered, the funds had to be reimbursed, either by the official who had authorized the expenditure or by remanding the wages and salaries of all the low-level officials and functionaries in the province.^[31]

Although the responsibility for the repayment of deficits was supposed to fall on the official who had incurred them, it was not always possible to determine who that person was. This was particularly true when shortages appeared in the financial commissioner's treasury. The manipulation of accounts that was the product of the informal funding system had its greatest impact at the provincial level. With well-placed gifts, a lower official could avoid impeachment for failure to fulfill his tax quota. At the same time, governors and governors-general frequently borrowed funds which the financial commissioner would be forced to cover up. Since even expenditures for public purposes were illegal if not authorized by the central government, high provincial officials had a powerful incentive to make the accounts of the financial commissioner's treasury as difficult as possible to understand.^[32] This became vividly clear when the Board of Punishments attempted to clear up a case of deficits in the Hupei provincial treasury.^[33]

From the beginning, the Board's interrogation of high-level Hupei officials did not promise a speedy solution to the disappearance of funds placed in their care. After much cajoling, the former governor-general, Man-pi, admitted that 100,000 taels had been missing from the treasury during his term of office. This, he claimed, had been repaid out of his own income from huo-hao and customary fees and that of the governor and financial commissioner. However, when questioned on this subject, the present financial commissioner, Chang Sheng-pi, reported that the funds had not been made up at all. Instead, the shortage had been camouflaged by Hsü Ta-ting, who had been appointed to replace the financial commissioner in whose term the deficit had arisen. In order to hide the fact that 100,000 taels could not be accounted for, Hsü had ostensibly fabricated allocations of 60,000 taels to make up a shortage in the funds set aside to purchase copper and 40,000 taels of feng-kung yin to purchase horses for the military. Chang Sheng-pi had then been forced to go along with this deception when he took office.

If the investigators thought they now had the truth, they were once again thrown into confusion by the testimony of the original financial commissioner, Chang Wen-rs'an. According to him, the money involved was not 100,000 taels, but 150,000 taels. Moreover, it was not a deficit, but the result of people's arrears. Nevertheless, as the official in charge of fulfilling the province's tax quota, Chang Wen-ts'an had personally undertaken to make up the funds, some of which dated back to the period before he had taken office. With money obtained from the sale of considerable family property, several loans, remittances from servants and treasury officers implicated in the case, and a grant of 20,000 taels of huo-hao from the governor and governor-general, Chang claimed to have been able to replace the entire amount in one year.

Who was really to blame? Certainly, after so much tampering, the records of the Hupei provincial treasury would yield few clues. Moreover, more than 100,000 taels were still missing. Presumably, these funds would end up like so many others for which responsibility was indeterminate. Such deficits were usually classified as wu-cho chih-k'uan . Their very existence was a clear indication of the laxity with which shortages in government treasuries had been handled under previous administrations. Time and again, officials trying to clear up the accounts in local and provincial yamen bemoaned the passage of years and often decades during which deficits had been allowed to grow unchecked and unreported. In some cases the official originally responsible for the shortages was long dead.^[34] Even if he was still alive, his personal and family property was often insufficient to make up deficits that could amount to thousands or tens of thousands of taels.^[35] The web of deception that had grown up around the system of turning over accounts created many cases of disputed responsibility. The position taken by officials in the field was that the original official should make good any shortages that arose during his term of office, even if he had already left government service. However, fearing delays in repayment and complications resulting from attempts to retrieve funds from outside the province in question, Prince I and the Board of Revenue held to a strict interpretation of the dynastic statutes. Once an official had issued a bond of guarantee stating that all revenues placed in his charge were present and accounted for, it was his responsibility to rectify any discrepancies that might later be uncovered. If an incoming official had signed such a bond knowing that his predecessor's accounts were not in order, it could

only have been in hope of some future illicit gain.^[36] In effect, this ruling made officials now holding a post exclusively responsible for the repayment of the deficits of all their predecessors who had not been impeached, except in cases where evidence existed that they had filed a report and had been coerced into accepting the chiao-tai of the outgoing incumbent.

Needless to say, the local officials at the time of the deficit investigations balked at having such a huge financial burden placed on their shoulders.^[37] However, their resistance to paying for the crimes of others constituted only one of the obstacles confronting officials trying to replenish provincial silver stores. Retrieval of missing funds by confiscation of the guilty party's family property was a slow process and involved one province in what amounted to litigation in another province. Officials in the guilty official's horne province had their own deficits to worry about and were usually unenthusiastic about engaging in the tedious process of uncovering a person's assets and selling them, especially when the proceeds would be sent elsewhere. Moreover, when the guilty party was an official who held higher office than the magistrate or prefect in charge of his case, it was often extremely difficult for the latter to treat him with the severity that was usually necessary.^[38] When responsibility for deficits that had not been reported, or which had arisen out of unauthorized expenditures at the provincial level, devolved upon one or two high officials, the sums were often so great that there was no realistic expectation that they could be repaid swiftly. Finally, in the case of wu-cho deficits, responsibility for repayment could not even be assigned.

Under these circumstances, many provincial officials turned once again to the system of informal funding. On the level of the chou and hsien, magistrates often forced the people to pay surtaxes to make up their deficits.^[39] Some higher officials continued to rely on contributions of wages and salaries from their subordinates (chüan-feng ).^[40] Because feng-kung was still a major source of funding for provincial expenses, and deficits usually far exceeded the funds these wages and salaries made available, most provinces also relied on contributions of officials' customary gifts and fees to make up shortages in the provincial treasury.^[41]

It was at this point that the Yung-cheng emperor dealt the decisive blow to local fiscal administration as it had functioned for several years. In the process of investigating deficits, the emperor became increasingly convinced that the demands placed on lower

officials by their superiors were one of the major factors leading to shortages in cheng-hsiang funds. Therefore, in an edict to all governors and governors-general, he issued a ban on contributions of wages and salaries. At the same time, he issued an edict ordering that steps be taken to eliminate the sending and receiving of lou-kuei .^[42] It might have been possible for officials to ignore the emperor's instructions and secretly continue to rely on these two sources of funds in daily transactions to cover expenses for which cheng-hsiang revenues had not been provided. However, when repayment of deficits was being handled directly by the Board of Revenue, it was impossible to camouflage the source of such huge sums. The hands of provincial officials were tied. A new solution, both to the problem of repayment of deficits and to that of preventing similar shortages in the future, had to be found.

The Reform of Local Fiscal Administration: Huo-Hao Kuei-Kung

The crisis generated by the emperor's ban on feng-kung and lou-kuei contributions once again focused official attention on the richest of all informal sources of revenue. References to huo-hao , or meltage fees, appear as early as the Sung dynasty.^[43] However, their importance as a form of tax surcharge is associated with the increasing monetization of the Chinese economy in the late Ming. Technically, the meltage fee was a charge added to regular tax remittances to compensate for the inevitable loss of silver that resulted when taxes were melted down into large ingots for transporting to the central government. Loss of silver also occurred because of the difference between the quality of silver received from taxpayers and the standard fineness required of taxes sent to the capital. Although the actual loss was rarely more than 1 or 2 percent, the proportion of meltage fees levied on land and head taxes increased steadily from the late Ming until the eighteenth century.^[44] At the same time, the term huo-hao came to describe almost any general percentage surcharge for which no other special term existed.

The amorphous nature of meltage fees and their diminished connection with actual silver depreciation led to their immediate ban by the new Manchu rulers of China.^[45] This proscription was continued under K'ang-hsi, who was not incorrect in equating huo-hao with the unauthorized appropriation of the people's re-

sources through the informal funding system.^[46] Under the K'ang-hsi emperor, laws were promulgated permitting commoners to report officials who collected taxes beyond the quota and specifying punishments for superior officials who covered up for subordinates who exacted huo-hao .^[47] Nevertheless, the magistrates' need for funds for daily living expenses and the wages of servants, clerks, and runners compelled officials to recognize the need to collect at least some extra revenue in the form of huo-hao . As we have seen, even K'ang-hsi had to concede that an official who demanded only 10 percent above the tax quota could still be recommended as a good administrator.^[48]Huo-hao therefore came to occupy an ambiguous position as both an illegal surcharge and a source of funds increasingly viewed as the private property of the chou and hsien bureaucracy. Notwithstanding frequent exhortations to high provincial officials urging them to supervise their subordinates and eliminate abuses in tax collection, huo-hao continued to be collected and rates continued to rise.

Once alternative means to eliminate deficits were obstructed by the emperor's ban on lou-kuei and salary contributions, several characteristics of huo-hao made it an obvious choice. As an openly declared percentage of regular taxes, huo-hao escaped the stigma of more deviously contrived methods of milking the common people through falsification of records, tampering with weights and measures, and extortion. Moreover, because the collection of huo-hao could be integrated into the system of "wrapping and depositing one's own taxes," officials would not be forced to rely on the complicity of clerks and runners to defraud the taxpayer. Integration of huo-hao into the legal fiscal administration would therefore not undermine the taxpayer's direct dealings with the state, a principle central to Ch'ing administrative ideology. The collection of huo-hao , although generally viewed as a form of peculation, had always been tied to real levels of local revenue and government fiscal requirements. Areas with large gross tax receipts customarily levied huo-hao at considerably lower rates than areas with very small tax receipts. This put a greater burden on the taxpayers in the agriculturally depressed parts of the empire, but guaranteed all magistrates comparable funds for at least a portion of their administrative expenses. Perhaps even more important to officials at the time was the fact that as a strictly local levy, huo-hao was not associated with bribery. Although revenues from huo-hao contributed to the pool of funds sent to higher officials in the guise of lou-kuei ,

it was not linked directly to the concealment of official malfeasance that critics of local fiscal administration saw as the main cause of deficits.

It was probably such considerations that had prompted officials such as Nien Keng-yao to recommend the legalization of huo-hao to clear deficits in the past. However, if the return of the meltage fee to the public coffers (huo-hao kuei-kung ) had been simply a program to eradicate government debts, it would deserve little more than a footnote in the annals of Chinese fiscal administration. The great achievement of officials seeking a solution to the immediate issue of debt repayment was their attempt, for the first time in late-imperial history, to attack the causes of deficits. Their efforts gave rise to a complete reorganization of local-government finance, in which huo-hao would play the pivotal role.

From Desperation to Innovation

Government toleration of salary contributions and lou-kuei had been founded on a monumental fallacy. Because these funds appeared to come out of the officials' own pockets, they were rarely associated with increased taxation imposed on the common people. On the other hand, huo-hao , the source of much of the lou-kuei that circulated within the bureaucracy, was nothing more than an illegal surcharge, a deplorable evil. The Yung-cheng emperor's attack on salary contributions and lou-kuei was aimed at improving state finances by rehabilitating official discipline and morale. How was it, then, that after decades of resisting such a step, the Ch'ing dynasty legalized the collection of huo-hao and transformed it into the basis for the rationalization of government funding at the local level? Huo-hao kuei-kung violated the supreme taboo, that against high taxes. How was the emperor able to implement a policy which to many Chinese officials appeared to sacrifice the people's livelihood for the sake of state finances?

The manner in which huo-hao kuei-kung became public policy bears some resemblance to the way in which the Single-Whip reforms were implemented in the late Ming. Initiative appeared to be local, transmitted to the central government by the governors and financial commissioners of individual provinces. No general order or legislation was enacted requiring empire-wide implementation, and when it was carried out, each province took into consideration local differences in establishing rates and methods of collection

and remittance. However, beneath the surface a crucial difference existed in the mediating and guiding role played by the emperor during the Ch'ing reform. In part, this was the result of the emperor's dynamic personality and personal commitment to fiscal reform, which we have already seen demonstrated in his attack on deficits. Closely related to these factors was the institution of a new communication device, the secret palace-memorial system, in this same period.^[49]

One persistent theme in late imperial Chinese history was the devolution of imperial control over the bureaucracy and the decision-making process. In the routine memorial system, or t'i-pen chih-tu , the correspondence of local officials passed through the offices of the Grand Secretariat, which not only selected those memorials that were deserving of the emperor's attention but drafted possible responses from which the emperor could choose his rescript. At times, this system enabled the central bureaucracy to manipulate policy by screening the information available to the emperor. It also produced a situation in which weak emperors could evade all responsibility for government decisions. In a political system where decision-making authority was ambiguous, this greatly limited the possibility for innovation, particularly that emanating from the provincial level. Moreover, it obstructed rationalizing reforms for which a major prerequisite is the most complete possible information about conditions as they really are.

Beginning in the last years of the K'ang-hsi reign, a second memorial system was instituted. Its full development was realized under the Yung-cheng emperor. This secret palace-memorial system, or tsou-che chih-tu , made possible unmediated communication between the emperor and his officials in the field. The existing central-government organs were completely by-passed and correspondence by this route was guaranteed total confidentiality. Because all high officials knew that all other high officials had the same privilege of secret communication with the emperor, the latter was assured of honest reporting from his subordinates. Their reports were for the emperor alone, hence officials could express their opinions and present their ideas freely, without fear of censure from their peers in the central government or of accusation that they were spurning widely held values. Moreover, the emperor, who composed most of his rescripts himself, was in a position to experiment with new ideas without the formal sanction that adhered to any imperial statement made in response to a t'i-

pen .^[50] Through private communications of this kind the emperor could also help mold official opinions and actions by means of subtle suggestions and criticisms while preserving provincial autonomy in areas where he felt it important. Thus, through the secret memorial system, both imperial and provincial decision-making power were greatly enhanced. One of the most notable examples of this may be found in the implementation of the huo-hao kuei-kung reforms.

Among many high provincial officials the prospect that their main avenues of funding would be cut off inspired panic rather than creativity. Some governors and financial commissioners even nurtured the hope that a persuasive argument could win exemption from the ban applied elsewhere. Kiangsi Governor P'ei Shuai-tu boasted that he had eliminated all customary gifts previously sent to his own yamen and that of the commissioners, circuit intendents, and prefects. However, his virtuous display of compliance with the emperor's wishes did not prevent him from receiving 12,000 taels in fees from the grain intendant and the financial commissioner. P'ei realized that his role as a Confucian administrator was to set a good example for those beneath him. To accept such fees would encourage his subordinates to do the same. Yet without some informal source of funding, neither he nor the financial commissioner would be able to operate their yamen.^[51]

Whereas P'ei sought imperial sanction for his continued acceptance of lou-kuei , other officials attempted to circumvent the prohibition of contributions of feng-kung . Liang-kuang Governorgeneral K'ung Yü-hsün feared that without contributions of feng-kung the provincial government would be forced to farm out purchases of materials and supplies to the individual chou and hsien. This could only result in the latter making up the costs by levying surcharges on the people. Interpreting the emperor's ban as a demonstration of concern for poorly paid officers and staff on the lowest rungs of the official hierarchy, K'ung proposed that Kwangtung continue to appropriate official salaries, but only those of officials of magisterial rank and above.^[52] Along the same lines, Chekiang Governor Huang Shu-lin proposed solving the problem of wu-cho deficits in his province by using contributions of feng-kung collected during the period before the imperial ban was issued.^[53]

In each of the above cases the emperor's rescript was noncommittal. On the one hand, he reiterated his concern that the collec-

tion of fees and salary contributions be terminated. On the other, he promised that the pleas of officials who really found themselves in financial straits would not go unheeded. Although only a few were aware of it at the time, Yung-cheng himself was becoming convinced that no provincial official could survive on the revenues legally made available to him by the state. Communicating with them through secret palace memorials, he allowed a small number of officials to experiment with the use of huo-hao to make up deficits and provide a more stable source of income for provincial administration.

The earliest evidence of a plan to use huo-hao for public expenses appears in a memorial from Hukwang Governor-general Yang Tsung-jen. Even before Yung-cheng issued his edict prohibiting the contribution of feng-kung , Yang expressed his fear that if lower officials were deprived of their salaries they would have no choice but to extort funds from the people. However, if no provision was made for provincial expenses, the costs of government projects would inevitably fall on the magistrates, who in turn would pass them on to the taxpayer. In order to solve this problem, Yang proposed that the chou and hsien be allowed to collect huo-hao at a rate of 10 percent. Of this, 20 percent would be sent to the financial commissioner for provincial expenses, 10 percent would be sent to the Board of Revenue to cover losses in the regular taxes remitted (yü-p'ing ), and 70 percent would be retained by the chou and hsien to meet their own administrative costs. The emperor's response to Yang's plan was extremely enthusiastic. He called it "completely correct and without a flaw," but warned Yang to be cautious in carrying it out.^[54]

At the same time that Yang Tsung-jen was implementing his plan in Hunan and Hupei, Shansi Governor No-min was formulating the scheme that would eventually become the model for the whole empire. No-min's original memorials have not survived. However, Shansi Financial Commissioner Kao Ch'eng-ling reported two years later that in a memorial dated YC 1, 5, 13, Nomin requested that the rate of collection for huo-hao in Shansi be standardized at 20 percent. Inasmuch as nowhere in the province was the rate lower than 20 percent, this meant an immediate saving for many of Shansi's taxpayers. The governor then estimated that the total huo-hao for the province would be approximately 500,000 taels. Of this, 300,000 taels were to be set aside for officials' expenses and for the public expenses of the whole prov-

ince. The remaining 200,000 taels would be deducted annually to make up wu-cho deficits.^[55]

Although they receive no mention in the official accounts of the huo-hao kuei-kung reforms, other provinces were also initiating programs utilizing this new method of funding. After receiving an ambiguous response to his own proposal to continue the use of salary contributions for public expenses, Chihli Governor Li Weichün memorialized the emperor, advocating limited use of huo-hao for this purpose.^[56] Shantung Governor Huang Ping began to investigate the collection of huo-hao in his province as early as the end of YC 1 (1723).^[57] His efforts to control rates of collection as a prelude to their use in deficit repayment was reported to the emperor early the next year.^[58] However, an elaborate scheme for sharing huo-hao on a provincial basis was already being attempted at this time. According to Wu Lung-yuan, who passed through Shantung's Yen-chou prefecture on his way to proctor the examinations in Kiangnan, the rate of huo-hao collection in Shantung had been fixed at 15 percent of the regular tax. For every tael of tax collected in Yen-chou an additional 0.03 tael was put aside for transportation expenses, 0.01 tael for the prefect, 0.007 tael for the Yen-ning circuit intendant, and 0.005 tael for the judicial commissioner. The remaining 0.098 tael (65.3 percent of the total huo-hao ) was being used to make up deficits. Once all the deficits were paid in full, the chou and hsien would be permitted to deduct an additional 0.03 tael for themselves, the remainder being allocated for public expenses.^[59] Surviving documentation makes it impossible to determine whether the situation observed by Wu Lung-yuan was typical of all administrative units in the province. Nevertheless, it is clear that some form of huo-hao kuei-kung was being practiced quite early in Shantung.

The same may be said of neighboring Honan. Toward the end of YC 1, Honan Governor Shih Wen-cho toyed with the idea of using the huo-hao of the whole province to make up wu-cho deficits. The bans on lou-kuei and salary contributions were not yet in effect, but Honan's feng-kung was already completely committed to military-supply costs and dike repairs. However, Shih still shared the common view that huo-hao was the exclusive property of the chou and hsien magistrates, and at this point felt it was unfair to require magistrates without deficits to contribute to the repayment of deficits for those who did have them.^[60] Instead, he decided to continue the use of lou-kuei to make up these shortages.

By the new year, lou-kuei was no longer a permissible remedy for government shortages and Shih proceeded to present the emperor with a detailed proposal for the use of huo-hao to clear deficits and furnish local officials with funds for their expenses. Governor Shih's plan made no attempt to alter the existing rates of huo-hao collection, rates which at the time varied between those paid by the gentry and those paid by commoners. He estimated that the average rate was about 13 percent. Given Honan's annual land- and head-tax quota of approximately 3,600,000 taels, this would provide the province with around 400,000 taels of huo-hao each year. From this total, each official would be granted personal expenses, as well as public expenses, and funds would be set aside for advance payments (p'ei-chih , probably to cover unexpected costs that would be paid back from the cheng-hsiang after approval was received from the Board of Revenue). Once the above were deducted, Shih calculated that between 150,000 and 260,000 taels would be left over to make up deficits and pay for provincial-level expenses.^[61]

The emperor's reply to Shih Wen-cho's memorial is very revealing. Unlike the noncommittal responses he made to officials who tried to bend the rules and continue the use of funds from the informal network, Yung-cheng's rescript openly praised Shih's plan. The emperor told him that this memorial could not be compared with the vague mutterings in his previous memorials. Yung-cheng pointed out that a high provincial official's duty lay precisely in this kind of thorough attention to fiscal matters. Only by calculating total available funds and determining how much was needed to make up deficits and to pay for officials' personal and public expenses could the smooth functioning of government be assured. Platitudes about morality and good government were not enough. "If one speaks in imprecise and confused generalities, uttering phrases such as 'taxes are important,' then what one says amounts to empty words on paper. What practical [result] does it have?"^[62]

What differentiated Shih's and others' discussions of huo-hao reform from previous proposals to clean up government was their focus on rectifying the dysfunctional aspects of the regular system of revenue sharing. In every case, huo-hao was recommended not only as a temporary means to repay deficits, but as a long-term source of official living and administrative expenses. The idea that virtue in administration was the product of factors other than the moral qualities of the men in office was not a new one. At least as

early as the late K'ang-hsi reign, memorials on good government referred to the need to nourish virtue (yang-lien ) by assuring officials adequate income. As we have seen, however, most officials in the central government persisted in the belief that the Ch'ing dynasty had already achieved this ideal through existing fiscal arrangements. It was not until the Yung-cheng period that the term yang-lien was applied to funds designated specifically to provide provincial officials with a substantial and autonomous source of revenue.

In the middle of the first year of the Yung-cheng reign, Shih Ichih wrote a memorial linking deficits to the insatiable demand for funds imposed by provincial governors and governors-general on their subordinates. These high officials made a great show at being virtuous (chiao-lien ), but even when they claimed to have eliminated lou-kuei and huo-hao in their jurisdictions, the amount they continued to extort was great. Shih's observations did not come as a revelation to the emperor, but his remedy reflected the sentiments of a growing number of practical reformers in the field. Shih felt that the reason corruption continued unabated was that the expenses of officials exceeded their share of legal tax revenues. No individual could possess the personal resources necessary to run a whole province. Rather than warning against greed, Shih proposed that the best way to eliminate the pretense of virtue and make it a reality was to liberalize these high officials' sources of yang-lien . Therefore, he requested that the emperor issue an edict to all the governors and governors-general in the empire to calculate the amount of huo-hao, chieh-li , salt, post, customs, and tribute administration fees for their provinces and present a list of them to the emperor. The emperor would then calculate the amounts and items that could be retained on the basis of the size and remoteness of the province and the harm and benefit pertaining to the collection of each kind of fee. Those funds approved by the emperor would be a source of yang-lien for the governor and governor-general, and all other informal sources of income would be eliminated.^[63]

The emperor did not issue such an edict at this time, but the concept of a regular source of yang-lien was integrated into most of the plans for huo-hao kuei-kung presented to the emperor. The term was used by Shih Wen-cho, Yang Tsung-jen, and No-min in their memorials to refer to the funds they proposed to set aside from huo-hao for provincial officials' expenses. Their initial pro-

posals were rather vague in this regard and it remained for specific procedures to be established for determining who would receive yang-lien , what amounts would be given each official, and how the new funds would be distributed.

The catalyst for this final stage of fiscal rationalization was introduced in Shansi late in YC 1. Governor No-min and Financial Commissioner T'ien Wen-ching decided to require that the chou and hsien magistrates remit their huo-hao in its entirety to the province's financial commissioner. The latter would then calculate the amount of public-expense funds and yang-lien to be issued to each official yamen. In all the previously memorialized plans the magistrates were allowed to deduct their own expenses from the huo-hao collected, remitting only that amount that was allocated for higher levels of government and to repay deficits. As we shall see, it was this innovation that really distinguished huo-hao kuei-kung from plans that merely coopted previously illegal levies into the regular network for the purpose of paying up deficits.

Reactions to Reform

Criticism of the new fiscal practices began to appear during the early part of YC 2. Experiments with huo-hao kuei-kung not having been publicized through official channels, most of those who commented on the reforms at this time were central-government officials who had been sent to proctor examinations in provinces where the reforms were being tested.

One of the main targets of these early critics was the newly proposed requirement that chou and hsien magistrates remit all of their quota of huo-hao to the provincial treasury. Censor Wang Chih-lin was not averse to the legalization of huo-hao or its use to provide provincial officials with yang-lien . He did object to the waste in resources that transportation to the provincial capital and back to the local administrative units would entail.^[64] Others were not so sanguine. Lo Ch'i-ch'ang, who witnessed the reforms in action while proctoring the examinations in Honan, expressed unqualified opposition to the remittance of huo-hao . During his travels, Lo had heard that in Chihli, Shantung, Shansi, Shensi, and Honan the governors and financial commissioners had demanded that all huo-hao collected on the land and head tax be deposited in the provincial treasury in preparation for expenses incurred in carrying out public business. In return for their sacrifice the chou

and hsien magistrates were given yang-lien . The amount allocated varied with the size of the jurisdiction, ranging from 600 to 1,000 taels per year. Although this may have solved the fiscal problems of high government officials, Lo reported that the magistrates were now complaining about inadequate funding from yang-lien .^[65]

Criticism was also leveled against the reforms by officials who felt that the issue of reform could not be divorced from Confucian principles of rule. A number of officials feared that legalization of huo-hao would result in higher taxes. If huo-hao were drawn upon to make up deficits, magistrates would use this as a pretext to increase the rates at which this surcharge was collected. One such critic invoked the ancient precept that stressed the rule of men over the rule of law (yu chih-jen, wu chih-fa ). If the emperor truly wished to solve the problem of deficits, he would be best advised to abandon these administrative schemes and concentrate on choosing virtuous men for the posts of governor and governorgeneral.^[66]

Probably the most famous opponent of huo-hao kuei-kung was Shen Chin-ssu. Best known as a philosopher and devoted follower of the Chu Hsi school, Shen's critique of the reforms also reflected a concern for the preservation of Confucian moral values. Unlike many other central-government officials, Shen had had a distinguished career as a local magistrate, a fact that added credence to his evaluation of the new policies. In 1723 the new emperor appointed Shen to the post of director in the Board of Civil Appointments, and in 1724 Shen was sent to proctor the examinations in Shantung.^[67] Shen also feared that the legalization of huo-hao would encourage official avarice and lead to heavier tax burdens on the people. However, his main concern at this time was the effect that huo-hao kuei-kung as a method of making up deficits would have on official morale.

In his memorial, Shen pointed out that not all deficits were the same. He felt that the governors and financial commissioners in each province should be ordered to undertake a general audit of all deficits. The deficits should then be divided into four categories: (1) those that were the result of an edict ordering officials to make public contributions (kung-chüan ) for construction or other projects; (2) those that were the result of funds being used on the official's own initiative for projects of benefit to the whole province; (3) those in which one official was taking the responsibility for deficits left by a former official; and (4) those that were the result

of embezzlement or shifting of funds by the incumbent official himself. Shen insisted that in cases where funds were embezzled by the incumbent himself, repayment should be made by that official and should be accompanied by severe punishment. In cases where an edict was issued to contribute funds the officials should be exempted entirely from repayment. He agreed that where an official was being held responsible for funds used for province-wide expenses or for deficits left by his predecessor, it was not fair to make him use his own wealth to make up the shortage. Nevertheless, Shen persisted in the belief that huo-hao was a strictly local source of funds. To use huo-hao to make up any but the official's own deficits would still be equivalent to confiscating his personal resources.

According to Shen, the desire of the reformers to restock their treasuries had led them to overlook the importance of moral factors in governing the empire. Even if an official had good cause to have accumulated deficits, the very fact that his accounts were not clean was a blot on his integrity and a sign that he did not know how to economize. Shen was convinced that, if such an official were left in office, not only would he fail to repay his previous deficits, but he would surely run up new ones in the future. The result could only bring more harm to the people and increase levies of huo-hao . Therefore, he recommended that any official with deficits be removed from office and another official be appointed in his place. The latter could then record the exact size of the deficit in the jurisdiction and calculate how much should be paid back each year from collection of huo-hao . Moreover, huo-hao rates would be fixed at no more than to percent of the regular tax quota, and any official exceeding this amount would be impeached for avarice (t'an-wu ) and severely punished.^[68]

The implication of Shen Chin-ssu's remarks was that, although the policy was not ideal, it was permissible to legalize the collection of a limited amount of huo-hao for the express purpose of repaying deficits. This was because the harm caused by deficits to official morality was greater than the potential damage caused by legitimizing huo-hao . However, collection could be sanctioned for only a specified time period and would be controlled by having a yearly quota calculated and remitted for deficit repayment. Given his conclusion that, even in extenuating circumstances, deficits were a sign of moral depravity, it is not surprising that Shen made no mention of using huo-hao to rationalize provincial expenditure.

Not all the reactions to the new policies were negative. Some

central-government officials were impressed with the reforms and urged the emperor to make improvements in their implementation. One such proposal came from Board of Punishments Vice-president Chang Ch'ing-tu, who was sent to proctor the examinations in Hunan. As we have already seen, in that province magistrates collected a uniform percentage of huo-hao , deducting their own share at the source. Chiang suggested that the chou and hsien continue to collect huo-hao at a rate of 10 percent, but that they be ordered to remit it in full along with their regular taxes. The governor and governor-general could then calculate allocation on the basis of each administrative unit's location and the complexity of its government business. In this way an official's yanglien would reflect his needs and not just the size of the tax quota in his area.^[69] Chang probably did not know that his was precisely the method being introduced in Shansi and Honan, a method others did not find so appealing.

Backlash in Honan

Much of the initial reaction against huo-hao kuei-kung was prompted by a genuine concern that Confucian principles of benevolent rule not be violated. However, concern for personal gain also motivated the reluctance of many officials to abandon the informal funding system. It is no accident that the magistrate's right to control huo-hao became the rallying cry of huo-hao kuei-kung 's most fervent critics. As long as huo-hao remained illegal and its collection unregulated, it could provide the magistrate and his superiors with a lucrative source of personal income. Even among those for whom profit was not an issue, the idea of supervised collection and allocation of meltage fees raised the spectre of new shortages at the local level. Only several years of successful implementation could dispel this fear. Consequently, the physical remittance of huo-hao to the provincial coffers became the main target of resistance to the reforms in the provinces where they were actually being implemented. By the summer of 1724, opposition to the theory of huo-hao kuei-kung was translated into opposition to the practice of huo-hao kuei-kung .

In Shansi this was manifested in the refusal of a number of magistrates to remit their huo-hao to the provincial treasury.^[70] Rapid impeachment of the recalcitrant minority soon brought the situation under control, and as a result, the reforms received little ad-

verse publicity. In Honan, magistrates also sought to forestall the remittance of huo-hao , but here they did so by joining forces with officials in the capital to discredit one of the earliest proponents of the reforms, the newly appointed Honan financial commissioner, T'ien Wen-ching. Requests for T'ien's impeachment and the simultaneous outbreak of gentry resistance to another of T'ien's policies focused the attention of officialdom on huo-hao kuei-kung and prompted the emperor's first public intervention on behalf of the reforms.

T'ien typified the kind of energetic, practical administrators who were selected for high provincial posts by the Yung-cheng emperor. Like the other officials instrumental in Yung-cheng's campaign to improve local government, T'ien had entered the bureaucracy by the irregular route.^[71] His rise to prominence was a direct result of imperial patronage.^[72] T'ien Wen-ching began his career as an assistant magistrate in 1684 and in twenty years of government service rose no higher than the rank of magistrate of an independent chou. In 1708 he was transferred to Peking and held several middle-level positions. During the last years of the K'ang-hsi reign, however, T'ien seems to have developed a close relationship with the prince who was soon to become the new emperor,^[73] and throughout his reign Yung-cheng referred to T'ien as one of his three most trusted officials.^[74]

T'ien's relationship with the Yung-cheng emperor was based in part on a common approach to administrative affairs. In their public lives, both men demonstrated a deep commitment to the aim of improving the institutions through which virtue could be revealed in government. Though reviled in his own day, T'ien was one of the great exemplars of ching-shih activism in office, and many of his memorials were republished in the nineteenth century. The minister and the ruler also shared an intense distrust of scholar-officials and the cliques that were often formed among successful examination candidates.^[75] This may explain their readiness to blame many of the problems inherent in Ch'ing fiscal administrations on favoritism and cover-ups among provincial officials. T'ien's long tenure in local government led him to place value on practical experience and not on the effete book learning of Confucian degree holders. He consistently emphasized administrative expertise in his evaluations of his subordinates and may have been responsible for the introduction of a system of official apprenticeship (shih-yung kuan ) adopted during the Yung-cheng period.^[76]

Moreover, T'ien's own writings on administrative matters were published by the government as part of its efforts to train local officials.^[77]

Although we do not know whether T'ien Wen-cbing was promoted to high provincial office with the specific purpose of advancing the cause of huo-hao reform, he was clearly an ideal vehicle.^[78] T'ien was sixty-two years old when he was rescued from obscurity and sent to Shansi as financial commissioner. As he himself often stated, he was too old to let career concerns influence his actions in office. As a Chinese bannerman without local ties or literati interests, his loyalty was to the emperor alone. T'ien's intimacy with Yung-cheng and his confidence that he was carrying out the emperor's desires contributed to his single-minded pursuit of reform.^[79] Nevertheless, his lack of tact made him one of the least popular of provincial officials. It is not surprising, therefore, that early resistance to the new policies came to focus on the person of T'ien Wen-ching.

Once the concept of the return of the meltage fee to the public coffers was established in Shansi, T'ien was transferred to the post of financial commissioner in Honan. Just as his arrival in Shansi marked the beginning of a systematic use of huo-hao to finance local government, T'ien's move to Honan was immediately reflected in changes in fiscal administration there. As we have seen, it was early in YC 2 that Governor Shih Wen-cho decided to use huo-hao to make up deficits and provide officials with a stable source of operating funds. With T'ien as his second in command, this was accomplished by means of full remittance of huo-hao to the provincial treasury.

Lo Ch'i-ch'ang's criticism of this method of local funding was soon echoed in a barrage of memorials calling for T'ien's removal from office. According to his impeachers, under T'ien's leadership "not a 'cent' is given to the chou and hsien." As a result, "the officials are poor and the people are experiencing hardship."^[80]

Shih Wen-cho responded to these accusations by pointing out that in return for sending their huo-hao to the provincial capital, all officials in Honan now received yang-lien as well as additional funds for public expenses. Because the magistrates were no longer called upon to provide their superiors with gifts and fees or submit to unscheduled calls for contributions to public expenses, their financial situation was now much better than before.^[81] T'ien's own reply was even more emphatic in its insistence on the importance

of full remittance of huo-hao . If huo-hao was deposited in locked chests along with the regular tax and sent in full to the financial commissioner, the magistrate would have no incentive to increase the demands he made on the people. Only in this way could high provincial officials ensure a steady flow of funds for all provincial expenses and halt the escalation of surcharges imposed on the taxpayer.^[82]

Yung-cheng had no personal doubts as to T'ien's integrity, but his reply to T'ien's memorial clearly reflects the experimental nature of the reforms at this point and the caution with which officials had to proceed in carrying them out.^[83]

Now that I see your explanation, although it is clear, you and the Governor seem to have managed this inappropriately, arousing suspicion and hate. Even more incredible, there are those who say this action of yours is worse than accepting bribes. Your hearts do not harbor such things, but in your actions there is perhaps a little to lead people to suspect that you do. Be careful to examine yourself, act fairly, and be flexible. You cannot let people have the slightest excuse to make trouble.

The trouble that Yung-cheng feared was not long in coming. By early July, 1724, T'ien's efforts to reform local finance were being linked to the oppression of the masses in the execution of public works. Within a short time the interests of local gentry and of opponents of huo-hao kuei-kung were joined in a scandal that could have brought an end to the rationalization of fiscal administration.

The incident that precipitated this crisis was the result of extensive dike repairs being undertaken in Honan. To ensure the success of the project, local effort had been enlisted to supply labor and materials. According to denunciations of T'ien's policies, the financial commissioner, by forcing the magistrates to send all their huo-hao to the provincial treasury, had left them no choice but to confiscate straw from the common people and compel them to transport it to the construction sites. In addition, despite the Ch'ing abandonment of corvée labor, both gentry and commoners were being conscripted to work on the dikes.^[84]

In fact, the situation in Honan was somewhat different from that described. Large numbers of peasants were transporting straw to the construction sites, but only because the government was offering a good price that would probably drop once the new harvest

was in. However, the real target of T'ien's opponents was the government's assessment of labor service on the dikes. In order to hasten completion of the project before the onset of the rainy season, T'ien had authorized the chou and hsien to apportion the necessary number of workers among the people on the basis of land holdings. One man was to be provided for every 150–200 mou of land owned. Each laborer would then be paid a wage calculated according to the amount and difficulty of the work he accomplished. Clearly, the assessment of labor on the dikes was aimed at the wealthier inhabitants of the local community, for few peasants owned this much land. T'ien himself acknowledged that those who would fill the quota were gentry households with large holdings and few adult males. As a result, "(the gentry will) hire poor people to work on the river for them. So apart from their wages, these people will also get paid by the rich to substitute for them. This is a case of the landless poor getting a means of livelihood, not a case of burdening the people."^[85]

Not everyone judged the situation as T'ien did. Before the month was over, 100 Honan sheng-yuan, chü-jen , and chin-shih had staged a protest against the labor assessment. Blocking the reentry of the Feng-ch'iu magistrate into the hsien city, they presented a series of demands aimed at halting the government's employment of hired labor on the dikes. However, their most important grievance was the government's failure to distinguish between gentry and commoners in taxation and assessment of labor service. Vigorous efforts to calm the protestors failed, and in late July lower-gentry discontent reached its climax in a boycott by all but twenty-three of the students enrolled to sit for the hsien examinations.^[86]

The examination boycott was taken very seriously by the Yungcheng emperor. Refusal to take the examinations was one of the few outlets available to the literati to express their displeasure with the government. In effect, such an act challenged the dynasty's right to rule. Despite efforts by the Honan educational commissioner and other high officials to gloss over the affair, the emperor insisted that the leaders be arrested, and in a secret rescript to T'ien Wen-ching, suggested that one or two be executed as an antidote to this kind of wicked behavior.^[87]

T'ien Wen-ching was also worried about the implications of the boycott. Throughout the ensuing month and a half, protest over the labor assessments continued to be linked to dissatisfaction

with huo-hao kuei-kung . Some of T'ien's own subordinates joined in the outcry against remittance of huo-hao to the provincial treasury, charging T'ien with using this innovation to line his own pockets.^[88] T'ien's anxiety was expressed in the growing defensiveness of his memorials, which sought again and again to show that huo-hao kuei-kung was a measure designed to end corruption and not contribute to it. The financial commissioner's fear that he had perhaps gone too far in the interest of reform was partially assuaged by the assurances the emperor appended to his memorials. Time and time again Yung-cheng urged him to ignore the hatred that his zeal had engendered.^[89] In the emperor's own words, T'ien was "one of those who really understands my wishes and I trust you to do your best."^[90]

The emperor's rescripts must have been comforting to this ageing bureaucrat, but Yung-cheng's greatest expression of support for reform came in actions, not words. Three months after these protests took place, T'ien Wen-ching was promoted to the post of governor of Honan.^[91] Of even greater significance, at the height of the conflict over reform in that province, Yung-cheng chose to make his support of the new policies public by presenting the Shansi version of huo-hao kuei-kung for the approval of the court.

From Experimentation to Empire-Wide Implementation

Prior to the summer of 1724, discussion of huo-hao kuei-kung had taken place exclusively through the medium of secret palace memorials. For the first year and a half of his reign, the Yungcheng emperor had been content to allow these officials in the provinces to examine local conditions and present plans for the solution of the fiscal crisis independent of any coercive action from above. Perhaps because of his recognition of the divergence of attitudes and interests between officials serving in the provinces and those in the central government, Yung-cheng preferred to grant provincial officials considerable autonomy in fiscal matters throughout his reign. During the early months of huo-hao experimentation, no edicts were issued on the subject beyond a reiteration of the ban on official extortion of gifts from one's subordinates.^[92] In his rescripts to the secret memorials of those who had put forth solutions to the deficit problem, he had encouraged those whose ideas coincided with his own, but had not advocated any specific course of action.

In mid-July the emperor decided to press the issue of huo-hao kuei-kung by having Shansi Governor No-min request its implementation in a memorial to the Grand Secretariat.^[93] Opposition to the reforms was mounting. In particular, magistrates in the field were gaining allies in their fight against loss of control over informal revenues. Yung-cheng could have avoided intervening in the controversy, but only by risking the abandonment of the reforms where they had already been carried out. Perhaps he felt that a demonstration of his own support for the new policies, however veiled, would force court officials to consider the merits of Nomin's plan. If so, he was soon disappointed. Rejection of No-min's request was swift and unanimous.^[94]

Only two individual responses to No-min's memorial have survived. Shen Chin-ssu, who had already expressed reservations regarding the use of huo-hao to make up deficits, offered new grounds for opposing legalization of this surcharge. In particular, he feared that by converting a surcharge which until now had had no fixed quota to a tax remitted to the provincial treasury, officials would come to regard huo-hao as a regular tax (cheng-hsiang ). In Shen's view, this would inevitably lead to additional levies of huo-hao in order to compensate for that which the magistrates were forced to send elsewhere. The result would only be hardship for the common people.^[95]

An equally emphatic condemnation was put forward by Ch'a Ssu-t'ing.^[96] Ch'a felt that huo-hao should not be legalized, but was willing to concede that it might be necessary to do as a temporary expedient for the purpose of clearing deficits. However, his greatest concern was the lack of any provision for Board supervision of huo-hao revenues in No-min's plan.^[97] As a remedy, he proposed that the financial commissioner of each province compile an account of the amount of huo-hao that could be collected each year and the amounts of the deficits that had to be repaid. On the basis of this information, a schedule of repayment could be drawn up and the Board could monitor the use of huo-hao by comparing the amount collected against the amount of deficits reimbursed. Knowledge of each province's capacity to clear its debts would then allow the Board to set a definite time for the completion of the task, after which remittance of huo-hao could be terminated.

Ch'a clearly had little faith in the moral qualities of most provincial officials, a fact that led him to oppose any plan to use huo-hao to pay for provincial and local public expenses. Not only

would this interfere with the repayment of deficits; it would also provide officials with the opportunity to embezzle government funds. Once again the issue of Board scrutiny of provincial fiscal affairs was crucial to Ch'a's objections. As presented by No-min, huo-hao allocations for public expenses would not be subject to Board approval and would not be reported in the annual accounts (tsou-hsiao ). If officials were able to embezzle millions of taels from the cheng-hsiang that was reported in the annual accounts, Ch'a could imagine only too well what would happen if the provinces were allowed to collect funds over which they alone had control.

The apprehensions expressed by these two men were valid and would continue to worry supporters of rational local funding for many years. The system of informal funding that was now threatened had grown out of the inability of the bureaucracy to cope with the tension between centralization and the need for local autonomy, and that between a belief in low taxation and the fiscal requirements of local government. The official response of the Grand Secretariat to No-min's memorial reflected their unwillingness to tamper with a system that had worked so well, albeit by masking these tensions and encouraging a double standard of morality in official life.

The Grand Secretariat's deliberations were published in the Peking Gazette (T'i-ch'ao ). Although the original text no longer exists, we can reconstruct their argument in the detailed rebuttal offered by the new Shansi financial commissioner, Kao Ch'eng-ling.

Kao Ch'eng-ling's memorial was one of the most convincing defenses of huo-hao kuei-kung written during this period.^[98] Kao began by attacking the basic misconception that huo-hao was the exclusive property of the chou and hsien. The key to his argument was the classical premise that all the land in the empire belonged to the monarch. The greatness of the kingly way (wang-tao ) being rooted in human feelings (jen-ch'ing ), the regular taxes are used to supply the state's expenses, whereas huo-hao is used to nourish virtuous officials. "Ruling men and feeding men are intimately connected." In other words, an official can be a good administrator only if he himself has the means to live. Although huo-hao is the people's money, it is also the dynasty's wealth, because all land belongs to the king. Therefore, according to Kao, the emperor, in his sympathy for his officials, decided to supplement the regular salary of all officials, including governors, commissioners, circuit

intendants, and prefects. These funds were not established for the sake of the chou and hsien alone.

Kao asserted that if, as the Grand Secretaries stated, hao-hsien belongs to the chou and hsien and should not be remitted to their superiors, then they do not understand that there is an intimate relationship between hao-hsien and customary gifts (chieh-li ). If hao-hsien is not remitted to the high officials, then their subordinates must send them chieh-li . Under these circumstances the lower officials weigh the power and influence of the governors, governors-general, commissioners, prefects, circuit intendants, and t'ing officials, and determine the amount of chieh-li to send them accordingly. Gifts are given on the occasion of the Dragon Boat Festival, Mid-Autumn Festival, New Year's, and birthdays. These are called "the four-festivals gifts." To the four-festivals gifts must be added gifts when paying a visit to the superior official (piao-li ). To the gifts on paying a visit must be added gifts of local products (t'u-i ). To the gifts of local products must be added gifts of seasonal delicacies (shih-hsien ). Lower officials send gifts to ingratiate themselves with their superiors. Once they have established a dependency relationship they can intimidate their superior to the point that they can do as they please with no fear of reprisal. Superiors crave the gifts of their subordinates and therefore cover up for them and say nothing when they are involved in malfeasance. Moral integrity is damaged and the official code of conduct is destroyed. The exploitation of the people and the deficits in the state treasury are a result of this.

Kao then anticipated the rebuttal of moralists by pointing out that banning gift-giving and selecting virtuous officials would not solve the problem. Even if chieh-li were prohibited, unscrupulous officials would find a way to defraud and give license to their insatiable demands. Even if there were those who were content with a simple life and wanted to break off the exchange of gifts, high yamen officials would have no other source of funds. They would have to do their jobs without salary while the chou and hsien peacefully enjoyed large profits. Who, Kao asked, could bear that?

For the above reasons, Kao advocated that the huo-hao of the chou and hsien be remitted to the provincial treasury. The high provincial officials would redistribute it, using everyone's huo-hao to supply everyone's yang-lien . This, Kao assured the opposition, was in fulfillment of both heavenly principle (t'ien-li ) and human feelings and was not prohibited by the law of the empire. He

stressed that when huo-hao was remitted, if the whole province incurred unavoidable public expenses, they could be allocated as needed without recourse to assessments on the chou and hsien (fen-p'ai ). If the high officials did not have to levy assessments, then the chou and hsien would not have a pretext to extort from the villages. In this way, remitting huo-hao could also eliminate private surcharges (ssu-p'ai ). Thus Kao sought to transform the very policy that K'ang-hsi refused to endorse for fear of being accused of approving ssu-p'ai into a major force against ssu-p'ai .

Kao then defended the reforms against accusations that it would burden the taxpayers. He showed that in the case of Shansi province the huo-hao in each chou and hsien had been investigated and reduced so that the rates of collection were now half as much as they had been in the past. Moreover, in keeping with the emperor's edict on tax arrears, collection was being carried out in installments over a period of three years so as to ease the people's ability to pay. Against arguments that a fixed huo-hao quota would lead to higher taxes, Kao pointed out that only by fixing the rate of collection could the people know how much tax they should pay and protect themselves against officials who collected taxes in any way they pleased. He linked the fixing of huo-hao rates with the practice of having the individual taxpayer deposit his own taxes in locked chests and reiterated that because the chests were locked and the rates fixed, the chou and hsien could not interfere. Therefore, as T'ien Wen-ching had earlier emphasized, magistrates would have no incentive to collect more than the quota. He also assuaged the fears of those who wondered what would happen to huo-hao collections in famine years by stating, sarcastically, that one did not have to be a sage to know that when the regular taxes could not be paid in full, neither would the huo-hao .

To those who complained that reductions in the amount of huo-hao retained by the magistrates would leave them with too small a sum for their expenses, Kao also had a riposte. On the one hand, if the share of funds received by the chou and hsien now seemed diminished it should be remembered that they would also save on expenses, because they no longer had to send chieh-li to their superiors and no longer had to find their own sources for public expenses. Kao felt that if officials followed the regulations and refrained from bringing large entourages to their posts, and practiced frugality and virtue, then their yang-lien , coupled with miscellaneous duties (tsa-shui ) and surplus, would be more than

enough for their food and clothing. In any case in Shansi, besides their yang-lien , chou and hsien officials also received funds for transportation and smelting taxes, and additional fodder for post horses beyond the Board quota. Busy chou and hsien at much-traveled crossroads were given over 2,000 taels in yang-lien , and that of even the simple and out-of-the-way chou and hsien was not a mere three fen per tael, as claimed by some critics.

Kao finished his memorial by assuring the opposition that using huo-hao to make up deficits did not constitute giving officials with deficits a free ride at the expense of their more virtuous peers. In Shansi the policy was still to impeach officials with deficits and retrieve the missing funds from their office and family property before declaring them wu-cho and repaying them with huo-hao .

Kao conceded that good government depended on men and not laws. In the final analysis there would be no way for him to guarantee that there would not be unscrupulous officials who would abuse the remitting of huo-hao to profit themselves. Nevertheless, he requested that the emperor order the governor and governor-general of each province to follow Governor No-min's memorial to carry out huo-hao kuei-kung . At the end of each year they could then calculate the amount of huo-hao collected, and on the last day of the year memorialize in a t'i-pen how much was disbursed for yang-lien , how much was allocated for public expenses, and how much was retained to make up deficits. In Kao's words, "under the emperor's scrutiny, who would dare embezzle?"

Armed with Kao Ch'eng-ling's penetrating analysis, Yung-cheng again submitted the idea of huo-hao kuei-kung to the court. This time he extended discussion to the Censorate and the Council of Ministers and Princes, which included the top echelon of all the Boards. He cautioned them to deliberate calmly, fairly, and justly or heads literally would roll. In case opinions were split, the emperor granted permission to present two or even three opinions.^[99]

Yung-cheng may have hoped that by expanding the arena of debate he would be presented with a mandate to implement the reforms. There could be no doubt in the minds of the officials asked to comment that the emperor favored this measure. Therefore, it is no surprise that the court submitted an endorsement of the plan. However, as the emperor's reply to their memorial shows, it was an endorsement with grave reservations that reflected the persistent differences between the emperor and his ministers on fiscal organization.^[100]

Originally huo-hao should not exist. However, the expenses and yang-lien of the whole province must come from it. Huo-hao has been collected for many years, but the chou and hsien magistrates still levy additional surcharges and embezzle state funds. Deficits amount to over several thousand taels [in each administrative unit]. The reason is that the magistrates send gifts from this money to their superiors and the superiors demand all their daily expenses from the magistrates. . . . Thus, magistrates have an excuse to satisfy their own greed and their superiors, because they too are involved in wrongdoing, tolerate the situation. These long-standing abuses must be eliminated. Would it not be better for the superiors to appropriate huo-hao to support the magistrates rather than having the chou and hsien save huo-hao to support the superior officials?

You memorialized to set a quota for each subordinate's huo-hao . But I think in each province there are large and small chou and hsien, places with high and low tax quotas. Where a chou or hsien is large and its taxes are great, if you collect huo-hao [at a low rate] it will be enough. But this will not be so in a small chou or hsien with few taxes. Only by not setting a quota can more funds be raised to handle situations of extreme need. On the other hand, if one year a locale's affairs are few and simple, huo-hao can then be reduced. . . . If one sets a quota, in the future it will become an established quota. It will only rise and never decrease. So we cannot set quotas for huo-hao .

As for your suggestion that we should let the magistrate deduct the funds he should receive and not remit it to be allocated back to him: Now when the chou and hsien magistrates collect taxes, the people wrap the money themselves and throw it in the locked chest. If, when the time comes, all the responsible officials open the chests together and examine the contents, and the huo-hao and the regular taxes are remitted together, not a bit can go into their own pockets. Then the magistrates will all know there is no benefit to themselves in collecting exorbitant huo-hao . In that case who will be willing to collect extra huo-hao? On the one hand huo-hao will be used to furnish yang-lien for all the high and low officials. On the other hand, by using it to make up deficits it will be beneficial to the state finances. If you let the magistrates take out the amount they should receive, they definitely will exact more surcharges beyond their quota and burden the people. [An

even more pressing reason is] if you have it all sent to the provincial governor there will be obvious evidence [of how much has been collected and disbursed]. But if it is deducted in the chou and hsien, it will be very hard to verify who is greedy and who is not. This is why the magistrates cannot deduct their hsien-yü .

You also said Governor No-min is incorruptible and very able and that Financial Commissioner Kao Ch'eng-ling's character is very good. You therefore suggested that these two be allowed to wholeheartedly deliberate and test their proposal in Shansi first. This is particularly impermissible. In the world, things either can be carried out or they cannot be carried out. If something can be carried out it can be carried out in the whole empire. If it [is a wrong policy], then it should not even be tried out first in Shansi. It is the same as in the treatment of an illness. If you try every medicine, one at a time, rarely will you get a cure. Thus, I cannot bear for you to use Shansi as an experimental province. Are you trying to say that in the whole empire no other governors or financial commissioners can compare with No-min and Kao Ch'engling. Are these two alone capable of implementing [such a policy]?

You also said that sending the huo-hao to the provincial treasury is not a method which can be carried out for a long time. I think whenever a law is promulgated there is no guarantee that it will never incur abuses. In the past we have had the rule of men and not the rule of law. However, the development of civil or military administration depends on one's policy. If the person [whose strategy it is] remains, the policy will develop. Laws must change to suit the situation. For example, if a person is sick you give him medicine according to the disease and you stop the medicine as soon as the disease is cured. Thus, now to remit the huo-hao is a policy to fit the situation of the present time. In the future, when the deficits are made up and the treasuries are full and officials all know how to be incorrupt, then we will not have to send up [the huo-hao ] and the [amounts collected] can also be reduced.

Are your deliberations for the benefit of the state finances? Are they for the good of the people? They are only for the good of the magistrates. You think the magistrates have their problems, but the high officials have their problems too. Your arguments must be fair and impartial. Furthermore, the court and the people

are one body. If the state finances are sufficient, then, if the people encounter a bad year, there will be enough to furnish them with relief. The people need not worry that there is not enough. This is why clearing the deficits is beneficial to the state's finances and to the people.

Foremost in the emperor's mind was the problem of eliminating corruption by rationalizing the local fiscal base. Yung-cheng located the source of local fiscal instability in the irregular funding practices of provincial officials and not in the inability or unwillingness of the people to pay taxes. More important, he recognized that the problem resulted from the inadequacy of regular fiscal arrangements on the local level. Although he acknowledged that huo-hao had its origins as an illegal surcharge, he now felt that its collection was essential to supply officials with yang-lien and provincial administrative expenses.

Traditional Chinese political economy, with its emphasis on low tax rates, carried with it an implicit injunction against raising the tax burden of the people. As we have seen in K'ang-hsi's rescript to Nien Keng-yao, the late emperor feared that raising taxes by legalizing huo-hao would be viewed as a sign of bad government. In traditional historiography, the three tax rises at the end of the Ming were seen as primary causes of the downfall of the dynasty. Rather than recognize the need for expanded revenues as population and services performed by the government increased, most rulers closed their eyes to the informal funding network, thereby giving tacit approval to corruption. The results were as Yungcheng described them in his edict. Now, for the first time, in place of exhortations to virtue, Yung-cheng hoped to supply the practical prerequisites for good government—a regular and sufficient source of provincial funds, administered through the provincial treasury, but with enough local autonomy to guarantee the flexibility demanded by changing local needs.

As for the method of implementing huo-hao kuei-kung , Yung-cheng clearly favored complete remittance of chou and hsien huo-hao to the provincial treasury. The arguments he presented in favor of this method were precisely those made earlier by T'ien Wen-ching and Kao Ch'eng-ling. However, complete remittance was a measure designed for the sole purpose of curbing corruption and should not be confused with other aspects of the reform. Many later officials interpreted Yung-cheng's remarks—that in the

future when deficits were made up, the treasuries full, and official morale had been restored, remittance would no longer be necessary and huo-hao rates could be reduced—to mean that huo-hao kuei-kung was a temporary expedient to be eliminated as soon as possible. This was a complete misunderstanding. What Yung-cheng meant was that once deficits were eliminated and efficient management restored to local finance, not as much huo-hao would be needed. Once rational financing had become the rule it would no longer be necessary to require complete remittance to guard against official peculation. However, the use of huo-hao to supply officials with yang-lien and public expenses was a policy of lasting benefit to the state and to the people.

Yung-cheng's rejection of the suggestion that huo-hao kuei-kung be carried out first in Shansi was not a rejection of the experimental approach, but rather an indication that he felt that the experiments prior to this point had been successful. The court could not have been ignorant of the fact that in addition to Shansi, remittance of huo-hao was already in effect in Hupei, Honan, Shantung, and Chihli. It is likely that the emperor saw their recommendation as an attempt not to experiment with the policy but to curtail its use in those places where it was already being carried out. By now, the emperor was ready to see the reforms extended throughout the empire.

Even more revealing was the fact that the emperor chastised the court for its reliance on virtuous officials to carry out the reforms. Implicit in the court's remarks was the Confucian interdiction against fa , laws or methods. Yung-cheng's confidence in the primacy of methods was expressed in his insistence that if a policy is good it can be carried out everywhere and if it is bad it should not be carried out at all. What had virtuous officials to do with the issue? The emperor recognized that all laws contained within them the potential for being violated. However, in his own words, "the development of civil or military administration depends on one's policy." In a subtle reinterpretation of the adage that good government relies on the rule of men over the rule of law, the emperor stated that what was important was not the virtue of men (for Yung-cheng had grave doubts about the virtue of most men in public office), but the policies they put forth. With new men, new policies could be implemented to benefit the country and the people. Moreover, as the circumstances of both change, so must the laws. Thus, here as elsewhere, the Yung-cheng emperor freed him-

self from the constraints of tradition to govern in the best way he saw fit.

The policy of huo-hao kuei-kung was not without its own philosophical underpinnings. Concern for the people's livelihood was a fundamental principle of both Confucianism and what Abe Takeo has called "Manchu benevolence." ^{[ 101]} Although on the surface the reforms of the Yung-cheng period appeared to raise taxes, in practice the emperor saw them as a means to lower the gross burden on the people by eliminating all other surcharges. Whereas proponents of Confucian statecraft placed their faith in the morality of the scholar in office and at home to promote the general welfare, early Manchu rulers harbored a deep distrust of the degree-holding elite. It was this distrust that prompted the Ch'ing reduction of gentry tax exemptions, as well as Yung-cheng's expansion of the number of non-degree holders in office.^[102] The emperor felt that court opposition to huo-hao kuei-kung reflected an exclusive concern for the profits that a magistrate could gain by milking the people of their taxes. This abuse of office holding as a source of personal wealth violated the major tenet of Manchu benevolence or what Frederic Wakeman has labelled Manchu paternalism—the unity of national welfare and people's welfare. This concept was expressed in Dorgon's compact (yüeh ) with the Chinese people when he first entered Peking in 1644.^[103] Inherent in the statements of both Dorgon and Yung-cheng was a belief that the corruption of Han officials was the main threat to an almost sacred bond between the dynasty and the people. Moreover, this bond was the foundation for the prosperity of all the inhabitants of the realm.

Embodied in Yung-cheng's fiscal program was a bold challenge to the vested interests of the Han scholar-gentry. Therefore, although his edict was an unqualified endorsement of the huo-hao kuei-kung reforms, the emperor still felt it necessary to proceed with caution. In the face of only grudging acquiescence on the part of officials in the central government and growing opposition on the part of many officials and gentry in the provinces, Yung-cheng was forced to leave the implementation of huo-hao kuei-kung to the discretion of the individual governors and governors-general in the provinces. Once again, the palace-memorial system provided the vehicle to ensure that discretion should be exercised in the way the emperor intended.

4—
Fiscal Rationalization and Local Adaptation

The Yung-cheng emperor's endorsement of huo-hao kuei-kung was the first step toward its implementation throughout the empire. A firm advocate of change, the emperor also recognized that reform of this magnitude could not successfully be accomplished by fiat. Relying on the principle of "adapting one's policies to the time and the place," Yung-cheng left the details of implementation to the governors, governors-general, and financial commissioners of each province. Moreover, the emperor was acutely aware of the conflicts that might arise between provincial and central-government officials on the issue of local financing. He therefore insisted that the funds made available by the reforms remain under the exclusive jurisdiction and supervision of the province in which they originated. During the first decade, the accounting of these provincial funds was completely divorced from the tsou-hsiao system used to regulate the principal taxes. However, as in the formative period discussed in the preceding chapter, the emperor lent a guiding hand through the mechanism of the secret palace-memorial system. By placing men whose approach to government he trusted in the highest provincial posts, and by privately chastising those whose methods he found at odds with the spirit of the reforms, only rarely did Yung-cheng have to exercise his role as supreme arbiter of administrative policy.

Finding a Model

Because no precise regulations were established at the outset to govern the implementation of the huo-hao kuei-kung reforms, some confusion surrounded its early stages. In particular, the provinces differed considerably in the speed with which they moved to institute the new structures for local financing. Resistance, less overt than before the emperor's edict, was also present, though it soon became clear that the method of implementation formulated in Honan and Shansi was to be considered the model for reform. Limitations imposed by differences in local resources, topography, and spending requirements did lead to variations in each province's adaptation to this ideal. However, despite the large amount of local autonomy granted in carrying out these reforms, the differences that emerged between provinces were primarily ones of degree. Moreover, unlike the reforms in tax collection and registration undertaken in the late Ming and early Ch'ing, the fiscal reorganization during the Yung-cheng period was under the strong direction of the governors and financial commissioners from the beginning, so that variations within provinces were minimal. Before turning to the problems encountered by individual provinces and their solutions, let us first examine the major structural elements common to all provinces and the principles that guided their implementation.

Behind all the structural innovations encompassed by the huo-hao reforms was the principle of "using public funds to carry out public business" (i-kung wei-kung ). On the surface, such a statement seems quite simplistic. However, in the light of our analysis of the previous funding system, it implies several very important advances in fiscal thinking. First, a distinction was made between tax revenues belonging to the state and those belonging to the province. The kung referred to in this phrase must be understood to mean provincial as well as public. Under the reforms a new category of taxes was created, which was generated in and used by the province itself. Taxes of this kind, emanating primarily from huo-hao , but including other types of levies, were called kung-hsiang . This term was used to distinguish them from cheng-hsiang taxes paid to the state and used to meet expenses considered to be the responsibility of the central government.

Secondly, i-kung wei-kung was a confirmation of the need to provide the administrative units at each level of provincial govern-

ment with their own legitimate sources of income to carry out their business. As we saw in the preceding chapters, even before the Yung-cheng period there was allusion to an ideal of funding adequate to meet an official's expenses and discourage recourse to peculation. Discussion of this kind, however, was almost always couched in very personal terms. Most often an official who feared encroachment on his informal funding sources would claim that items such as huo-hao or lou-kuei were necessary to permit him to "nourish his virtue" (yang-lien ). In fact, the virtue nourished by such means was tenuous at best. Because no attempt was made to calculate the actual expenses incurred by officials at various levels of the administration, the funds required to nourish his virtue were determined by the official himself, a decision tempered only by what the people providing those funds were willing to tolerate. Inasmuch as neither the official's personal expenses, nor most administrative and service costs were provided for through legitimate government channels, no distinction was made between public and private means or ends in the operation of the official yamen. Huo-hao kuei-kung was thus most of all an effort to eliminate corruption by setting limits on any official's access to funds and by providing him with enough public funds so that he would no longer have a pretext to levy additional surcharges or extort additional lou-kuei from his subordinates. Thus, the term kung signified not only a distinction between central and provincial funds, but heralded a strong attack on the privatization of fiscal affairs as carried out under the informal funding network. This separation of public and private was also manifested in the way funds were allocated, differentiating for the first time between an official's private costs and the expenses incurred in governing his area.

The notion of using public funds to pay for public expenses manifested itself in another way. Because officials were to be granted funds from the provincial treasury, the intricate network of personal relationships upon which the informal funding system revolved would be eliminated. An official's expenses would be based on a rational determination of local needs rather than on a particular man's skill at milking those under his jurisdiction. Funds would accrue to the office and to the locale, not to the man. Even in those few instances where funds which had previously been lou-kuei were returned to the public coffers rather than being eliminated, they became impersonal levies, and could no longer be

relied upon to provide the donor with special considerations. Thus, salt fees once negotiated with, and paid personally to, the governor became just one of a number of levies which entered the pool of funds in the provincial treasury to be disbursed throughout the province. Likewise, the funds sent by the chou and hsien magistrates to their superiors became regular, fixed remittances and could no longer be seen as buying them protection from scrutiny over their activities in office.

A final corollary of using public funds for public expenses was an expanded view of the government's responsibility for public projects. With the huo-hao kuei-kung reforms, a vast amount of funds that had previously been diverted to the pockets of local officials was now made available for public purposes. Government re sources could now be applied not only to the limited number of large projects and military undertakings provided by the central government's grants of regular taxes, but to smaller, more localized projects which previously had been neglected or financed through irregular tax levies and contributions.

The two main structural innovations introduced during the huo-hao kuei-kung reforms were "nourishing virtue silver" (yang-lien yin ) and "public-expense silver" (kung-fei yin ). Putting aside for the moment the sources of these funds, we can see that the division of provincial resources into these two categories provided both a fixed-base allocation for each level of government and a more elastic, project-oriented fund under the supervision of the provincial governor and financial commissioner. This system of financing was designed to suit the requirements of a bureaucracy in which administrative costs were fairly stable and service expenditures were often temporary, large, and unexpected.

Yang-lien yin has been translated as "supplemental salaries."^[1] This designation is useful in calling attention to the huge increase that yang-lien yin represented over the emoluments granted officials from the regular tax revenues.^[2] However, the term salary tends to connote a separation between public and private income that may be misleading in the context of Chinese imperial political economy. A more useful distinction is that between "inner" and "touter."

A provincial official both lived and worked in his yamen. While in office there was nothing he did that did not pertain to his capacity as an official. Individual temperament might lead to differences

in consumption, but a higher level of what we would call personal living standards was expected of an official at the upper level of government than of one at the lower level. The entourage he brought with him to office would be larger, as would the staff he required to help him govern. Although the yamen itself was divided into inner quarters that housed the official's family, personal servants, and secretaries, and an outer section in which government business was conducted, all who worked and lived within this august compound, by virtue of their service to the official, were vital to his functions as an agent of government. This is clearly so in the case of private secretaries who, although hired by the official himself, functioned in much the same way as a modern administrative assistant does in the United States. However, even a personal servant might assist in the conduct of official business when a trusted agent was necessary. Thus, just as previously there was no provision of funds for any of the diverse activities within the yamen (besides wages for a limited number of yamen runners), the total allocation of yang-lien yin was seen as providing for all yamen activities. It was used for the personal living expenses of the official and his family and salaries for the staff, as well as for more clearly administrative expenses associated with stationery costs, routine government procedures, tax collection, catching criminals, and so on.

These "inner" expenses may be contrasted with those costs incurred for defense, or in the provision of funds for construction projects or services of benefit to the local population and the provincial government. The latter, including wall and bridge repairs, granary construction, waterworks, and the like, were "outer" expenses, and, as we shall see, were paid for out of the second category of funds, the kung-fei yin .

The first problem faced by those carrying out the reforms was to determine the amount of yang-lien yin for each official and how the officials were to receive their share. These issues went to the very heart of the reforms. As we have seen, two approaches had been tried during the period before Yung-cheng's edict calling for empire-wide implementation. In both Honan and Hupei, funding by means of customary fees and gifts was banned and in its place the governor of each of the two provinces sanctioned the collection of a flat percentage charge on all land and head taxes. This charge, called huo-hao , was then to be used to provide officials

with yang-lien yin and kung-fei yin and to make up wu-cho deficits.

The Hupei Model

The rudimentary huo-hao remittance scheme first proposed by Hupei Governor Yang Tsung-i in mid-1723 was refined during the next year and extended to include Hunan. This new plan was designed not merely as a temporary device for eliminating deficits, but as a lasting system for providing officials at many levels of government with yang-lien yin . By the middle of 1724 the rate of collection of huo-hao was set at 10 percent. Thus, for every tael of regular taxes (cheng-hsiang ) collected, an additional 0.1 tael was appended for provincial expenses. For every 0.1 tael of huo-hao the chou and hsien received from the taxpayers, 0.068 tael was remitted to provide funds for their superiors. Allocation of this 0.068 tael was broken down as follows: 0.015 tael for the financial commissioner's yang-lien , 0.004 tael for the yang-lien of that chou or hsien's circuit intendant, 0.01 tael for its prefect, 0.003 tael for its subprefect (t'ung-chih ), and 0.03 tael for kung-fei . The remaining 0.032 tael was retained by the chou or hsien magistrate for his own yang-lien . Those chou and hsien with taxes of 5,000 taels or less were exempt from remittance because their total huo-hao would be 500 taels or less.^[3]

This system was a great advance over the informal funding network it superseded. However, it contained a number of disadvantages that led the emperor to bypass it as a test case when presenting the reforms to the court. In fact, an examination of the weaknesses in the Hupei method can serve to underline some of the major objectives of the reforms as they were finally formulated in Honan.

First, the Hupei plan was not equitable in its distribution of funds. An official was not granted a fixed amount of yang-lien yin based on need, but rather received a percentage of the taxes collected in his area. Because the taxes collected by the chou and hsien ranged from several thousand to tens of thousands of taels, some magistrates would be left with only a few hundred taels whereas others would enjoy several thousand. Moreover, by allowing those areas with a tax quota of 5,000 taels or less to retain all of their huo-hao , a chou or hsien with 4,900 taels of taxes

would retain 490 taels, but a chou or hsien with 5,100 taels of taxes would have to share its huo-hao in the same way as a very wealthy area, leaving the magistrate with only 164 taels with which to operate his yamen. Inasmuch as tax collection was one of the most important and demanding functions of local government, and a high tax quota often did signify that an area was "busy" in other respects as well, it is not unreasonable that some officials favored granting yang-lien yin as a percentage of taxes collected. It is also true, however, that poorer areas were in need of far more funds than they could generate internally. Their inhabitants were less likely to be able to contribute to projects such as waterconservancy works, and because many poorer areas were located in remote or mountainous regions they often had a higher incidence of banditry and incurred greater costs for the transportation of goods and taxes and for travel on official business.

The second weakness of the Hupei plan lay in its method of yang-lien distribution. Higher officials continued to receive their own operating expenses in the form of small contributions directly from the chou and hsien. This not only was less convenient than receiving one's funds in a lump sum, but also invited a reversion to the kinds of extortion and blackmail that had been rampant under the informal funding system. In addition, because the magistrate collected the huo-hao and deducted his own share before remitting the share designated for his superiors, there was no institutional check to prevent him from collecting more than he was supposed to. In effect, the fiscal management of the entire province was diffused in the hands of scores of magistrates. Under these conditions the potential for rational supervision of local finances at the provincial level was small indeed.

A final weakness of the Hupei plan was its failure to provide any yang-lien yin at all for a number of provincial officials, most obviously the governor and governor-general. Neither were provisions made for the grain, post, and salt intendants and the educational commissioner. On a lower level, yang-lien was allocated for the subprefects but not for the assistant prefects (t'ung-pan ), who, as Examiner Chuang Ch'ing-tu pointed out, had many expenses connected with their responsibility for the tribute grain.^[4] A number of other officials were excluded from the original reforms everywhere, and we shall discuss provision of yang-lien for them later.

The Honan Model

The second approach to the allocation of yang-lien is best represented by the method instituted in Honan under the guidance of Financial Commissioner T'ien Wen-ching.^[5] Prior to T'ien's arrival in Honan, Governor Shih Wen-cho had already issued an injunction against the collection and remittance of customary fees to high government officials.^[6] In addition, the collection of a fixed percentage of huo-hao was legitimated. However, until T'ien Wenching's arrival, the focus of attention was on using huo-hao to clear up wu-cho deficits, and the regulations governing the use of these new funds for yang-lien yin and public expenses were vague.

In a memorial dated the third month of Yung-cheng 2 (1724), Governor Shih declared that all huo-hao from the chou and hsien would be remitted to the public coffers, in other words, to the financial commissioner's treasury. Although high officials would no longer have kuei-li and chou and hsien officials would no longer have huo-hao , all officials of the rank of hsien magistrate and above would now be granted yang-lien yin in sufficient amounts to meet their expenses. In this way, officials could perform their duties without concern about their finances.

Total huo-hao revenues at a rate of 10 percent of the land and head tax were estimated at 400,000 taels. On the basis of this figure the following schedule of fixed yang-lien allowances was formulated:^[7]

In addition, Shih requested yang-lien for the governor based on the amount of funds that that office used to receive under the old system, approximately 30,000 taels.

Besides paying for each official's yang-lien , Shih stated that huo-hao would also be used to provide for the public expenses of the whole province as well as to supplement certain fixed local costs such as the purchasing of fodder for post horses and goods sent to the capital, the travel expenses of officials sent to the chou and hsien to supervise the opening of the tax envelopes, the travel expenses of messengers on official business, and the transportation expenses incurred by the chou and hsien in remitting their taxes. The total annual expenditure of huo-hao for yang-lien and public expenses was estimated at 245,000 taels, leaving about 150,000 taels to make up deficits and repay funds borrowed from the regular tax for such purposes as buying additional tribute grain.^[8]

The emperor commended Governor Shih on his plan, adding a note of satisfaction that Honan was now following the method first implemented in Shansi under Governor No-min. Interestingly, the only criticism Yung-cheng offered was that the yang-lien of certain chou and hsien might not be large enough. He suggested that after all the deficits were cleared, yang-lien rates be reevaluated,^[9] confirming the view that the emperor did not see huo-hao kuei-kung simply as a temporary expedient to overcome the deficit crisis.

The Honan plan represented a considerable improvement over the method of implementation utilized in Hupei. All huo-hao was remitted to the provincial treasury for redistribution in the form of yang-lien and kung-fei allowances. Coupled with the practice of sending officers from outside the chou and hsien to supervise the opening of the tax chests, this marked a great step forward in the control of surcharges and corruption on the lowest levels of administration. As T'ien Wen-ching pointed out, because magistrates

kept none of the land and head taxes they collected, they no longer had any incentive to collect at rates beyond the legal quota.^[10] Moreover, as all officials received their yang-lien in quarterly installments, directly from the financial commissioner,^[11] they were relieved of both the inconvenience and stress of receiving their operating expenses in dribs and drabs from their subordinates. In addition, yang-lien was given first priority in allocating huo-hao , so that officials were guaranteed receipt of their expense fund regardless of difficulties arising from natural disasters, arrears, or high provincial-level expenditures. By divorcing funding concerns from the relationship between officials and their supervisors, the Honan system eliminated at least one of the sources of extortion and favoritism characteristic of financing under the informal funding network.

A second advantage of the Honan method of distributing yang-lien was that it provided a more equitable allocation of funds based on need and not on rank or the tax income of a particular area. This too was made possible because all the huo-hao in the province was pooled and reallocated in fixed amounts. The attention given to need is illustrated particularly in the case of magistrates' yang-lien . The chou and hsien were divided into three ranks based on the degree of public business in the area, location at important crossroads, and distance from the provincial capital, and yang-lien was granted accordingly. Circuit intendants were given larger stipends in view of the extensive travel their jobs entailed. This in turn relieved the chou and hsien of the responsibility for housing, feeding, and providing transportation for supervisory officials' tours of inspection. The K'ai-feng circuit intendant and prefect were given additional funds reflecting the added burdens placed on officials in the capital precincts. The high yang-lien granted the financial commissioner is particularly noteworthy, not only as a manifestation of the increased prestige and power of that post in the Ch'ing, but because it specifically included added provisions for the financial commissioner's yamen's kung-fei . This kung-fei was in response to the expanded accounting responsibilities that the implementation of huo-hao kuei-kung would entail.^[12]

It should be noted that as conceived in Honan, yang-lien was not meant to cover all the expenses that an official would incur in office. At the very least, besides those expenses customarily viewed as a provincial responsibility, such as river works and major con-

struction projects, the chou and hsien were to be provided with additional funds to cover travel and transport costs and the costs of tribute goods and post-horse feed. All of these expenses had previously been major drains on local resources.^[13]

Finally, the Honan plan was more comprehensive in its provision of yang-lien for all officials at the level of magistrate and above. Later, yang-lien would be extended to lower officials, with Honan once again taking the lead. From the start, yang-lien rates were not considered immutable. Not only could the more flexible funds stored in the provincial treasury as kung-fei be drawn upon to increase yang-lien allocations, but, as we shall see, new sources of funds could be earmarked for provincial use now that the concept of "return to the public coffers" (kuei-kung ) had received the imperial blessing.

The methods of implementation used by the remaining provinces combined elements of both the Honan and Hupei plans. Professor Abe Takeo, in his pioneering work on huo-hao , divided the methods of remittance that appeared in the course of the reforms into three categories: total remittance, remittance of most huo-hao , and remittance of little huo-hao .^[14] Total remittance followed the method first utilized in Honan and Shansi. The huo-hao was collected by the chou and hsien at a fixed rate and sent in full to the financial commissioner for allocation in the form of fixed yang-lien allowances and flexible public-expense disbursements. In the second form, huo-hao and yang-lien rates were also fixed, but the chou and hsien were permitted to deduct their own yang-lien before remitting their huo-hao to the financial commissioner. Other provincial officials then received their yang-lien directly from the provincial treasury. The last category most closely parallels the method first utilized in Hupei. The chou and hsien collected their fixed rate of huo-hao but remitted only that portion designated for provincial kung-fei . They retained their own yang-lien and were responsible for distributing any yang-lien due higher officials to the officials individually. Table 4.1 summarizes the methods of implementation used and how they changed during the course of the Yung-cheng reign. It should be noted that throughout this period there was a definite trend toward closer conformity with the Shansi-Honan model.

In the remainder of this chapter we examine the reforms as they were carried out in each province and the problems encountered which led to local variations. In chapter 5 we then discuss some of

the major themes and continuities underlying local variations and evaluate the implementation of the reforms in general.

Suiting One's Policies to the Time and the Place

Part of the appeal of huo-hao kuei-kung was its simplicity. The legalization of one surcharge would provide provincial officials with a predictable and reliable source of funds for all local-government expenses not covered by the regular system of revenue sharing. Because huo-hao was levied on the preexisting land and head tax, it required no new bureaucracy to administer its collection and entailed little additional effort to compile registers and accounts. Likewise, it could be integrated into the system in which the individual wrapped and deposited his own taxes, diminishing the magistrates' reliance on corrupt yamen runners and clerks and enabling the government to notify the taxpayer of the precise amount of huo-hao due. Because it did not consist of taxes of diverse origin, budgeting allocation of huo-hao income was a relatively easy task and left little room for manipulation by officials seeking to profit from the new system of funding.

The successful implementation of such a system was predicated on several conditions. Most important was the existence of a large

land- and head-tax quota. If the agricultural productivity of a province was small, even a very high percentage charge on the regular tax would yield inadequate revenues to provide for all of its administrative needs. Moreover, resort to high surcharges would risk arousing discontent among the local population. The second requirement was that the land and head tax be readily collectable. If tax evasion or arrears were endemic, then even a high quota on which to base huo-hao levies would not guarantee local officials a stable source of income from year to year.

For these reasons, huo-hao kuei-kung , as originally formulated, was most successful in a few provinces in north China where experimentation had already been under way prior to the summer of 1724. In Shansi, Honan, and Chihli annual cheng-hsiang tax quotas all exceeded 3 million taels. Shensi's more modest quota was offset by a somewhat smaller bureaucracy and a lower level of necessary government services.^[15] Equally important, most taxes in these four provinces were collected from independent, smallholding peasants. Land tenancy was not unknown in the north, but played a far smaller role in the rural economy there than in the south. In addition, the earlier reforms in land-tax collection had proceeded further in the north, and the relative similarity in land quality and the slower progress of commercial agriculture made tax rates and collection more uniform than in China's major rice-producing provinces. The rapid surrender of much of north China's bureaucracy to the Manchu conquerers also provided the Ch'ing with intact land and tax registers for that region, whereas prolonged resistance and the flight of local officials left southern yamen records woefully incomplete. For decades, southern-gentry opposition to Manchu rule continued to be expressed in the form of tax evasion, and even where remedied by legal means, this greatly impaired the efficiency of local tax-collection efforts.^[16] Differences in land tenure and the existence of large concentrations of wealth may also account for the greater power exercised by the elite in south China. Although the Honan examination boycott of 1724 caused the dynasty considerable embarrassment, the gentry of Feng-ch'iu were unable to rally sustained support for their cause.^[17] Whereas gentry interference in tax collection and involvement in tax farming was a constant theme in south China, in the north it is rarely mentioned prior to the nineteenth century. Floods and drought always threatened, but the imperial system of tax re-

missions could respond with exemptions and postponements so that arrears were far less common than in the south, where both tax resistance and submagisterial corruption presented serious obstacles to reform.

The problems posed by the structure of society beyond the north China plain delayed implementation of huo-hao kuei-kung in many provinces for several years. Differences in tax base, tenancy relationships, commercial activity, and the structure of local power led to the evolution of different patterns of reform. In agriculturally underdeveloped provinces such as Yunnan, Kweichow, Kwangsi, and Kansu, land- and head-tax quotas which totalled less than the huo-hao of more advanced provinces were inescapable barriers to the adoption of the Honan model. Only by the creative use of local commercial and mining revenues were these provinces able to provide adequate income for local government.

However, backwardness was not the only obstacle to reform. The high tax quotas of provinces along the southeast coast were not simply a reflection of higher agricultural productivity. Cheng-hsiang tax rates were themselves higher in the south, a fact which militated against the huo-hao levies of 13 to 20 percent that were common in the north. Thus, despite their greater wealth, several southern provinces consumed an inordinately large portion of their total revenues on yang-lien , leaving little for productive provincial expenditure. (See table 4.2.) What remained was often wiped out by shortages incurred in the process of tax collection or coopted by the central government in the form of Board fees and military expenses. Here, too, revenues from commercial duties and surplus income from government monopolies came to play a part in ensuring the successful implementation of the reforms. Of equal importance was the central government's own efforts to rescue the reforms by providing grants to supplement provincial budgets and by relieving some of the burden of defense responsibilities formerly relegated to kung-hsiang .

The adaptation of the original Honan model that was achieved in each of these provinces attests to the remarkable flexibility permitted by "provincial management of provincial revenues." By the end of the Yung-cheng reign, every province in China had implemented its own variation of the original theme of local fiscal rationalization, making use of its own particular resources and satisfying its own special needs.

The Sweet Smell of Surplus: Huo-Hao Kuei-Kung in North China

The most striking manifestation of huo-hao kuei-kung 's success was the appearance, within a few years of its inception, of large surpluses in provincial revenues. After years of struggling with deficits, provincial officials in north China found themselves with more money than they knew how to spend.

The most dramatic reversal of provincial fiscal fortunes occurred in Honan. Here, Governor T'ien Wen-ching's careful con-

trol of corruption combined with a large and relatively easily collected tax quota to produce an accumulated surplus that reached as much as 600,000 taels.^[18] Because the repayment of deficits had also meant replenishing granary stores, T'ien first applied this windfall to the construction and repair of granaries throughout the province.^[19] Cautious lest a natural disaster occasionally prevent the collection of taxes in full, the governor set the remainder aside as a cushion for the future. At the emperor's urging, a portion of these idle funds was put to more immediate use. Public expenditure for the repair of city walls, temple and public-building construction, and famine relief were substantially increased, as was the yang-lien of those officials whose posts brought them into closest contact with the people.^[20] As a result, the long-neglected physical substructure of the province was greatly improved and low-level officials in Honan became among the highest paid in the empire.

As a major proponent of the huo-hao kuei-kung reforms, T'ien Wen-ching was convinced of the link between adequate salaries and the ability of officials to carry out their responsibilities honestly and efficiently. Therefore, he moved next to utilize his province's large surplus income to provide yang-lien for officials below the rank of magistrate. These so-called minor officials performed duties that brought them into constant touch with the people. Many were charged with catching criminals and with supervising the pao-chia organizations of the local populace, and they were frequently sent on missions outside the chou or hsien capital, all of which involved heavy expenses that should not have to be passed on to the local population. T'ien first offered this proposal in 1725, but was turned down by the emperor on the ground that other provinces might not have enough funds to follow Honan's example.^[21] That Yung-cheng approved the governor's plan three years later was an indication not only of the success of the reforms in Honan, but also of the emperor's growing confidence that other provinces would soon be able to reward their minor officials in the same way.^[22] This innovation provided assistant and submagistrates of independent chou an annual yang-lien of 120 taels and jail wardens and assistant and submagistrates of regular chou and hsien 80 taels, at a cost to the province of 10,600 taels per year.^[23]

The above increases in huo-hao expenditure do not seem to have seriously decreased Honan's provincial surplus. In 1739, Honan still had uncommitted revenues in excess of 400,000 taels.

As we shall see, by the beginning of the Ch'ien-lung reign, officials in other provinces where the reforms were not as successful in providing for all provincial expenses began to eye Honan's abundant treasury stores with envy, an envy that was to undermine Yung-cheng's principle that the huo-hao revenues of a particular province were to be its inalienable property.^[24]

Honan was not alone in producing a surplus of huo-hao revenues. By the end of the Yung-cheng reign, similar successes could be claimed by Shansi, Shensi, and Chihli as well.^[25] Like Honan, they achieved their success without having to resort to significant infusions of funds in addition to huo-hao . In this respect, they constituted the only provinces in which huo-hao kuei-kung functioned in the manner envisioned at its inception. Nevertheless, even in these provinces the implementation of the reforms did not proceed without problems.

Although Shansi had been the test case for the introduction of huo-hao kuei-kung , officials there were never able to enforce compliance to a fixed rate of huo-hao . Under Governor No-min the rate of 20 percent that was established seems to have served more as a guideline than as a definite quota.^[26] His successor, Chueh-lo Shih-lin, lowered the rate to 13 percent as the provincial debt was gradually paid off. However, even then the rate continued to range in practice from 14 to 20 percent.^[27] Despite this lack of uniformity, the reforms do appear to have held magistrates to a maximum levy of 20 percent. In at least one instance where this limit was exceeded, the magistrate, Wang Meng-hsiung, was criticized by the people of his county and a special investigating team was able to force him to lower his rate to 17 Percent.^[28]

Violations of a standard rate of huo-hao was less troublesome than magisterial resistance to the reforms. In some counties, magistrates remained wary of the province's promise of funds for local projects and continued to collect surcharges on an ad hoc basis.^[29] In others, magistrates resisted remittance of their huo-hao to the provincial coffers.^[30] In the first year of the reforms, this cost the province 60,000 taels of its almost 500,000-tael huo-hao quota.

Many of these problems were alleviated when Chueh-lo Shih-lin took over as governor in 1727. Following a thorough investigation of provincial revenue management over the preceding five years, he established strict guidelines to prevent excessive expenditure.^[31] For expenses of a relatively predictable nature, quotas were set. (See table 4.3.) This marked an important step in the rationalization of

provincial financing. Whereas earlier in the reforms the only effort at accounting had been to add up the funds expended and saved each year, as the reforms began to be accepted by all officials and remittances became more reliable the governor was able to budget allocations for the coming year, based on accurate information on spending and a consistent flow of silver into the provincial treasury. Nevertheless, there was still a great deal of flexibility in these allocations in that the ultimate decision to disburse funds still remained the responsibility of the provincial governor.

Chihli and Shensi also suffered from initial confusion in the implementation of the reforms. It was not until YC 7 (1729) that Chihli instituted a standard 13-percent rate of collection and complete remittance to the provincial coffers.^[32] Although a modified version of direct remittance had been introduced in Shensi by Governor-general Nien Kung-yao, his successor, Yueh Chung-ch'i, discovered that little effort had been made to hold magistrates to the rate established at that time, a more generous 20 percent.^[33] Indeed, under Nien, local officials had been permitted to increase or decrease their huo-hao levies in response to the size of the deficits incurred in their chou and hsien, a policy that differed little from the common practice prior to the reforms.^[34]

The haphazard management of huo-hao in Chihli and Shensi was rectified with the help of vigorous provincial leadership appointed later in the Yung-cheng reign. Enforcement of direct remittance of huo-hao to the provincial coffers ended most of the problems stemming from excessive exactions from the taxpayers.

In Shensi, Yueh Chung-ch'i took the added precaution of ordering the accepted rates of huo-hao posted in every city and village, in the belief that an informed public was the best defense against official malfeasance. New regulation scales were issued to every chou and hsien to prevent the use of weighted balances to manipulate tax rates, and village headmen were instructed to report any magistrate accused by the people of overcharging.^[35]

In all four northern provinces, a little over half the provincial revenues were earmarked for allocation as yang-lien . This left a sizable allowance for those expenses designated as kung-fei or public expenses. In Honan, Shensi, and Shansi, officials received special allowances of kung-fei in addition to those funds set aside for the public expenses of the whole province. We are fortunate to have a complete list of the expenditures made from Shensi's kung-fei in YC 4 (1726). From this we can see that kung-fei yin did indeed pay for expenses of a provincial nature, including repairs to yamen buildings, watchtowers, examination halls, transportation costs of officials on provincial business, city repairs, and special purchases and expenses incurred by high-level provincial officials. (See table 4.4.) Governor Yueh pointed out that some of these

costs would previously have been raised through levies on the villages in the areas where projects were being carried out. By dividing the expenses among all the people in the province, the inhabitants of frontier areas and areas with a high concentration of official yamen were thus given a measure of much-appreciated relief.^[36]

Only Chihli appears to have had inadequate funds for public expenses. A large portion of its kung-fei was initially remitted to the Board of Revenue in the form of fees that the province had to pay when submitting its taxes.^[37] This was partially remedied by the absorption into the provincial budget of 7,000 taels in weighing fees previously paid at tax-collection time to the officers and clerks in charge of the provincial treasury.^[38] However, in the following year this added revenue was more than offset by the allocation of over 8,600 taels to provide minor officials with yang-lien . It was not until YC 12 (1734) that Chihli was able to join the other northern provinces in providing local officials with public-expense allowances as well as much-needed increases in their yang-lien .^[39] This was accomplished without an increase in huo-hao as a result of a decrease in the fees demanded of the province by the Board.^[40]

The success of the reforms in these four provinces demonstrated without question that the rationalization of local-government finances was an attainable goal. However, only creative adaptation of the original model would enable this success to be translated to the majority of provinces beyond the north China plain.

Overcoming Backwardness: Huo-Hao Kuei-Kung on the Periphery

Unlike the flat, fertile north China plain where huo-hao kuei-kung was first conceived, China's southern and western frontiers were largely mountainous regions which, in the early eighteenth century, contributed little to the fiscal resources of the state. Large areas of the border provinces were occupied by aboriginal peoples whose integration into the mainstream of Chinese life had barely begun. Even Szechuan, which is now one of the most populous and agriculturally productive provinces in China, suffered in the early Ch'ing from the devastation of land and inhabitants that was the legacy of Chang Hsien-chung's war against the last rulers of the Ming. Only Kwangtung could boast a land- and head-tax quota comparable to those found in southeast and central China. However, poor inland communications made direct remittance of

huo-hao to the provincial treasury impractical.^[41] Elsewhere on the frontiers of the empire, inadequate revenues combined with rugged terrain to necessitate a completely new approach to the reforms mandated by huo-hao kuei-kung .

Szechuan was the only border province to attempt to follow the Honan model in implementing the rationalization of localgovernment finance. This was not an easy task. In YC 5 (1727), when Shensi-Szechuan Governor-general Yueh Chung-ch'i sought to apply the lessons learned in Shensi to his more southerly jurisdiction, he found that the latter yielded an annual regular tax quota of only 333,567 taels.^[42] In order to generate sufficient huo-hao to supply every official in the province with yang-lien , the rate of collection was set at an unprecedented 30 percent. In this manner, the taxpayers generated an extra 1000,070 taels for provincial use. However, because Szechuan's yang-lien quota consumed 99,414 taels (see table 4.5), little was left for expenses of a more general kind. To remedy this, Yueh and the Szechuan governor, Hsien Te, proposed the continuation of an informal levy known as pang-t'ieh kung-shih , formerly collected illegally from the villages to supplement the costs of running the local yamen.^[43]

Such a high rate of huo-hao was not desirable. Even the officials responsible for its introduction declared their intention to lower it as soon as land reclamation had sufficiently increased the tax base. The emperor himself expressed displeasure at the way huo-hao kuei-kung had been accomplished, but decided to grant approval pending a thorough survey of the taxable land in the province. In the meantime, Yueh once again turned to an informed public as a safeguard against further exactions. Signs were posted in all towns and at major crossroads notifying the public of the new rates and of their right to report any official guilty of exceeding them.

By the summer of 1729 the land survey was under way. Behind this investigation of taxable land in Szechuan was the assumption that far more land had been returned to cultivation than was indicated in existing county records. Indeed, as we have seen, as long ago as the K'ang-hsi reign, some officials had suspected the people of Szechuan of scheming to keep their land off the government's tax rolls. Therefore, it should have come as no surprise when the people of Wan hsien seized upon the arrival of the land-survey team in their vicinity to protest against the government's taxation policies. What had not been expected was that the heavy rates of pang-tieh now collected in concert with huo-hao would become the focus of popular discontent. Convinced that the real target was the land survey itself, Governor Hsien removed the inept censor heading the team in Wan hsien and dispatched a more-experienced investigator in his stead. The disturbance was quickly brought under control and the governor felt vindicated in his assessment of the crisis.^[44]

The connection between the land survey and anger over huo-hao and pang-tieh was not as tenuous as portrayed by Hsien Te. So long as a substantial portion of Szechuan's reclaimed land was hidden from government scrutiny, some segments of the population were probably quite willing to tolerate high levels of extra-quota fees. During the K'ang-hsi period, Nien Keng-yao had reported that Szechuanese taxpayers were supporting a level of surcharges as high as 50 percent of the regular tax. However, the prospect of accurate tax registration and high huo-hao was too much for the local inhabitants to bear. Soon after the Wan hsien protest, the taxpayers of Szechuan found an ally in the governor of neighboring Kweichow province. In an impassioned memorial to the emperor, Chang Kuang-ssu reported that whereas in his tax-poor province huo-hao was only 20 percent of the regular tax, in Sze-

chuan, the combination of huo-hao and pang-t'ieh raised the additional liability in some places to an outrageous 55 percent. Chang felt this could not be justified once the present land survey had augmented Szechuan's tax rolls, and he suggested that the huo-hao rate be lowered to match that in Kweichow. If this was insufficient to provide for all the province's needs, then magistrates could be authorized to collect a pang-t'ieh of no more than 5 Percent.^[45]

Governor Hsien was not averse to such a plan, but intended to await the completion of the land survey before making any adjustments in existing rates. However, Chang Kuang-ssu's plea on behalf of the Szechuan people was soon echoed in an imperial audience by a prefect newly arrived in the capital from a term in Szechuan. As a result, early in 1730 Hsien received notification from the Board of Revenue to lower Szechuan's huo-hao and eliminate pang-t'ieh .^[46]

The governor reluctantly complied with the order, but in subsequent communications with the Board he expressed concern that the levels of revenue generated by the new 25-percent huo-hao would be insufficient to meet Szechuan's administrative needs. Although the Board claimed that it had not meant huo-hao to be lowered until the new quotas for regular taxes were in effect, it could not sanction a rise in the rates so soon after a cut had been announced. Instead it recommended the temporary continuation of pang-t'ieh levies and proposed that discussions regarding further adjustments be held after the new land registers were compiled.^[47] For his part, the emperor attributed the mishandling of the affair to the incorrect referral of huo-hao matters to the Board of Revenue.^[48] The confusion created in Szechuan only strengthened Yung-cheng's conviction that provincial funding had to be kept independent of Board supervision.

The problems encountered in implementing huo-hao kuei-kung in Szechuan served to underscore the difficulties entailed in rationalizing local-government finance in provinces where land- and head-tax revenues were small. In the end it was commercial taxes, in the form of surplus duties on tea and salt, that salvaged the reforms in Szechuan and saved the populace from the reinstitution of pang-t'ieh .^[49]

The availability of surplus revenues from the taxation of commerce and government-run industries was the key to the successful implementation of the huo-hao kuei-kung reforms in each of the other provinces of the northern and southwestern frontier as well.

In Kansu, Governor Hsü Jung estimated that he would require 79,740 taels to pay for the yang-lien and public expenses of the officials in his province.^[50] Even at a rate of 30 percent, huo-hao would not yield sufficient revenues on the basis of the 254,000 or so taels Kansu collected annually in regular taxes.^[51] In order to supplement the income from huo-hao , Hsü incorporated into the provincial budget some 20,000 taels of surplus duties from the frontier trade (shui-hsien ) and 11,900 taels of customary fees sent to the governor each winter by merchants applying for licenses to do business.^[52] These additions to the provincial coffers were still insufficient to provide Kansu with the level of funding found elsewhere in the north. Therefore, in YC 7 (1729), the governor appealed to the emperor to allow some of Shensi's surplus huo-hao to be transferred to Kansu.^[53] Although the alienation of huo-hao from one province to another was generally not permitted, the fact that Shensi and Kansu had long been administered as a unit made an exception possible in this case.^[54]

Perched at the northern and southernmost extremes of China, Fengtien and Kwangsi were handicapped by the lowest huo-hao quotas in the realm. In Fengtien, a land and head tax of only 34,592 taels initially inhibited any efforts to institute huo-hao kuei-kung . However, in 1729 Yung-cheng ordered the provision of yang-lien for all officials not previously covered by the reforms. Even allowing for much lower levels of yang-lien than allocated elsewhere,^[55] the funds available from huo-hao were clearly insufficient to cover all expenses incurred by Fengtien's bureaucratic personnel. Extra revenues from the cart-license fee levied when merchants and commoners carried goods over the border into China proper helped augment the yang-lien of the metropolitan prefect and provided wages and stationery expenses for the clerks in several other yamen.^[56] However, annual income in Fengtien was still too small to permit the allocation of any funds for public expenses.

This problem was remedied by the emperor in YC 10 (1732). On the condition that it not interfere with the collection of the regular commercial-tax quota, Yung-cheng allowed Fengtien's officials to levy a 30 percent surcharge on goods carried by merchants through the Shan-hai pass separating China from Manchuria. It was calculated that this impost could yield between 30,000 and 36,000 taels annually. Revenues derived from this source were immediately applied to the costs of compiling the prefectural gazetteer, paying laborers at the customs stations and

guarding dikes, and repairing public ferries and roads leading to the Fengtien capital.^[57]

In Kwangsi a low regular-tax quota was exacerbated by the 2-percent rate of huo-hao established by Li Fu when he was governor of this remote province. Only 6,000 taels could be returned to the public coffers each year, all of which was set aside for emergencies affecting the whole province.^[58] Nevertheless, later governors were able to tap enough surplus revenues from maritime customs duties, excess commercial duties, and surplus income from government-owned copper mines to pay all officials yang-lien and kungfei and supplement the costs of suppressing an uprising by the province's Miao minority in the latter part of the Yung-cheng reign.^[59]

Yunnan and Kweichow provide the best evidence of the flexibility with which the reforms could be carried out in areas with unfavorable economic conditions. With their large, unassimilated aborigine populations and land- and head-tax quotas of only 172,525 taels and 66,661 taels respectively,^[60] neither could rely on huo-hao alone to meet the expenses of a developing border region. Nevertheless, by making use of income from commercial taxation and the rich mining resources within their jurisdiction, even these two remote provinces were able to show a fair level of success in joining the agricultural heartland in the task of fiscal reform.

Available data concerning the implementation of huo-hao kueikung is more complete for Kweichow than it is for Yunnan. We do know, however, that as early as YC 1 (1723), officials in Yunnan were considering ways to provide funds for provincial expenses. One of the richest sources was Yunnan's numerous copper mines. Although taxes on copper production were the sole property of the central government,^[61] provincial control over copper extraction and marketing yielded a sizable income for local use.

By the early part of the Yung-cheng reign, the Ch'ing government was beginning to look favorably upon private operation of mineral extraction.^[62] However, lack of capital and poor labor productivity still discouraged large-scale merchant investment in such enterprises.^[63] Miners were considered shiftless and unruly, and honest and efficient managers were also difficult to find. Transportation posed another problem. Ores had to be carried from the mines by teams of pack animals driven by aborigine tribesmen who alone had sufficient knowledge of Yunnan's rugged terrain. Risk of loss was great. If they were attacked by bandits or their

animals collapsed under their burden, the porters would simply abandon the copper and flee. Some no doubt themselves stole the load they were hired to carry. Even the disposal of this essential commodity posed problems. The market in copper was extremely erratic, necessitating the warehousing of bulky ores for several years at a time, only to see the entire stock sold out in a matter of months when demand suddenly increased.^[64] As a result, government still dominated the mining industry in the southwest.

Despite these difficulties, Yunnan was able to net a profit of 20,000 taels from its copper mines, all of which was used to supplement the province's meager income from huo-hao .^[65] Looking forward to the day when most mines would be privately run, provincial officials also proposed a fee of 20 percent of the output of such enterprises as an additional contribution to the provincial coffers.

Copper was not the only commercial source Yunnan tapped to enhance the funds available for public expenses. Salt fees traditionally sent by merchants to the governor and governor-general generated more than 64,000 taels for yang-lien and kung-fei . Government operation of the salt wells themselves yielded another 65,721 taels after remittance of imperial salt quotas and the payment of wages and expenses at the wells.^[66] Provincial officials also expressed hopes of considerable profits from the management of Yunnan's developing silver-mining industry.^[67]

Given the low land- and head-tax quota and the availability of such large sums of nonagricultural surplus income, it is not surprising that Yunnan's high officials first responded to the huo-hao kuei-kung reforms by instituting a system of limited remittance to the provincial treasury. Whereas revenues from salt and mining were used to pay for provincial-level projects and the yang-lien of high-ranking officials, the chou and hsien were allowed to retain their own huo-hao to cover local expenses. Middle-level officials, such as prefects, and some chou and hsien magistrates obtained the greater portion of their expenses from surplus commercial duties above the small quotas due the central government.^[68]

The emperor was not pleased with this system, perhaps because it amounted to little more than the legalization of existing patterns of funding under the informal network. In early YC 3 (1725) he issued an edict instructing the officials in both Yunnan and Kweichow to account for all revenues available from huo-hao , surplus duties, mining, and rents on imperial lands in preparation for their

remittance in full to the provincial coffers. Once the total resources available in each province were known, their high officials were to allocate yang-lien for every post from governor to magistrate, calculated on the basis of remoteness from the provincial capital and the level of official business involved.^[69]

We do not know how these funds were ultimately allocated in Yunnan, but we do have remarkably detailed accounts of how Kweichow's income was derived and the manner in which it was redistributed after the reforms. Table 4.6 shows that of the 60,894 taels returned to the provincial coffers in Kweichow, only 18 percent, or 10,792 taels, were derived directly from huo-hao on the land and head tax. Even if we include the income from surcharges on levies of rice and beans, the total revenues from traditional agricultural sources were only 33 percent of Kweichow's provincial funds under the new system.

The fact that most of Kweichow's provincial revenues were obtained from commercial sources and a grant of rents from imperial lands argues strongly against claims that the huo-hao kuei-kung reforms were a central-government ploy to extract an even greater portion of the extra-quota income collected by the provinces. By approving the accounts submitted by governors such as Shih-li-ha, the emperor was actually confirming the right of provinces to a

portion of the customs and salt revenues originating within their borders. It is clear that the emperor's concern was not to commandeer additional income for the central government, but to ensure that each province had the means to carry out the rational management of local finances with the same success that was possible in areas with reliable, high land-tax receipts. This is particularly striking when we realize that 64 percent of Kweichow's estimated annual income from salt and commercial taxes was now to be retained by the province for its own use.

Even with the addition of funds from imperial-lands, salt, and miscellaneous duties, Kweichow's provincial revenues were barely enough to meet all its fixed costs. (See table 4.8.) A comparison with yang-lien statistics for other provinces shows that an official in Kweichow received several hundred taels less than that of an average official of the same rank in other, wealthier provinces. (See table 4.9.) Moreover, only 1,190 taels remained to meet the elastic public expenses of the province, such as yamen and road repairs and water-conservancy projects.

Events of the next few years placed intolerable strains on Kweichow's already tight finances. Jurisdictional reorganization made Yung-ning hsien a part of Szechuan province, resulting in a loss of over 3,600 taels in duty surpluses. In addition, Tsun-i prefecture was transferred to Szechuan and five other hsien were made a part of Hunan province. At the same time, several new prefectures and chou were created in areas previously under aborigine rule, increasing the province's burden of yang-lien and public expenses.^[70] A further budgeting problem resulted from the fact that so much of Kweichow's income depended on surplus revenues from the salt and customs administration. Both of these sources were highly variable, depending on the volume of trade during a given year. The estimates of salt and duty surpluses used by Shih-li-ha in drawing up the Kweichow budget were those of the year before the reforms were implemented. Income in later years did not always reach the levels of the first few years of the Yung-cheng reign. During the intervening period it was often necessary to reduce the yang-lien of all officials by as much as 10 percent to make up for unforeseeable losses in revenues.^[71] As a result of all these circumstances, Kweichow province was faced with a shortage of approximately 20,000 taels by YC 8 (1730).

In order to solve this problem and provide yang-lien for newly created posts and for miscellaneous officials, Governor Chang Kuang-ssu devised a novel plan that would enable the province to profit from its rich mining resources, as Yunnan had done earlier in the dynasty.^[72] Within its borders Kweichow had mines at Nanlung fu and Wei-ning chou that produced a form of lead known as wo-ch'ien . Because Yunnan had opened a mint and needed lead for its smelting factory, these Kweichow mines had been placed under its jurisdiction. By 1730, the Lo-p'ing mines in Yunnan were producing sufficient lead for Yunnan's needs and the lead that had earlier been sent from Kweichow was lying idle. To remedy this, the officials in Yunnan had memorialized to borrow funds from their provincial treasury to buy up the surplus lead and ship it to Hangchow for sale. For every 100 chin of lead shipped north, they netted one tael, after deducting the purchase price and transportation costs. The profits were deposited in Yunnan's public-expense fund.

According to Governor Chang, Kweichow was now undertaking to establish its own smelting operation and requested that all produce from the mines within its territory be placed once again

under Kweichow's control. The province would use the lead to supply its own smelting facilities and dispose of any surplus after central-government production quotas were filled in the same manner as Yunnan. He estimated that Kweichow would be able to purchase the entire surplus from the lead mines, 4 to 5 million chin annually. In addition, 700,000 chin would be collected by the province in the form of taxes. Because the smelting factories would be able to handle only about 160,000 chin when they first opened, the remaining lead from taxes could also be sold. Calculating a net profit of 1.5 taels per 100 chin , Kweichow could obtain between 50,000 and 60,000 taels in additional income per year. Governor Chang requested the emperor's permission to keep this sum to make up the shortages in the current provincial budget and to provide 28,200 taels to increase the yang-lien granted to each official and allocate yang-lien for miscellaneous minor officials and newly created posts along the Miao border. This would leave as much as 11,800 taels for those cost-elastic projects previously neglected under the administration of Shih-li-ha.

Only one obstacle stood in the way of this plan. Whereas Yunnan had advanced the initial capital for its own merchandising experiment from surplus funds in the provincial treasury, Kweichow's treasury was already suffering a deficit. Therefore, Chang also requested to borrow from the 100,000 taels set aside by the central government for land-reclamation projects within Kweichow's border regions. Originally these funds had been designated to provide working capital for Han Chinese opening new land to cultivation. However, Chang claimed that most of the land was actually being reclaimed by Miao who were supplying their own tools and labor and had not requested public assistance. Chang proposed to borrow 60,000 taels from this fund. This could then be repaid over a period of three years out of the profits from the sale and shipment of lead. The emperor's only comment on Chang's memorial was to question whether educational officials would be provided with yang-lien , hence it may be assumed that this plan was indeed carried out.

The use of government entrepreneurship to increase provincial revenues in Yunnan and Kweichow was an extremely important step in the development of Chinese fiscal policy. Although the projects undertaken entailed merchandising endeavors and not manufacturing per se , they may have set a precedent for the provincial enterprises established during the nineteenth century. Although

government monopolies were as old as the imperial system itself, this was the first time that provinces engaged in profit-making projects for the sole purpose of providing funds for their own provincial coffers. The existence of surplus revenues in the Yunnan treasury with which the province could initiate its own leadmerchandising scheme indicates that the reforms in that province, aided by income from copper mining and salt production, were a success. With Kweichow's own leap into the arena of provincial entrepreneurship, that tax-poor province, too, was able to join the agriculturally rich provinces to the north in providing a steady and secure source of income for its officials and to carry out public works.

The absorption of surplus salt, customs, and miscellaneous commercial tax revenues into the budgets of these border provinces also demonstrates that what mattered most in the implementation of the huo-hao kuei-kung reforms was the changes that they made possible in the operation of local-government finance. Huo-hao as the vehicle for these changes was only secondary. Moreover, because most of these funds belonged by statute to the central government, it is clear that the emperor's commitment to fiscal rationalization outweighed even the desire to augment central-treasury reserves.

Commercial duties played an important part in the funding of provincial government in the provinces along China's eastern seaboard as well. However, here in the economic heartland of China, even more overt grants of aid from the central government were necessary to offset the obstacles presented by tax evasion and the heavy demands placed on provincial resources.

Poverty Amidst Riches: Huo-Hao Kuei-Kung in Shantung and the Southeast

For the first time in imperial history, the fiscal reforms of the Yung-cheng period addressed the problem of inadequate funding of local government. As such, they provided the Chinese bureaucracy with a powerful weapon against corruption within its own ranks. However, huo-hao kuei-kung could not alter the structure of local society. In those provinces where the influence of the local elite was strong, the proponents of reform met their most intractable adversaries. Rationalization of local-government finances threatened to break the control that members of the gentry exer-

cised over magistrates by releasing them from their dependence on tax farmers for revenues beyond the legal quota. In addition, huo-hao kuei-kung meant the equalization of surtaxes, ending the lower rates enjoyed by the privileged strata of the rural population.^[73] In many parts of China, yamen clerks, local gentry, and village strongmen waged a quiet but unrelenting war against fiscal rationalization in the form of slander campaigns and, even more damaging, outright resistance to the payment of taxes.

Shantung province provides an excellent example of the problems confronted by officials in an environment where their authority over the tax-collection process was incomplete. Although it was a north-China province with a large tax base, Shantung was not able to emulate the Honan model of reform until the last years of the Yung-cheng reign. The policy first instituted in Shantung set the rate of huo-hao at 18 percent. On a regular tax quota of approximately 3 million taels, this yielded an annual income of about 540,000 taels. Of this, 200,000 taels were set aside to make up deficits, another 200,000 taels were used to provide officials with yang-lien , and the remaining 140,000 taels were appropriated for the public expenses of the whole province.^[74] On the surface, at least, Shantung seemed well on its way to matching the performance of Honan, Shansi, and Shensi in the effective implementation of reform.

However, the policy of remittance established in Shantung set it apart from its fellow provinces to the west. The chou and hsien deducted their own yang-lien before sending their huo-hao to the provincial capital. Higher officials received their yang-lien directly from the financial commissioner, but this was almost the only revenue he controlled. Of the 140,000 taels designated for provincial public expenses, the chou and hsien were also permitted to retain 120,000 taels to cover their own kung-fei and the costs of remitting their taxes to the provincial treasury.^[75] Consequently, only 20,000 taels were left for construction projects and famine relief administered for the province as a whole.

This policy undermined the efficacy of the reforms in several ways. Just as T'ien Wen-ching predicted, magistrates, when left to deduct a portion of their huo-hao revenues themselves, tended to persist in the collection of illegal surcharges (ssu-p'ai ). Some magistrates were so determined to acquire this additional income that the people were forced to pay the illegal fees before remitting their regular taxes. When the burden became too great and the regular

quota went unfilled, the magistrates simply reported a high level of people's arrears (min-ch'ien ).^[76] Because huo-hao was remitted along with regular quota, deficits in the cheng-hsiang also meant deficits in the huo-hao revenues sent to the provincial treasury. According to the original plan, it was anticipated that Shantung would be able to repay all her previous deficits by YC 4 (1726).^[77] By YC 6 (1728), however, the combination of old deficits that had not yet been repaid and new deficits incurred since the reforms went into effect brought Shantung's total outstanding tax payments to an astonishing 2 million taels.^[78]

Shantung's total tax quota was among the highest in the empire. Yet despite the ample revenues that an 18-percent huo-hao should have yielded, the problems encountered in collecting these taxes forced Shantung's officials to turn to additional sources of funds beyond the land and head tax. Unlike the other provinces in north China where an added incentive to engage in corruption was eliminated by banning most customary fees and gifts, Shantung province integrated many of these former informal levies into her revised fiscal system. A continual debate was carried on among provincial officials over which of these sources were legitimate and which should be eliminated. As a result, not only is it impossible for us to determine the annual income from these alternative sources, but even officials at the time were unable to predict their revenues for any given year.^[79]

In YC 6 (1728), the emperor appointed Honan Governor T'ien Wen-ching governor-general of Honan-Shantung, a post created especially for him. T'ien's most important task in this capacity was to use the techniques developed in Honan to clean up the administration of his new jurisdiction.

T'ien Wen-ching attributed Shantung's fiscal difficulties directly to the strength of the local elite of the province. Unlike Honan, Shantung was plagued by vestiges of the outlawed li-chia system, under the guise of which influential members of the local community, often with the aid of hired thugs, collected the land and head tax at several times the legal rate.^[80] In some parts of the province, magistrates had completely abdicated their responsibility for tax collection in return for a guaranteed share of the booty wrenched by these petty tax farmers from the common people. Although this corrupt alliance was a product of the days when the informal funding system was the only means available for financing local administration, the networks of power it supported in the country-

side made both magistrates and their extrabureaucratic cohorts almost impossible to control.

True to the principles he espoused in Honan, T'ien first moved to institute complete remittance of huo-hao to the public coffers as a check on the manipulation of tax income. Impartial inspectors were appointed to examine the locked chests set up to receive the payments of landowners and relieve them from dependence on tax farmers. At the same time, strict measures were implemented to prevent high officials from encouraging low-level graft by demanding gifts and fees from their subordinates.^[81] Nevertheless, gentry resistance to the new policies remained formidable and, as in Honan, was often aimed directly at the person of the governorgeneral. When the emperor ordered T'ien to visit Shantung in 1729 to investigate a case of corruption, his detractors leaped at the opportunity to discredit him in the eyes of the public he sought to release from their control. His admonitions to the local bureaucracy to clean up fiscal malpractices and carry out reform were met with rumors that T'ien had come on a mission to raise people's taxes and had even issued a new bamboo measuring rod inches shorter than the one normally used as a standard in assessing the size of the people's land holdings.^[82]

Efforts to disparage huo-hao kuei-kung by linking it to augmentation of the peasants' tax burden was a common ploy during the early years of the reforms. However, the persistence of T'ien's adversaries was sufficient to cause him to postpone complete remittance of huo-hao to the provincial coffers. Unable to duplicate his performance in Honan, the governor-general bad to be content with prohibiting the extra fees collected by magistrates and lowering the rate at which huo-hao was levied to 16 percent.^[83] Moreover, instead of allowing the chou and hsien to retain a percentage of their huo-hao as public expenses, Governor Fei Chin-wu ordered that the amount of kung-fei deducted by each magistrate be calculated according to need and that transportation expenses (chieh-fei ) be allocated on the basis of an area's remoteness from the provincial capital. In mid-1730, Financial Commissioner Sun Kuo-erh finally moved to complete the process by requesting that huo-hao be remitted to the provincial treasury in full and that the magistrates receive their yang-lien as a grant from the financial commissioner.^[84]

Although the refinement of huo-hao kuei-kung in Shantung had required almost seven years to achieve, officials in that province

had begun as early as the first year of the Yung-cheng reign to confront the problem of corruption in their midst. In the provinces of Kiangnan and the southeast coast, officials were much slower to respond to the emperor's invitation to reform.

Kiangsu, with the highest land- and head-tax quota in the empire, had the unique distinction of being the only province expressly denied the emperor's permission to carry out huo-hao kuei-kung . It was not until YC 4 (1726) that the governor thought to broach the subject. Even then, his main concern was to find a source of revenue to reimburse provincial deficits. However, it was not the high level of deficits in Kiangsu that made Yung-cheng reluctant to permit its experimentation with his reforms. Those deficits had reached 2 million taels by 1726,^[85] but were no worse than the deficits already recorded in Shantung. More important were the province's 8,810,000 taels in arrears, a figure in excess of the combined arrears of all the other provinces in the empire.^[86] Although the emperor had sanctioned the repayment of these outstanding taxes in manageable installments, magistrates had proven themselves incapable of retrieving more than 100,000 taels per year.^[87]

The remarkable inability of Kiangsu's officialdom to meet its tax quotas was a reflection of fiscal corruption and disorganization that went far beyond that uncovered in Shantung. Early in his reign, Yung-cheng had begun to receive reports of irregularities in Kiangsu's tax rolls, the widespread contracting of tax payments, and various practices utilized by clerks and members of the gentry to disguise landownership and redistribute tax liabilities to their own profit.^[88] Until these malpractices could be investigated and brought to a halt, any efforts at reform of local funding would be in vain. Chronic arrears would preclude the collection and remittance of adequate supplies of huo-hao . Furthermore, without regularization of the methods of taxation it would be impossible even to tell how much of the enormous arrears in the province was really the result of nonpayment by the people and how much was already lining the pockets of officials and their staff. If the latter could circumvent payment of over 8 million taels in taxes owed the central government, far better controls would be necessary to guarantee remittance of huo-hao quotas to the provincial coffers.

In order to clear the way for reform here and elsewhere in Kiangnan, Yung-cheng first decreed a thorough investigation of the region's arrears and tax-collection procedures. (See chapter 6.)

Only when this investigation was well under way was permission granted to undertake huo-hao kuei-kung .

The task of implementation was aided by the appointment of Kiangsu's financial commissioner as governor. Ch'en Shih-hsia had previously served in Honan and Chihli and had had considerable experience with the reforms. After a brief dispute with his replacement as financial commissioner over the method of remittance to be used, Ch'en's plan for complete remittance to the provincial treasury was promulgated in 1728.^[89] A 10-percent rate of huo-hao was levied on all land and head taxes and on several smaller levies traditionally sent to the grain, river, post, and salt intendents.^[90] An initial schedule of yang-lien quotas was submitted to the emperor by Ch'en Shih-hsia just prior to his retirement from his post. The emperor's criticism of these rates was integrated into the final schedule outlined in table 4.10. Except for the yang-lien of the financial and judicial commissioners, which the emperor judged to be too high, the salaries of almost every official were revised upwards in the final list. Practices in other provinces were also influential in determining the manner in which yang-lien was ultimately distributed. For example, originally no provision had been made for the fourteen subprefects and assistant subprefects in charge of river conservancy. The governor later decided to follow the precedent set in Honan and Shantung where yang-lien was granted to these low-level functionaries in order to free them from the necessity to overreport construction costs on waterworks as a means of acquiring funds for their yamen and travel expenses.^[91] In addition, the yang-lien of Kiangsu's chou and hsien magistrates was raised to match that of the better-funded provinces. A fund for local public expenses called pei-kung yin was also established, from which each chou and hsien was allocated an additional 200 to 600 taels annually.

Kiangsu's total tax quota at the time the reforms were introduced was 3,719,000 taels. Of this sum, 2,271,000 taels were remitted directly to the financial commissioner, yielding 272,100 taels of huo-hao . Another 997,800 taels were sent to the grain, river, post, and salt intendants. They were allowed to retain 30 percent of the huo-hao derived from these taxes to make up the silver content of the taxes they sent to the central government. The remaining 69,800 taels were also forwarded to the financial commissioner. Kiangsu thus had a total of 341,900 taels with which to pay all yang-lien and public expenses.

Despite its large tax quota, Kiangsu had less income available for provincial expenses than did provinces with comparable quotas in the north. Table 4.11 shows the fixed expenses of the province in a typical year. Over 80 percent of these funds were spent in fulfilling obligations to the central government or to subsidize projects paid for out of cheng-hsiang . After allocations were made for fixed expenses, Kiangsu had only 36,000 taels for the elastic expenses of the province itself. This could hardly cover the costs of city-wall, road, embankment, and garrison repairs encountered by provincial officials each year. Moreover, this sum might not be available when the need arose. Even after the implementation of the reforms, tax arrears continued to cost the province an average of 70,000 taels in huo-hao per year.

In order to provide a reliable fund for provincial expenses, Governor Yin-chi-shan requested in 1731 that customary fees remitted by Liang-huai salt merchants to the salt controller be incorporated into the provincial budget. These so-called yen-kuei yin-fei totaled over 100,000 taels annually. With them, Kiangsu was able to guarantee the availability of funds for repairs and also joined the prov-

inces in the north in providing most minor officials with yang-lien .^[92]

Official confusion and lack of initiative could be just as important a deterrent to reform as tax arrears and fiscal corruption. Under the informal funding network, high officials in Kiangsi were the beneficiaries of a source of income not affected by Yungcheng's proscription against the acceptance of customary fees.^[93] Along with the regular tax, each hsien in the province remitted an additional 0.0175 tael per tael, known as p'ing-fei . On an average annual tax quota of 1,500,000 taels, this yielded over 26,000 taels, which was distributed among the governor, governor-general, and financial commissioner, and used for both yamen and public expenses. Magistrates were allowed to collect huo-hao to cover their own expenses, leaving little incentive for high or low officials to lobby for its return to the public coffers.

When a plan for remittance of the meltage fee was finally devised for Kiangsi in YC 7 (1729), it contained the most unwieldy system of yang-lien allocation to be proposed by any province. Those hsien with a total regular tax quota of 9,000 taels or less

were not required to remit any of their huo-hao . Those with between 10,000 and 60,000 taels were to remit 61 percent. Those with even higher tax quotas were to remit as much as 70 percent of their huo-hao , keeping the rest as yang-lien . The yang-lien of higher officials was calculated as a percentage of the huo-hao collected by the magistrates within their jurisdiction, deducting an amount equivalent to any salt fees each might receive.^[94]

The emperor's reaction to this scheme was not enthusiastic, and he immediately suggested that it be reconsidered to make yang-lien quotas more closely reflect the real fiscal needs of the local administrative units. What he received in reply was almost comical. Instead of determining quotas for groups of hsien on the basis of their remoteness from the capital and the complexity of their public business, the financial commissioner simply revised the percentages of huo-hao to be retained, calculated the amount they would yield in each place and converted that amount into a fixed quota. The yang-lien figures obtained in this manner were so diverse that it clearly would be impossible to enforce provincial supervision of county accounts.^[95] The yang-lien of higher officials followed a similar pattern. (See table 4.12.) Once again the emperor admon-

ished Kiangsi's high officials to compile a schedule of salaries that would both be equitable and discourage embezzlement and fraud. The result of these new deliberations was a system of complete remittance of huo-hao , collected at a rate of 10 percent of the regular tax. Most magistrates, and many prefectural officials as well, enjoyed an increase in yang-lien under the new plan. Twenty-eight military officials were granted yang-lien for the first time. And all

of this was accomplished at a saving to the province of forty taels.^[96]

The emperor's influence was also an important factor in the outcome of the reforms in Anhui. Here, complete remittance at a rate of 10 percent was proposed from the start. However, in Anhui it was not yang-lien quotas but the allocation of kung-fei that caused confusion among provincial officials. Governor Wei T'ing-chen's provisional budget appears in table 4.13. Those items marked with an asterisk were rejected by the emperor as unwarranted. If we compare the items that the emperor questioned with those he left standing, we can see that he approved funding for all administrative expenses, including stationery costs, clerk and runner

wages, and so on. What he did not sanction was thinly veiled compensation for revenues which previously would have been obtained as customary gifts and fees. Among these were gifts received from subordinates when a high official was engaged in supervision of government projects, and gifts in kind, such as annual contributions of brocades to the governor's yamen. Moreover, the cost of entertainment and travel was considered part of the expenses covered by an official's yang-lien . Additional compensation for what amounted to fulfillment of one's duties was not compatible with Yung-cheng's conception of the huo-hao kuei-kung reforms.

Nor did good intentions and an understanding of the reform process guarantee the successful implementation of huo-hao kuei-kung . In Chekiang it was excessive zeal that created difficulties in the management of local finances. Chekiang's huo-hao levies had traditionally ranged between 5 and 8 percent. These low rates were due in part to the many different fees from the salt administration and commercial duties that were available to the province's bureaucrats.^[97] However, when Huang Shu-lin came to office as governor in early YC 2 (1724), he not only followed the emperor's instructions and banned the collection of customary fees but also lowered the rate of huo-hao to a uniform 5 Percent.^[98] Subsequent officials found the income from huo-hao inadequate for their needs, a predicament aggravated by the cumulative arrears that prevailed throughout the region. With only 130,000 taels from which to allocate yang-lien and kung-fei , Huang's successors begged for special permission to continue the arrogation of feng-kung contributions banned by Yung-cheng the previous year. Unable to persuade the emperor of the justice of their cause, the governor and financial commissioner resigned themselves to implementation of huo-hao kuei-kung on a very limited basis. Chou and hsien officials continued to collect huo-hao at a rate of 5 percent. Three percent was retained for their own use and the remaining 2 percent was remitted to the financial commissioner's treasury for projects of benefit to the whole province.^[99] No attempt was made to provide yang-lien for higher officials, who continued to receive customary fees from salt merchants.^[100]

It was not until Li Wei, one of Yung-cheng's three most trusted officials, was appointed Chekiang governor in 1727 that order was brought to the implementation of the reforms in that province. However, the full credit cannot go to Li Wei. Even with the administrative skill and integrity that the emperor clearly felt he would

bring to his post, Li could not create funds where none existed. As his predecessor, Shih Wen-cho, had once remarked, trying to use huo-hao to manage all of Chekiang's expenses was like trying to cook a meal without any rice (wu-mi chih-ch'iu ).^[101] Rather, it was the emperor himself who brought to Chekiang the opportunity for rational fiscal administration.

The instrument of the emperor's largesse was an edict issued in the winter of YC 5 (1727). In it he bestowed two favors upon the people of Chekiang. In order to relieve the burden of taxes, an important factor in the generation of arrears, Yung-cheng reduced the so-called excessive tax quotas (fu-liang ) of Chia-hsing and Huchou prefectures.^[102] This, of course, lowered even further the basic tax upon which Chekiang's already meager huo-hao was levied. Therefore, the emperor also granted Chekiang the extraordinary privilege of keeping for its own use 5 percent of the cheng-hsiang quota usually remitted to the central government by the nine prefectures not affected by the tax reduction. In effect, Chekiang's huo-hao was raised to 10 percent without any additional imposition on the people. An extra 100,000 taels was made available, with only one condition for its use—that it be part of a program of complete remittance and contribute to the equitable determination of yang-lien to all officials in the province.^[103]

Yung-cheng's gift to Chekiang enabled the province to eliminate customary fees and allocate yang-lien on the basis of need. Moreover, for the first time, the province was endowed with enough funds to repay its deficits, subsidize repairs to warships, and defray the costs of disaster relief, water conservancy, and other emergencies that might arise.^[104]

Indeed, it was only by means of imperial gifts and the absorption of revenues not generated from the land and head tax that many provinces in the southeast were able to cope with shortages in the funds available for public expenses. Kiangsi's lower cheng-hsiang quota yielded far less huo-hao than was available in Kiangsu or Chekiang. However, a smaller tax quota did not signal a smaller bureaucracy, and Kiangsi's officials required almost as much for yang-lien as the other two. In addition, more than half of Kiangsi's public-expense funds were drained out of the province in the form of Board fees and costs incurred in transporting tribute and taxes to the imperial capital. (See table 4.14.) Huo-hao alone could not provide for all of the province's needs. Fortunately, officials in Kiangsi had access to various commercial sources, which

substantially increased their discretionary income. Several tens of thousands of taels were accumulated each year from salt fees and surplus revenues from the Chiu-chiang and Kan-chou customs.^[105] The province was also able to set aside a few thousand taels each year from the sum the central government allocated to cover the costs of shipping taxes to the capital or as assistance to other provinces.^[106] Concern that Kiangsi might still suffer from want of funds

for public expenses and yang-lien for minor officials prompted the Board of Revenue to consider a special grant of regular tax funds in this province as well. Therefore, in YC 8 (1730), Kiangsi was permitted to retain part of its deed tax (t'ien-fang ch'i-shui ) and surplus income from the unloading tax (lo-ti shui ).^[107] The Board's generosity did not end there. In the following year, a surplus appeared in the province's annual report of huo-hao income and expenditure, submitted to the emperor in a secret palace memorial.^[108] Most of the unspent funds were attributed to huo-hao saved from "food money" (hao-hsien fan-shih chih-sheng yin ). Inasmuch as such fees generally designated extra payments made to cover the Board's administrative costs, it is likely that a substantial portion of Kiangsi's obligation to this agency of the central government was waived in the interest of local fiscal stability. By YC 11 (1733) this had meant a cumulative saving to the province of approximately 112,800 taels.^[109]

Officials in Fukien were not able to circumvent their fiscal difficulties with comparable ease. Fukien's cheng-hsiang quota was the lowest in the southeast. Even at a rate of 14 percent, huo-hao produced only as much income as some provinces spent on yang-lien alone.^[110] After a brief flirtation with a system of little remittance, the emperor's obvious disapproval impelled Fukien's officials to establish quotas for yang-lien , but of necessity they fell far below the standards enjoyed elsewhere in the region.

Fukien's problems were compounded by its own obligations to the central government. These were in the form not of Board fees but of expenditures for coastal defense. Fukien was a major outpost for China's maritime forces and bore a large share of the responsibility for military provisions and the construction and repair of warships.^[111] So critical to the empire's security were these charges against Fukien's treasury stores that it was the only province compelled to vary its allocations of yang-lien year by year, according to what remained after the allocation of kung-fei .

Once again, the intervention of the central government was crucial in rescuing the province from financial disaster. In 1729 an annual grant of 11,320 taels in surplus commercial duties and bridge tolls was recommended by the Board of Revenue. In addition, Fukien was allowed to retain 30,739 taels in revenues from imperial lands in Taiwan.^[112] These funds enabled the province to improve its yang-lien quotas (see table 4.15) and provide salaries for minor officials below the rank of magistrate. More important were

the Board's efforts to release Fukien public-expense funds for the province's own use. Convinced that the construction and repair of warships and military barracks and other defense-related projects was too great a drain on Fukien's meager resources, the Board instructed the governor and governor-general to submit a routine memorial on all such expenditures so that they might be reimbursed from cheng-hsiang .^[113]

Even this degree of Board magnanimity was insufficient to relieve Fukien of all its fiscal woes. Like Kiangnan, it had a long history of politically as well as economically motivated tax evasion. Few comments on Fukien by its nonnative administrators lacked references to the province's scheming and unruly population, a reflection as much of current tensions as of Fukien's historic role as the last bastion of armed resistance to the Manchu regime. Although the impact of arrears on huo-hao revenues was often a matter of acrimonious debate among Fukien's officials, it was generally recognized that approximately 30 percent of the province's land- and head-tax quota was outstanding every year.^[114] Some of the deficiency this created was offset by the repayment of past arrears, but even the most conservative estimates assumed an irretrievable loss to the province of almost 20,000 taels of huo-hao annually. In mid-1730 the emperor discovered that his grant to Chekiang exceeded its needs by precisely this amount, and an annual subsidy of 20,000 taels per year was transferred from Chekiang to Fukien.^[115]

The efforts of the Yung-cheng emperor to guarantee the fiscal health of each province and the successful implementation of the huo-hao kuei-kung reforms even extended to the allocation of direct cash grants of central-government revenues to areas judged to have special funding needs. The precedent for such gifts was set early in his reign, when Wan-p'ing and Ta-hsing hsien in Chihli were presented with 10,000 taels each in recognition of the additional responsibilities entailed by their proximity to the metropolitan capital. In YC 8 (1730) the emperor broadened the scope of these grants to provide one-time allocations of between 200,000 and 300,000 taels to regions throughout the empire having unusually high costs due to military expenditure.^[116] Although Kiangsu, Kiangsi, and Anhui were not militarily sensitive areas, their importance from an economic and political point of view induced Yung-cheng to authorize grants of 100,000 taels to each to provide for emergencies and guard against the delays that might occur

if money had to be requisitioned from the central government in times of urgent need.^[117]

By the end of the Yung-cheng reign, the legitimation of a single surcharge on all land and head taxes had evolved into a complex program, adapted to the unique fiscal requirements of each province in a large and varied imperial state. Customary fees had been brought under control, and commercial and industrial revenues had been openly enlisted into the service of local government. Even a portion of the central government's own tax quota had been diverted to provincial coffers in the interests of fiscal reform. Problems still remained, and in the ensuing decades their resolution would more than once test the viability of the new fiscal order. Nevertheless, when the third Ch'ing emperor died, a system of rational local finance could stand as one of his greatest bequests to his successor and to his subjects.

5—
Local Variations and Underlying Principles

Together, the development of the palace-memorial system and the operations of local finance under the huo-hao kuei-kung reforms reflect an important transformation in the structure of decision making during the early Ch'ing. Both were manifestations of the Yung-cheng emperor's efforts to manipulate the lines of official communication in order to by-pass the organs of central bureaucratic decision making and establish a direct link between the emperor and his representatives in the provinces. The purposes of this realignment of power were twofold. On the one hand, it allowed the emperor far more initiative and control than had been enjoyed by any previous Ch'ing emperor. More important, however, was the creativity that this change in the power structure permitted in the reform of administration. Without Yung-cheng's strengthening of the palace-memorial system, from an occasional avenue of information on crop and weather conditions to an institutionalized link between emperor and official for the reporting of all aspects of local affairs, the huo-hao kuei-kung reforms could never have taken place.

Information transmitted in secret palace memorials first alerted Yung-cheng to the real fiscal instability of local administrative units. It was in his replies to palace memorials that the emperor was able to orchestrate personally the direction of reform. Despite

court opposition to the ideas embodied in No-min's and Kao Ch'eng-ling's memorials, the palace-memorial system made it possible for the reforms to spread throughout the empire without the interference and the potential obstruction of central-government bureaucrats, whose administrative philosophy was far more conservative than that of the emperor or his trusted officials in the field. The variations in local adaptation of the reforms and the ambiguities in the system of reporting and accountability utilized must be viewed in this light.

The Yung-cheng emperor was concerned that provincial autonomy in managing huo-hao funds be maintained in order to ensure the flexibility necessary for effective administration of local government. For this reason he was determined to prevent the Board of Revenue from imposing the item-by-item supervision over huo-hao accounts that it exercised in auditing regular taxes. However, the emperor's instigation of changes in certain provincial reform proposals shows that he was not content to leave implementation entirely up to the provinces themselves. Why, then, did the emperor not simply issue an edict declaring a uniform pattern for reform in all provinces? First, Yung-cheng firmly believed in the principle of "suiting one's policies to the time and place."^[1] To assume that the specific mechanisms for provincial funding appropriate to a province with high land-tax revenues and relatively little commercial income, such as Shansi, would be effective in a place where agricultural production was limited and resources from mining played a major role in taxation, such as Yunnan, would have been disastrous. The only way to ensure the success of a plan for legalized and rationalized local finance adapted to the particular needs of individual provinces was to allow the provinces to develop their own plans for reform. At the same time, by using the palace-memorial system to advise provincial leaders as they designed their reform proposals, the emperor could guarantee that despite local differences, each province would adhere to the basic goals of huo-hao kuei-kung .

We are now privy to many of the secret exchanges by which the emperor and the provincial bureaucracies worked out the formula that we call the return of the meltage fee to the public coffers. However, to officials at the time, particularly those in the central government, it is likely that the emperor's guiding role was far less apparent. Moreover, Yung-cheng himself was often deliberately noncommittal, leaving the day-to-day decisions in the hands of the

officials at the top of the provincial hierarchy. This strategy was designed not only to encourage provincial officials to take a greater role in local fiscal decision making, but to deflect from the emperor's own person the responsibility for any subsequent failure of the reforms. By appearing not to have a personal stake in the decisions made in connection with the implementation of the reforms, Yung-cheng could step in and initiate changes if they were necessary, without seeming to contradict his own orders. It was for this same reason that the original proposal for reform was introduced as a "grass roots" plan from the Shansi bureaucracy and not in the form of an imperial edict.

From our previous descriptions of the huo-hao kuei-kung reforms we can see that differences in economic resources, regular tax obligations and even the commitment of local officials to reform led to variations in the timing and structure of local fiscal reorganization. The emperor's own insistence on local initiative and planning in order to retain some of the flexibility inherent in the informal funding system might lead us to view the reforms as lacking a unifying set of principles, regulations, and goals. In fact, nothing could be farther from the truth. Local accommodations were made in the items of revenue included in local fiscal planning, and variations did exist in the proportions of huo-hao set aside for yang-lien and provincial expenditures. Implementation was delayed in a number of places and not every province adopted the same method of huo-hao remittance. An examination of the reforms as a whole, however, indicates a strong unifying conception embodied in principles and regulations which, though not always promulgated in the form of administrative statutes, were nevertheless known and followed by officials in every province. It is to these general guidelines and an assessment of the reforms that we now turn.

Public-Expense Funds

Public Expenses and the Central Government

One of the most important concepts implicit in the huo-hao kuei-kung reforms was that provincial and local administrative units should have their own regular sources of funds, provided from clearly defined tax sources and immune to exactions either from the central government or higher administrative units within

the province itself. Such funds came to be known as kung-hsiang . Items for which they were spent were usually called kung-yung or kung-shih . Kung-hsiang revenues allocated for kung-shih were known as kung-fei .^[2]

Provincial and local revenues were designated as kung-hsiang in part to distinguish them from the cheng-hsiang under the control of the central government. This distinction was important not only to avoid the much-feared conversion of provincial-expense revenues into regular taxes, but also to demarcate central and provincial responsibilities in financing projects on the local level. The huo-hao kuei-kung reforms were not, as it is sometimes assumed, devised to relieve the central government of those fiscal responsibilities that it had previously assumed in the provinces. Examination of the scattered evidence of cheng-hsiang allocation in the provinces shows that, to the contrary, the share of local expenses borne by the central government probably increased during this period, as part of a comprehensive drive to eliminate deficits by providing mechanisms to guarantee local solvency.

Nor did the existence of a regular source of provincial revenue lead the central government to abdicate its primary responsibility for the more costly outlays connected with military security. To the contrary, the demise of the informal funding system stimulated a greater consciousness of the need to prevent such expenditures from burdening the common people. A dispute that arose in Kansu between Shen-kan Governor-general Ch'a-lang-a and Kansu Governor Hsü Jung illustrates this point. In order to meet urgent demands for military supplies in his province, Hsfi Jung had ordered the accelerated collection of taxes in arrears. Resistance by the province's taxpayers prompted Ch'a-lang-a to file a complaint with the Board of Revenue against his overly zealous subordinate. In his recommendations to the emperor on this matter, Board President Prince I condemned Hsü's actions and set forth the government's position on funding for military campaigns. "If there is a shortage of funds for military expenses and provisions, the appropriate measure is to request, in a t'i-pen , that the central government allocate additional funds."^[3]

The 1720s and 1730s were a period of heightened central-government awareness of its role in funding major relief and construction projects. Yung-cheng's condemnation of officials who solicited feng-kung contributions for disaster relief was in part a reflection of his commitment to the use of cheng-hsiang for this

purpose.^[4] Likewise, when reviewing Fukien's allocations of huo-hao , Prince I was disturbed at the extent to which that province was using its own furlds for expenditures that were rightly the responsibility of the central government. He therefore ordered the governor and governor-general to pay for expenses such as warship construction and the erection of military barracks out of cheng-hsiang and to report such allocations to the Board of Revenue in a routine memorial.^[5]

Soon after the Yung-cheng emperor's death, the question of the role of the central government in paying for public-works projects was raised again. The issue was not whether the government should pay for such projects in the provinces, but whether the practice of assessing contributions from the local population to pay for construction projects should be allowed at all. In a memorial to the new emperor, President of the Board of Works Lai-pao and other Board members pointed out that it was the usual practice for the Board of Works to allocate cheng-hsiang funds for projects such as river conservancy (ho-fang ) and the construction of dikes (t'i-an ), city walls (ch'eng-yuan ), floodgates, and embankments (cha-pa ). However, in Chihli, Shantung, Kiangnan, and Szechuan officials occasionally paid for these items out of special levies on the people's landholdings. A list of the projects for which such levies had been authorized was enclosed to assist the emperor in deciding whether this exception to the rules governing fiscal responsibility should be allowed in the future.^[6] In advising the emperor, the members of the Board clearly favored banning these so-called contributions and placing full responsibility for major allocations on the central government. Their arguments were truly in the spirit of the huo-hao kuei-kung reforms, for although they recognized that many such contributions were made voluntarily, out of a sense of public-mindedness, the Board feared that any fiscal policy based on irregular contributions would invite extortionary exactions from the local population by higher officials.^[7]

Concern that cheng-hsiang revenues be readily available to meet the central government's obligations in the provinces even led to changes in the method of storage and distribution of regular tax funds. Steps had already been taken in this direction during the late K'ang-hsi period, when "surplus" central-government revenues were placed in treasuries in Manchuria, as well as in those of each Manchu general-in-chief, garrison commander, and brigade general.^[8] The transfer of these funds out of the imperial capital to

depositories in the provinces was designed to facilitate funding in case of a military emergency. However, the later expansion of this policy to include civilian treasuries was a result of the government's desire to avoid delays in the allocation of cheng-hsiang revenues that might lead to a greater burdening of the common taxpayer. In YC 5 (1727) it was decided to distribute additional funds to the provinces of China proper, to be stored in provincial treasuries. The provinces were classified in three groups, each receiving between 100,000 and 200,000 taels. In CL 41 (1776) the quotas of stored revenues were increased.^[9] These funds were in addition to the quotas of retained taxes (ts'un-liu ) deducted by each province from the regular taxes remitted to the capital each year, and represented total deposits, not an annual deduction. Unlike ts'un-liu funds, these deposits were still the property of the central government, set aside for emergency expenses in the provinces. Deposit in the provincial treasuries was designed to avoid the costs and delays involved in transporting silver to the provinces when it was needed for an emergency.^[10] Expenditure was probably much more closely scrutinized than were funds allocated to the provinces directly, and appropriation without authority (shan-tung ) was punishable by beheading.

In addition to the funds stored in the provincial treasuries, the Yung-cheng emperor also initiated a policy of depositing central-government revenues in lower-level treasuries. At first no quota was set for the amounts to be stored, but in YC 8 (1730) these deposits were regularized. The amounts to be stored in prefectural, chou, and hsien treasuries were divided into four grades, ranging from 100,000 to 300,000 taels.^[11] Central-government funds distributed among local administrative units in this manner were known as fen-ch'u chih-k'uan .^[12]

The central government's concern for the efficient allocation of cheng-hsiang revenues was an outgrowth of the attention focused on local finance during the huo-hao kuei-kung reforms. The principle that public projects should be paid for out of public funds, duly collected and budgeted for such purposes, forced the central government to reevaluate its own proper contribution to the administration of the provinces. Rather than encouraging the central government to encroach on local funds, the reforms seem to have served to highlight the respective responsibilities of each level of the government. The funds designated as kung-hsiang were meant to be used for quite different purposes than cheng-hsiang and the

methods of controlling, reporting, and allocating these funds were indicative of their unique nature.

Relieving the Burdens on the People

One of the main purposes of the huo-hao kuei-kung reforms was to relieve the local population of the burden of indiscriminate surcharges previously levied on the pretext of supporting local administration. By simply providing officials, runners, and clerks with adequate living expenses, it was assumed that the motivation for much of the corruption associated with the tax-collection process would be eliminated. More specifically, the provision of yang-lien led to the abandonment of many fees that were not extortionary but had been crucial to the operation of the informal funding system.

One striking example concerns fees previously levied to support provincial education commissioners. These men were charged with the duty of supervising both the hsien- and provincial-level examinations in their provinces and with monitoring the education and behavior of degree-holders and government students. Educational commissioners were among the officials whose hidden expenses were high and whose frequent travel brought them into constant contact with the local administration. In most provinces it was the responsibility of the magistrates to provide for the education commissioner's expenses, as well as the costs of administering the examinations.^[13] This was usually accomplished by demanding fees from the inhabitants of the chou or hsien. In Shensi this fee was called k'ao-p'eng kung-yin and was collected from the people at the time that the examinations were administered in their jurisdiction. In Szechuan the education commissioner's costs were met by gifts of ceremonial presents (chih-i ), usually of three or four taels each, sent to the commissioner by students taking the examinations. With the institution of yang-lien for the education commissioner these fees were banned and at least one extra burden was lifted from the shoulders of the local population.

We have already noted the central government's commitment to the provision of funds for major public-works projects in the provinces. However, minor projects remained the responsibility of the provincial governments themselves. In the past, despite the statutory elimination of corvée labor, it was often necessary to call upon the local people to contribute both the labor and supplies for

these projects. The availability of kung-hsiang revenues enabled provincial officials to carry out on a cash basis most transactions involved in local construction and administration.

In Kwangtung province, for example, repairs on garrison buildings and watchtowers had previously been handled by apportioning the responsibilities for the supply of labor and materials among the local inhabitants. The governor-general was naturally worried that such levies would place inordinate demands on the people's resources. But he had more practical concerns as well. The equal division of labor services among the people was an inefficient way to carry out repairs. Corvée labor on military projects aroused tensions between the soldiers and the people, and the people's lack of specialized skills led to inferior workmanship and costly delays. Procurement of supplies by means of popular "contributions" also gave clerks and runners too great an opportunity for extortion. The governor-general therefore requested permission to follow the precedent already set in Shantung (and, in fact, in many other provinces as well) of allocating the funds for repairs from the kung-hsiang .^[14] The emperor's response to Governor-general Shi Yü-lin's request indicates that the use of huo-hao funds for public projects was widespread and implicit in the original intent of the reforms.^[15] "Public expense funds exist expressly for you to use in the province for this purpose. What reason would there be not to carry out a plan such as this one, of such benefit to the soldiers and to the people? Deliberate and carry it out."

An examination of the sample budgets submitted by the provinces to the emperor^[16] shows that funds were invariably set aside for repairs and minor construction projects such as that mentioned above. Thus, many of the surcharges to which the people were subjected under the guise of contributions for public works were eliminated. Moreover, by prohibiting the contribution of officials' and runners' salaries and wages for construction projects, the reforms eliminated one of the main pretexts for levying nonspecific surcharges as well.

The Allocation of Kung-fei

The term kung-fei , as we have seen, referred to all funds collected by the province that were not remitted to the central government as regular taxes or distributed to individual officials in the

form of yang-lien . The manner in which these funds were budgeted and disbursed varied from province to province. However, allocation can be divided into four general categories: allowances given to chou- and hsien-level officials to cover "outside" expenses, allowances granted to high officials to cover the cost of yamen operations, funds set aside in the provincial treasury for local- and provincial-level irregular and emergency expenses, and funds set aside to cover the fees and expenses incurred in sending taxes, goods, and reports to the central government.

Kung-fei Allowances to Chou and Hsien Officials

As we have seen, the portion of huo-hao revenues allocated to chou and hsien magistrates in the form of yang-lien was to be used for the official's living expenses and for the costs of operating his own yamen. Expenses incurred outside the yamen, most notably for famine relief and for construction projects, were not expected to be covered by the official's yang-lien .

In at least some provinces, a separate allowance was granted to the chou and hsien for the public expenses in the magistrate's jurisdiction. The separate allocation of kung-fei to chou and hsien magistrates in Shantung was instituted as early as YC 3 (1725). Of the total annual receipts of huo-hao in that province, 140,000 taels were budgeted for the public expenses of the whole province. However, in accordance with Shantung's policy of partial remittance of huo-hao , chou and hsien magistrates were allowed to retain 22 percent of their total huo-hao income for their own kung-fei and the costs of remitting taxes to the financial commissioner.^[17] This was beyond the yang-lien allocated to them by the financial commissioner. Thus, of the total funds set aside in that province for public expenses, 120,000 taels were retained by the chou and hsien themselves and only 20,000 taels were deposited in the provincial treasury for the expenses of the province as a whole.^[18]

The magnitude of chou and hsien public-expense allowances in Shantung was unusual. In general, kung-fei allocations to lower-level administrative units were small and seen as supplementing yang-lien and providing for emergencies. Beyond this, it would be the responsibility of the province to handle major expenditures. For example, when the governor of Kansu decided to increase the very low rate of yang-lien payments in that province, magistrates, who were receiving a uniform payment of 600 taels in yang-lien

per year, were given differentiated kung-fei allowances as well. These allowances were based on the complexity of public business in their jurisdictions and ranged from an additional 600 taels for the Lanchow magistrate to 360 taels for medium chou and hsien and 160 taels for out-of-the-way chou and hsien.^[19] It was also the practice to grant separate allowances of kung-fei to the chou and hsien in Shensi. We have no evidence of the size of these allocations. However, Shensi Governor Chang Pao stated that "chou and hsien officials are each given public expenses besides yang-lien , on the basis of the complexity or simplicity of [official business] in their locale."^[20] Although yang-lien and kung-fei were listed together in the provincial budget submitted to the emperor,^[21] the fact that they were viewed as separate items is underscored by the provision that officials holding concurrent posts be given only half the yang-lien of the second post but receive the full allotment of kung-fei for both posts.^[22] Inasmuch as yang-lien was used for an official's personal expenses and the internal operations of his yamen, he would not need as much yang-lien as two officials holding two different posts. However, the "outer" expenses of hsien would not decrease during a temporary vacancy in the magistrate's office.^[23]

The same motivation seems to have been at work in the decision to grant kung-fei to chou and hsien officials in Chihli province. Sometime after YC 7 (1729), a tutor in the Imperial Academy named Ts'ui Chi memorialized regarding the need to provide magistrates with special funds for emergency expenditures.^[24] Ts'ui pointed out that unexpected matters occasionally arose on the local level for which funds were needed immediately. If the chou and hsien had already collected the taxes in their areas, they normally would transfer these regular taxes for emergency use. If, however, there were no newly collected taxes to transfer, or the official feared impeachment for the transfer of funds, he would usually borrow the necessary funds from saltshops, pawnshops, or rich households, or exact a temporary levy from the people. These borrowed funds were called ling-pi . After the emergency had been resolved (and presumably the funds to meet the emergency had been allocated by the provincial government), the borrowed funds were returned. However, sometimes less than the original amount was returned or a long period of time was allowed to elapse before restitution was made. As a result, both the people and the merchants were burdened. Moreover, Ts'ui was convinced that "treacherous merchants and wealthy and powerful households" were using this situation to coerce officials into alliances for their own benefit.

Ts'ui sought to rectify this situation by calling on a precedent set by the emperor in YC 7- "Shifting funds" had led to chronic deficits in Ta-hsing hsien and Wang-p'ing hsien in Chihli's Shun-te prefecture. In order to relieve the fiscal pressures on these two hsien, both of which lay within the precincts of the imperial capital, Yung-cheng had granted 10,000 taels to each to be stored in the hsien treasury and used in the event of an emergency. The magistrates could report the allocation to the metropolitan prefect and issue the necessary funds without waiting for the latter's approval. After the emergency had passed, the funds would be returned to the hsien coffers so that a balance of 10,000 taels would always be on hand.^[25]

Although he recognized that other chou and hsien did not require the reserve of funds granted to Ta-hsing and Wan-p'ing, Ts'ui felt that the problem of shifting funds and borrowing from the people was a serious one throughout the empire. He therefore requested that the emperor order each governor and governor-general to rank his subordinate chou and hsien according to the complexity of government business and allocate to them a portion of the huo-hao , unloading tax, property-deed tax and so on that was remitted to the province for public expenses. Ts'ui suggested that the busiest areas, situated at the crossroads of major transportation routes, might receive 1,000 taels. The next busiest could be given between 500 and 800 taels. The simplest, according to Ts'ui, need not receive any emergency funds at all.

Because the emperor did not issue a rescript to Ts'ui's memorial, it is difficult to know how Yung-cheng reacted to this suggestion, or whether widespread publicity was given to Ts'ui's views. It is clear, however, that even after the reforms, officials at various levels of the government continued to be concerned that magistrates would not have enough funds to carry out public business without some supplement beyond their original yang-lien quotas.

In YC 12 (1734), Chihli Governor-general Li Wei addressed himself directly to the problem of shifting funds at the chou and hsien level. When the huo-hao kuei-kung reforms were originally implemented in Chihli, many chou and hsien had been granted a reduced level of yang-lien because it was assumed that they would continue to receive a portion of surplus revenues and customary fees (hsien-yü lou-kuei ) from the post stations within their territories. In the course of the reform movement in Chihli, these fees were also eliminated, resulting in shortages in the affected chou and hsien. In the YC 12 huo-hao report to the emperor, Governor-

general Li proposed yang-lien increases in those areas to bring their yang-lien into line with the rest of the province.^[26]

The governor-general did not stop there. Li Wei also pointed out that in Chihli many magistrates continued the practice of advancing funds for public business (yin-kung p'ei-to ). In order to pay for non-yamen expenses, magistrates were also forced occasionally to contribute funds from their yang-lien . To remedy this problem, Li requested that each magistrate be allowed to retain "silver for carrying out public business" (pan-kung yin-liang ) based on the complexity of public business associated with his post.^[27] By requesting a special fund for chou and hsien public expenses, Li once again affirmed the principle of the separation of yang-lien from "outside" expenses.

It is not clear whether other provinces made separate allocations of kung-fei directly to the chou and hsien. For example, the Shansi budget submitted by Kao Ch'eng-lin in YC 3 (1725) indicates that 10,158 taels were set aside for miscellaneous construction and other costs incurred by the chou and hsien.^[28] However, it is not possible to determine from this document whether the funds were sent to the magistrates along with their yang-lien or were kept in the provincial treasury in anticipation of annual requests for construction funds by local governments.

The mere fact that we do not have direct references to chou and hsien kung-fei allowances outside of Chihli, Shensi, Szechuan, and Shantung does not mean that they did not exist. The gazetteer of An-yang hsien in northern Honan contains a detailed discussion of the various local expenditures that were previously met through the collection of irregular surcharges. According to the authors of the gazetteer, once the policy of huo-hao kuei-kung was implemented, all of these items were subsumed under a single huo-hao surcharge of 15 percent of the hsien's annual regular taxes.^[29] If we assume that all emoluments listed for ranking officials and teachers in Confucian schools, student stipends, sacrificial expenses, and alms for the widowed and poor were paid for from retained cheng-hsiang taxes (ts'un-liu ), the hsien still required 6,171.78 taels in huo-hao allocations to pay for normal yamen operating expenses and "outside" costs incurred by the hsien administration. It is likely, therefore, that a sizable allowance was granted to the hsien beyond the magistrate's own yang-lien . If this was the case in Honan, it may also have been true of other provinces for which documentation has not survived.

Kung-Fei to Higher Provincial Officials

In almost every province for which a provincial budget still exists, there is evidence that high-level officials were granted regular funds beyond their yang-lien quotas. In most cases, these funds took the form of special allocations to cover the costs of stationery and clerical wages. Usually they were budgeted separately and may have been thought of as elements of provincial public expenses, because these grants did not extend to officials below the rank of financial commissioner.

In Anhui province, the proposed budget submitted in YC 7 (1729) by Governor Wei T'ing-chen^[30] included provisions for separate payment of the wages of runners and clerks in the offices of the governor-general, governor, Tartar general and financial commissioner, as well as wages for the standard-bearers and bodyguards attached to the Tartar general and governor-general. Certain stationery costs in the governor's, governor-general's, and financial commissioner's yamen were also paid for with provincial funds. Even the educational commissioner was provided with stationery costs, and stipends were made available for the clerks and runners in his yamen when they were sent outside the provincial capital on missions.^[31]

The Kiangsu budget was not as generous as Anhui's in providing for the expenses of all high officials. It did, however, include substantial funds for stationery, gunpowder, and clerical wages for both the governor's and the governor-general's yamen. The only specific grant made to the financial commissioner was for stationery expenses.^[32] In Kiangsi, similar allocations were provided for the governor and governor-general, in addition to which the financial commissioner and judicial commissioner were given wages for their clerks and runners. The latter was also allowed a special fund to cover his statutory expenses.^[33] Specially budgeted kung-fei in Kweichow seems to have been limited to the offices of the governor and governor-general. Although the province contributed only toward the stationery expenses of the governor-general's yamen, provisions made for the governor's yamen included stationery supplies, clerical salaries, military supplies, firewood and lamp oil, and salaries for his private secretaries.^[34] At least part of these extraordinary allocations may be accounted for by the fact that at the time this budget was drawn up, no measures had yet been taken to provide the governor with yang-lien .^[35]

Both the amount of yang-lien granted to high provincial officials and the existence of special allowances of kung-fei to cover yamen costs indicate that considerable care was taken to ensure the fiscal solvency of officials residing in the provincial capital. This, of course, reflected the great expenses incurred in posts which had the responsibility of managing the affairs of the whole province. However, it also cannot be ignored that the original critique of deficits and corruption, out of which the reforms grew, placed a large part of the blame on high officials. If the government was to guarantee that high officials took their duty of supervising the fiscal behavior of their subordinates seriously, it was imperative that these same officials be completely removed from financial dependence on the chou and hsien.^[36]

The Kung-Fei of the Whole Province

Every province had a special category of revenue stored in the provincial treasury for the purpose of meeting provincial public expenses. This silver reserve is best viewed as a kind of provincial discretionary fund. In every province but Fukien, the amount of huo-hao revenues to be deposited in this fund was determined not by a calculation of annual expenditure requirements, but on the basis of the surplus available after the deduction of official yang-lien and other fixed expenses.^[37]

The logic of such a system becomes apparent when we examine the kinds of projects that were paid for from the provincial publicexpense fund. The reports of Governor T'ien Wen-ching to the emperor present a representative sample of kung-hsiang expenditures in Honan province. In 1725, huo-hao funds were used to buy a plaque for a state-sponsored temple. Cash was also lent to poor peasants whose land had been inundated by sand after severe floods, so that they could buy water buffalo to assist in plowing under the sandy soil. In addition, 8,000 taels were allocated to improve the military camps in the province.^[38] In 1727, 612 taels were used to buy eight government boats to outfit a ferry crossing at a strategic communications point, and an unspecified amount was issued to rebuild a temple.^[39] A grant of 2,000 taels was given to the Nan-ju circuit intendant to aid in flood relief in Ku-shih hsien, an area frequently affected by floods and small-scale bandit activity.^[40]

In response to imperial criticism of the size of Honan's huo-hao surplus, several major projects were initiated in 1729. To pay for

the repair of city walls, 65,550 taels were disbursed to nineteen chou and hsien. These funds were earmarked to cover the costs of both labor and materials. Another 5,836 taels were allocated to build 2,400 new examination cubicles to provide for the growing numbers of students coming to the capital to take the provincial examinations. Over 4,700 taels were used to construct a new jail for the judicial commissioner's yamen.^[41] Finally, 6,750 taels were spent to repair Honan's granaries, making it possible for the chou and hsien to replace existing buildings made of straw and mud with new facilities constructed of brick.^[42] In 1730, huo-hao funds were allocated to update and revise the provincial gazetteer.^[43] Moreover, a project to repair city walls was undertaken again in 1731, this time focusing on replacing walls that had been made of mud with more durable brick structures.^[44]

The situation in Honan was not exceptional. In discussions of provincial budgets submitted to the emperor, memorialists frequently mention appropriations for famine relief, river works, construction projects, and military procurement as falling into the category of provincial public expenses. In provinces where the chou and hsien officials were given separate kung-fei allowances, it is likely that some of the more distinctly local projects would have been handled without provincial-level intervention. However, in places like Honan, where huo-hao funds were more tightly controlled at the provincial level, magistrates generally had to petition for kung-hsiang funds, as was the case in Ku-shih hsien's appeal for famine relief. The reforms resulted in the transfer of much of the provincial surplus from the local to the provincial level, reducing the discretionary fiscal power of the magistrate. This loss of initiative on the local level was somewhat offset by the fact that control over large public-expense funds made it possible for the provincial government to initiate projects that individual chou and hsien might not have undertaken. Moreover, as in the case of citywall building or the large-scale construction of examination halls, economies of scale in the purchase of materials were made possible that would not have existed if projects were carried out on a piecemeal or strictly local basis, as in the past.

In the absence of detailed yearly accounts of provincial huo-hao expenditures, it is difficult to determine whether the benefits of kung-hsiang funds were distributed equitably among subordinate administrative units. For the few expenditures we do know about in Honan, it seems that huo-hao funds were indeed dispensed lib-

erally throughout the province. In the case of city-wall and granary construction, poorer areas were probably the most affected, because the emphasis was placed on upgrading construction that had earlier been undertaken with poor materials. However, there was a constant danger that concentration of the provincial surplus in the capital could result in a disproportionate allocation of funds to those chou and hsien within the capital environs. This seems to have been the case in Shensi province's 1726 budget,^[45] and it may reflect the fact that other chou and hsien in the province were given separate kung-fei allowances.

The institution of a provincial discretionary fund was in keeping with the demands of a political economy that required flexibility at all levels. Whereas administrative expenses were viewed as permanent, the expenses covered by provincial kung-fei were seen as temporary, often arising without warning and necessitating large outlays of money. Although floods or drought occurred frequently in many parts of China, no one could predict where or when they would strike next. On the other hand, projects such as city-wall construction or the building of charity cemeteries might be necessary only once in a generation. Upon completion of such structures, maintenance was usually turned over to local authorities. Therefore, it would not make sense for the provincial government to establish permanent bureaus of wall repair, with their own staff and budget. Where continuous funding was necessary, as in the case of boat construction in Fukien, or periodic dike maintenance throughout the empire, separate budgetary categories were established.

This is not to say that projects of the kind described above were carried out on an entirely ad hoc basis. Supervision of a particular undertaking was generally assigned to a permanent staff, headed by a nearby prefect or a circuit intendant^[46] and including expectant officials who used the opportunity to practice and demonstrate their administrative skills.^[47] Before allocations were made, estimates of material and labor costs were submitted by officials at the site.^[48] These estimates were subject to approval by the governor and, if central-government funds were involved, by the Board of Revenue as well.^[49] Officials in charge of a project were under constraints of time as well as money, as deadlines were usually set for completion in order to avoid delays and corruption.^[50] If a project took more than one year, accounts were submitted annually to ensure that the funds were not being wasted and that the final report

would not result in huge cost overruns. At the completion of the project, a final account of expenses was submitted and any surplus funds were returned to the provincial kung-hsiang .^[51]

Payments to the Central Government

One final category of kung-hsiang expenditures should be mentioned—funds spent in fulfillment of provincial obligations to the central government. It has been stressed that the huo-hao kuei-kung reforms were not an attempt by the central government to appropriate a larger portion of the local government surplus. Nevertheless, aside from those funds used to repay deficits, a part of the provincial kung-hsiang , and in some places a sizable part, did go directly or indirectly to the central government. It should be kept in mind that all of these expenditures would have been made whether or not the reforms had been implemented. Huo-hao kuei-kung had the effect of transforming what had been informal fees paid personally by high provincial officials into fixed allocations disbursed from the kung-hsiang of the whole province. Another burden was thus lifted from the shoulders of the governors, governors-general, and commissioners, reducing their own need to exact bribes and fees from their subordinates.

Payments to the central government fell into three broad categories: fees to central-government boards when submitting taxes or accounts, transportation fees for the shipment of those taxes and accounts, and costs of supplementing the Board allocations for the purchase of goods sent to the capital. Although we have direct evidence of such payments for only six provinces, it is reasonable to assume that some provision of this kind was made in every province. This is particularly so since the yang-lien of central-government officials was eventually budgeted from such provincial remittances.

The size of the allocations to the central government appears to have depended on the customary fees originally sent to the boards, the province's responsibilities in providing goods for the capital, and the size of its tax quota. The portion of the provincial budget devoted to such costs in the Kiangnan region was particularly high. Of a total of 71,144 taels allocated for public expenses in Anhui, over 50,000 taels were spent for purposes of no benefit to the province. These included fees to the Board of Revenue and its Board of Scrutiny at the time of the annual accounting, fan-shih to

the Board of Punishments, the costs of remitting taxes to the Board, and supplemental funds for the purchase of goods costing more than the Board price.^[52]

In Kiangsu, approximately 56,000 tads were set aside to pay expenses incurred in the remittance of taxes and to supplement the costs of dyes and pewter sent to the capital.^[53] However, inasmuch as that province's total kung-fei budget was considerably larger than Anhui's, payments to the central government did not cut as deeply into the Kiangsu surplus. This could not be said of Kiangsi, where almost half of the fixed public expenses of 62,204 taels were allocated for Board fees and the transport of regular taxes and shipments of goods to the capital in Peking.^[54] Chihli, as the site of the imperial capital, bore a particularly large responsibility for Board expenses in the form of remittance fees. If data for 1729 are typical, that province also spent about as much on Board fees as it did for all other provincial public expenses.^[55] These figures, of course, refer only to the fixed allocations in each province. Because all four provinces had surplus huo-hao and kung-fei which could be used as a discretionary provincial fund, the proportion of provincial revenues siphoned off by the central government, although high, was not as enormous as it first appears.

Outside these two regions, allocations of Board fees were probably considerably lower. Particularly in newly developing areas with low agricultural productivity, the central government was hesitant to make unreasonable demands on local resources. Of the 79,740 taels spent by Kansu province for yang-lien and public expenses, only 2,000 taels were sent to the Board of Revenues and Punishments in the form of clerical fees.^[56] Similarly, Kweichow spent only slightly more than 2,200 taels for this purpose.^[57]

Accounting and Accountability

The manner in which officials reported huo-hao income and expenditure to the central government provides an excellent example of the Yung-cheng emperor's reliance on personal authority in dealing with the provincial bureaucracy. Just as the decisions leading up to the adoption of huo-hao kuei-kung were taken outside the normal bureaucratic channels of communications, so the emperor divorced the day-to-day administration of the new provincial funds from routine scrutiny by the Boards. The collection and allocation of regular taxes were subject to rigorous Board sur-

veillance throughout the tsou-hsiao system of annual accounting. However, provincial authorities were specifically instructed not to report their huo-hao income to the Boards, but only to inform the emperor of their decisions via secret palace memorials.^[58]

All deliberations concerning rates of collection, distribution of funds in the form of yang-lien or kung-fei , methods of remittance, and so on were undertaken by the provincial governors, governorsgeneral, and financial commissioners. Although there is no record of the deliberations themselves, formal input from the lower levels of government probably was minimal, confined to requests for grants of aid for specific projects. Once the basic plan for huo-hao reform was established in a province, the governor or financial commissioner was required only to submit an annual report to the emperor in the form of a yellow register (huang-ts'e ) similar to those submitted to the emperor for regular taxes.^[59] These yellow registers merely listed total funds, following the format of the fourcolumn account. In this manner the emperor was kept informed of the aggregate income and expenditures in a province, leaving the details of disbursement to the provincial authorities themselves.

Occasionally a more elaborate report of expenditures was made, particularly when a change in official personnel in the provincial capital took place.^[60] These reports were similar to the chiao-tai reports submitted as evidence of proper management of regular tax funds when an official left office. In the case of both yellow registers and the more irregular end-of-term reports, the emperor usually returned the memorial without comment or with a simple "noted" (lan ) appended to the end of the account.

The implications of this arrangement were twofold. On the one hand, the Yung-cheng emperor displayed his continuing concern for the fiscal independence of the provincial governments. If the huo-hao kuei-kung reforms were to succeed, the provincial governments had to be assured the same flexibility in allocating funds that had been possible under the informal funding system. If the finances of the provincial treasury were subject to the same itemby-item scrutiny that was applied to the disbursement of cheng-hsiang funds, it would have been impossible for the provinces to respond quickly to emergencies or react to changing local conditions. The emperor acknowledged that he could not know the conditions within the provinces in the same way that a local official could. It was therefore logical to leave the details of huo-hao expenditure to these men.

Yung-cheng also seems to have had a special antipathy toward the interference of court officials in local fiscal affairs. During the debates surrounding the implementation of the reforms, the conservatism of officials within the capital bureaucracy, who were divorced from the daily operations of local administration, became readily apparent. Although the emperor was quick to shift the personnel holding the highest offices in the provinces, filling these crucial posts with men whose integrity and administrative philosophy he could trust, it was far more difficult to alter the composition of the central bureaucracy. Many of these men were holdovers from his father's reign, and their approach to local finance reflected the K'ang-hsi emperor's reluctance to tamper with the balance of fiscal power by legalizing huo-hao . Moreover, the emperor may have felt that these men, who had viewed deficits only in terms of corruption, and had failed to recognize the real shortage of funds on the provincial and local level, would not have been sympathetic to the spending needs of provincial administration.

The emperor did give the governors and financial commissioners broad discretionary powers in the allocation of provincial funds. In fact, there is reason to believe that he was willing to tolerate far more local independence than even the officials themselves recognized. The yellow registers of annual accounts were the only reports specifically required of officials by the emperor. Yet the emperor was constantly being barraged by reports ranging from those of the chiao-tai type mentioned above to two- or three-line notifications whenever a large sum of money was allocated for any purpose. These reports are symptomatic of a form of bureaucratic paranoia common among officials during the Ch'ing period. Even when given authority in allocating provincial funds, high provincial officials felt the need to report their actions to the emperor. In all probability, they hoped to cover themselves in case their successors questioned the way in which funds were handled during their tenure. This fear grew out of the heavy emphasis placed on an official's handling of tax matters in his career evaluations. Consequently, even though huo-hao revenues were not subject to the same sanctions and review that were applied to regular taxes, the association between the mishandling of taxes and impeachment seems to have been too ingrained in the officials' collective psyche to allow them to exercise their autonomy in such matters.

In contrast to the often compulsive reporting by high provincial officials themselves was the emperor's own responses to these re-

ports. When the emperor chose to issue any rescript to a memorial on the handling of huo-hao funds it was usually to urge the memorialist to take greater responsibility for decision-making himself. Sometimes he simply stated that he was leaving the decision up to the memorialist, as in the case of a report by the governor and governor-general of Kwangtung regarding allocations of yang-lien for submagisterial-level officials.^[61] More often the emperor simply indicated that he would not interfere in local fiscal affairs by stating that he took no interest in memorials of this kind,^[62] or that his habit was merely to glance at such memorials and nothing more.^[63] Occasionally, Yung-cheng even displayed some exasperation when officials continued to request imperial approval for decisions that should have been their own. After a long memorial from the Kiangsi governor requesting raises in certain officials' yang-lien allowances, the emperor wrote in an extremely informal style, "Okay, if it should be raised then raise it."^[64]

This is not to say that the management of huo-hao revenues was completely unsupervised. Notwithstanding his protestations to the contrary, the emperor did keep a close watch over the management of provincial finances through the palace-memorial system. As we have seen in the discussion of the implementation of the reforms, Yung-cheng was not averse to criticizing the manner in which a governor was handling the remittance, allocation, or disbursement of huo-hao revenues. The emperor also played a major role in guaranteeing that allocations were equitable and served the goal of redistributing the resources of a province where they were needed. Moreover, as the only agent privy to knowledge of the reforms in every province, the emperor also played an integrative role, informing officials as to how the reforms were being carried out elsewhere and ensuring that there were no undue discrepancies in the rates of yang-lien being paid to officials of comparable rank and responsibilities in different provinces.^[65] Thus, although the actual management and allocation of yang-lien was in the hands of the governor and financial commissioner, these funds were still ritually attributed to the emperor as the ultimate source of all revenues and laws in the empire. Even when an official decided on his own yang-lien , it was not uncommon to send a memorial thanking the emperor for his magnanimity in granting the official this new source of funds.^[66]

The emperor's occasional intervention in provincial financial affairs was not intended as an encroachment on provincial authority.

In view of the structure of authority in the Ch'ing legal system, it played an important function in strengthening the power of the provincial leadership as well as enabling the emperor to check abuses in the exercise of that leadership. But without the assignment of some general supervisory functions to the Board of Revenue, the provincial authorities had no means to enforce their control over the funds emanating from the local level. Because huo-hao revenues were not integrated into the regular Board of Revenue statutes, the provinces were allowed a far greater degree of autonomy in handling huo-hao than they had in handling cheng-hsiang . This assured considerable flexibility in the allocation of funds. However, it had one serious drawback, which had its roots in the dual nature of the Ch'ing legal system. Judicial cases involving crimes committed by commoners could be tried and adjudged at all levels of the bureaucracy. Although appeals could be made to the court and the emperor, provincial officials did not require central-government intervention in handling most cases. The same was not true when crimes involved misconduct in office. In such cases, although the official's superior in the provinces could recommend sanctions, the decision to apply sanctions against an official lay with the Boards and the emperor. Moreover, impeachable crimes had to be defined in the statutes of the various Boards, which taken together constituted a code of official discipline. In order to preserve provincial fiscal autonomy, huo-hao matters had not been integrated into the official code, hence corruption or recalcitrance in the implementation of huo-hao kuei-kung could not be dealt with through the usual channels used in cases of violation of the disciplinary code. High officials had only one way to deal with malfeasance in the newly created huo-hao kuei-kung system, which was to report the malfeasance directly to the emperor in a secret palace memorial and request that special sanctions be applied against the official involved.

Once deficits were paid up, as pointed out in chapter 4, magistrates in Shansi were reluctant to continue remitting huo-hao to the provincial treasury. This problem was not limited to Shansi province alone. In YC 10, Hupei Governor Wang Shih-chün sought imperial guidance in handling a similar problem in his province. Between YC 1 and YC 3, twelve hsien magistrates failed to remit a total of 4,232 taels of huo-hao to the provincial treasury. Although the amount was small, these cases were indicative of the difficulties encountered in trying to control huo-hao revenues. The

governor had clear authority to impeach a subordinate and demand that he repay cheng-hsiang deficits, but he did not know if he could do the same in the case of deficits in the kung-hsiang .^[67] The emperor assured Wang that deficits in the provincial publicexpense fund were an even more serious matter than shortages in cheng-hsiang revenues, and advised against leniency in dealing with these magistrates. Stern procedures for retrieving the funds were therefore set in motion, procedures which the governor probably would not have invoked without the authorization of the emperor.

The Yung-cheng emperor's extraordinary attention to the details of everyday administration is well known. Even a cursory examination of the thousands of memorials bearing his voluminous personal comments provides convincing evidence of his care and understanding in supervising local-government affairs. In his effort to bypass normal bureaucratic channels and deal with local fiscal matters through direct communications with provincial officials, Yung-cheng had no choice but to handle personally the numerous memorials received at his quarters every day. There is evidence, however, that the emperor did not shoulder the entire burden himself. The few extant disposition slips submitted to the emperor on local fiscal matters by Prince I, in his capacity as President of the Board of Revenue, indicate that occasionally the emperor did turn to his most trusted fiscal advisors in deciding provincial fiscal issues. Even though secret palace memorials were to be viewed only by the emperor and the memorialist, Yung-cheng did sometimes show tsou-che concerning huo-hao and yang-lien matters to the Board of Revenue. Furthermore, these secret palace memorials were, in rare instances, even copied and kept in the Board's files.^[68] Nevertheless, the Board only acted in an advisory capacity when reviewing these memorials and did not have the same powers of scrutiny that it exercised in the acceptance or rejection of chengh-siang accounts.

In YC 9 (1731), the system of reporting was modified to include additional input by the Board of Revenue. According to a Board communiqué, all governors and governors-general were ordered to inform the Board of any increases made in yang-lien quotas. The Board would then deliberate whether or not the proposed increases should be approved, and their recommendation would be sent to the emperor. The Board also indicated that it would inspect the accounts of huo-hao funds spent and saved when they were

reported annually to the emperor.^[69] This does not appear to have been an order requiring provincial officials to report their annual huo-hao accounts in a routine tsou-hsiao tse , but rather was an announcement of the Board's intention to continue its practice of informally advising the emperor on the accounts submitted as secret palace memorials. It was not until the Ch'ien-lung reign that the real discretionary powers of the provincial government in the management of local finances were encroached upon by the central government.

Yang-Lien

Changes in Yang-Lien Allocations

The yang-lien rates established in the early years of the Yungcheng reign were not viewed as static quotas. Throughout the first decade of the reforms there were numerous instances of yang-lien being increased and occasionally diminished in order to meet the needs of officials in the field. Sometimes these increases were in the form of across-the-board raises for all officials in a province. At other times an increase was granted to an individual official whose expenses were agreed to be larger than the funds provided when huo-hao kuei-kung was first implemented in his province. Often the original deliberation to raise yang-lien came from the provincial officials themselves, but a change in rates could be called for by the emperor and, after 1731, at the instigation of the Board of Revenue in its advisory capacity.

The most striking across-the-board raises in yang-lien occurred in Fukien and in Honan. The rates of yang-lien originally formulated in Fukien were considerably lower than those for comparable posts in other provinces in central China. This was the result of both the low rate of huo-hao collected in that province and the high proportion of provincial funds that had to be allocated for administration and defense. As we have seen in chapter 4, the emperor was concerned that allocations of yang-lien in Fukien would be insufficient and ordered that surplus revenues from official lands and miscellaneous commercial taxes be used to compensate for shortages in provincial funds. Over 40,000 taels were required to bring Fukien's yang-lien rates into line with the needs of local officials.^[70] When the surplus revenues granted by the emperor proved inadequate to meet all the costs of yang-lien and kung-fei , an addi-

tional grant of 20,000 taels was made to the Fukien treasury from cheng-hsiang revenues disbursed by imperial decree to Chekiang province.^[71]

The circumstances leading to increases in Honan's yang-lien were the opposite of those found in Fukien. The success of the huo-hao kuei-kung reforms in Honan was reflected in an everincreasing surplus in the provincial treasury. In 1728 the emperor reacted to this accumulation of revenues by ordering Honan Governor T'ien Wen-ching to devise ways to spend more of the provincial income for the benefit of the people and administration. T'ien responded by suggesting increases in the yang-lien quotas disbursed to lower-level officials. The governor felt that the 3,000 taels annually allocated to each circuit intendant was adequate. He did, however, ask for an increase of 400 taels for chou and hsien magistrates and 300 taels for independent chou magistrates. This raised the yang-lien as follows: independent chou magistrates to 1,800 taels, large chou and hsien to 1,400 tads, medium chou and hsien to 1,200 taels, and small hsien to 1,000 taels.^[72]

There are also numerous examples of yang-lien rates being increased for individual posts. Such raises were generally proposed by the provincial governor or financial commissioner. In YC 5 (1727), Li Wei requested that the Chekiang educational commissioner's yang-lien be raised from 1,000 taels to 2,500 taels.^[73] In YC 9 (1731) Kiangsi Governor Hsieh Min proposed that several border hsien and small hsien with large populations of "tent people" be reclassified as medium and large hsien and their yang-lien raised accordingly.^[74] These changes were made in recognition of the difficulty entailed in controlling border areas and in preventing the people from fleeing into neighboring provinces to evade taxation. The grain and post intendants in Kiangsi had originally been in charge only of tribute grain and post and salt affairs. During the Yung-cheng period, however, each was given jurisdiction over three prefectures and was made responsible for all judicial matters in their respective subordinate chou and hsien. As a result, both intendants were faced with greatly increased travel expenses and the costs of hiring additional private secretaries expert in local administrative and legal issues. Therefore a raise in yang-lien was also requested for them.^[75]

Changes in the remittance of customary fees could also give rise to modifications in the allocation of yang-lien . When the reform first went into effect, the director-general of water conservancy

(Tsung-ho ) received contributions from salt merchants which constituted the major portion of his yang-lien . When these contributions were eliminated, the director-general was left with only a small grant from merchants and surplus silver from the river treasury, totaling around 5,000 taels. This was not enough to cover the costs of travel to inspect construction sites, assisting the governor, rewarding soldiers attached to the river administration, hiring private secretaries, and so on. Therefore, a request was made that a raise in yang-lien be granted for this post.^[76] The elimination of fees as a means to support river administration and the substitution of increased yang-lien also reflected the gradual trend toward greater fiscal rationalization in local administration.

Officials sometimes requested raises in yang-lien for their own posts. In YC 6 (1728), the Hupei financial commissioner complained that the approximately 11,600 taels he received for yang-lien and operating expenses were too little in view of the high costs of remitting accounts to the Board and the numerous private secretaries and servants needed to man the province's top financial yamen. Hsü Ting claimed that because of the inadequacy of the funds provided in the provincial budget, he had been forced to rely on assistance from his father to cover his official expenses. This would not have been at all unusual under the prereform system of financing. Moreover, the usual response of emperors to claims such as Hsü's would have been an admonition to exercise greater frugality in his personal and official life. That Hsü Ting's own fears of such a reply proved unfounded is testimony to the transformation of the concept of fiscal responsibility after the reforms. Instead of chastising Hsü for excessive expenditures, Yung-cheng praised the plan to raise his yang-lien and stated that it had never been the policy of the dynasty to require an official's family to support him financially in the fulfillment of his administrative duties.^[77]

By the end of the Yung-cheng period, officials were sometimes quite militant in insisting upon their right to receive adequate yang-lien . Szechuan Governor-general Huang T'ing-kuei was particularly angry about the way in which his request for a raise in yang-lien was handled after authority to judge such requests was extended to the court. Huang memorialized that the Szechuan governor-general's yang-lien was only 6,000 taels and requested that it be augmented by 4,000 taels. He received a reply from the Grand Secretaries Chang T'ing-hu and Chiang T'ing-hsi by court letter. Enclosed was an imperial edict which stated that the sum of

6,000 taels was really insufficient. However, since Huang was concurrently in charge of the post of provincial commander-in-chief, the edict ordered the grand secretaries to inquire whether or not Huang received yang-lien for that second post, and expressed anger that he had failed to mention this additional yang-lien in his memorial.^[78]

Huang devoted an entire page of his rejoinder to insisting that he was not attempting to profit personally from his official responsibilities, pointing out that the post of provincial commander-inchief carried no yang-lien , but only about 2,800 taels a year from emoluments and military rations. Huang stated that he had already run into shortages and had been forced to shift funds and to borrow 1,000 taels from the governor, which he had not yet been able to repay. Moreover, much of the funds from his military command had gone to pay for two military expeditions to Liang-shan, to reward soldiers, and to hire laborers and coolies. Finally, Huang complained that although the duties of the governor-general were the most numerous, the governor received 19,000 taels annually in yang-lien and the financial commissioner received 10,000 taels. He thereupon closed his memorial by doubling his original request to 20,000 taels per year for the governor-general's yang-lien , a request which the emperor acknowledged and accepted with a simple response of "noted."^[79]

In general, the emperor left the determination of yang-lien rates to the provincial officials themselves. When he did intervene, it was most often when yang-lien was being established for a post for the first time. In such cases the emperor acted as an advocate for the official concerned to ensure that the yang-lien decided on was sufficient for his needs. For example, when the Anhui, Kiangsu, and Kiangsi governors were consulted to provide yang-lien for the Liang-chiang governor-general, it was suggested that the total contribution from the three provinces be 14,000 taels. Yungcheng felt that this would not be enough and hinted that 20,000 taels would be a more appropriate figure.^[80] Imperial intervention was also instrumental in guaranteeing adequate yang-lien for the Hunan governor. When Financial Commissioner Chu Kang suggested that the governor be granted a quarterly allowance of 1,000 taels, for an annual yang-lien of 4,000 taels, the emperor replied:^[81]

How can 1,000 taels be enough for the governor's needs? Previously, when Pu-lan-t'ai memorialized to

sell his family property [in order to finance his administration] I sent a rescript vehemently stating that this was not right. When the new governor takes office, you should discuss adequate yang-lien [with him]. Your present stance makes you appear to be offering a pretense of virtue. I do not make it a principle to have high provincial officials perform their duties on an empty stomach [i.e., without adequate compensation]. As Hunan Financial Commissioner these last few years you understand the situation thoroughly. . . . Handle matters fairly.

In rare instances an official requested that his yang-lien be lowered. In YC 11 (1733), the Shansi governor reported that the new acting judicial commissioner had spent only 7,000 taels of his annual 10,000-tael yang-lien . He therefore requested that the extra 3,000 taels be returned to the provincial public-expense fund.^[82] This, however, was reduction in that individual's yang-lien and probably did not mean a revaluation of the yang-lien quotas for all future judicial commissioners in the province. The only instance where the downward adjustment of yang-lien might have had an institutional impact was T'ien Wen-ching's refusal to accept additional yang-lien when he was elevated from the post of Honan governor to that of Honan-Shantung governor-general.^[83] However, this post was created expressly for T'ien and was eliminated soon after his death.

Extension of Yang-Lien

The trend toward fiscal rationalization was not reflected merely in increased yang-lien for posts where the original quota had proven inadequate. As the reforms became established, the number of officials to whom it was granted was also extended to include all the ranking civil officials in the provinces. We have already seen how the omission of yang-lien for the governor and governor-general of several provinces was quickly rectified.^[84] In many provinces it was several years before yang-lien was provided for education commissioners. This was because they frequently had an informal source of funds that was considered legitimate and served as a substitute for yang-lien . In general, these funds were linked to the administration of the civil-service examinations. For example, whenever the education commissioner in Shensi went to an area to hold the lower examinations, gifts were supplied by the

chou and hsien whose residents were among those attending the examination. These gifts were called k'ao-p'eng kung-ying or contributions of supplies to the examination hall, but were really obtained by levying surcharges on the local populace.^[85] In Hupei the education commissioner received his funds from the sale of the examination booklets used by students to write their answers. For each set, a fee of two or three taels above the cost of the paper was charged, yielding as much as 6,000 taels per year.^[86] These sources of funds were so widely accepted that in YC 5 (1727), when the Kwangtung education commissioner, Yang Erh-te, requested an increase in his yang-lien , the governor memorialized that the request should not be granted, because it was expected that the education commissioner would be provided with funds by the chou and hsien in which he conducted the examinations.^[87]

The emperor left it up to the individual provinces to decide whether or not to grant yang-lien to education commissioners. On the whole he felt that to grant them four or five thousand taels a year was appropriate.^[88] Moreover, Yung-cheng hoped that providing them with yang-lien would be an incentive to greater diligence in the execution of their duties.^[89] The reason the emperor did not insist on the suspension of contributions for the support of these officials and the institution of empire-wide yang-lien payments for education commissioners may be traced to his fear that some officials would nevertheless continue to exact fees for holding the examinations. Because this fee was paid directly by those involved in the examinations, it was not something that could be easily stopped by applying official sanctions. There was no way to ensure that such a widely accepted practice could indeed be eliminated by fiat.^[90] Despite the emperor's lack of leadership in this matter, efforts were made eventually in several provinces that had relied on informal funding practices to bring the funding of education commissioners into the mainstream of the reforms.^[91]

When the huo-hao kuei-kung reforms were originally formulated, yang-lien was provided only for civil officials of the rank of magistrate and above. Those officials closest to the people, including assistant magistrates and the numerous officials attached to the main administrative yamen, such as registrars, jailors, education officers, and other minor officials of the eighth, ninth, and unclassified rank (tso-tsa ) were not included in the new system of funding. The reasons for this omission were twofold. On the one hand, officers below the rank of magistrate were not considered

part of the main body of the bureaucracy. Often these posts were filled by people who had not passed the highest levels of the examination system or who had obtained their rank by purchase. Except in rare instances, they were not eligible for promotion to higher positions, and although holding office, they were viewed by the rest of officialdom as not far removed from the clerks and runners below them. Moreover, they did not command the sort of multifunctional yamen found at higher levels in the bureaucracy and were not seen as requiring the extensive staff found under officials of higher rank. Economic constraints also played a part in their exclusion from the yang-lien rolls. During the early years of the reforms, there was a constant fear that huo-hao revenues would not be sufficient to cover all the costs of provincial administration. To include these low-level functionaries in the ranks of officials supplied with yang-lien might have been judged to be an inordinate burden on provincial resources.

This proposition is confirmed by the Yung-cheng emperor's response to the first attempt to bring minor officials into the yang-lien budget. Honan Governor T'ien Wen-ching first proposed providing yang-lien for low-level functionaries early in YC 3 (1725). He pointed out that these officials had important duties and were constantly being sent out on government missions. Their material rewards were meager, however, and their rank low, providing little incentive for them to abide by the official disciplinary code. T'ien, who himself had risen from the rank of assistant hsien magistrate, felt that many of the men in such posts were extremely talented, and he proposed that if the emperor extended his benevolence to them by granting them yang-lien , they naturally would respond by displaying more virtue and devotion in the execution of their duties.^[92]

The emperor was not as confident as T'ien of the efficacy of such a policy. Although once again insisting that all such decisions were up to the governors and governors-general, Yung-cheng accepted the popular notion that very few of these low officials were really distinguished civil servants. The extraordinary few who were could be rewarded under a recent imperial act of grace that permitted their recommendation for promotion to higher rank. Rather than grant yang-lien to these lowly subordinates, the emperor felt it would be better to increase the yang-lien of the magistrates so as to enable them to improve the general level of administration through-

out the chou and hsien. Yung-cheng's real concern, however, was reflected in the last lines of his rescript:^[93]

In this instance I have unavoidably been overly severe. You should weigh the situation further and not immediately take my words for the truth. . . . I am even more concerned that this matter does not concern Honan province alone. If there are provinces whose huo-hao is insufficient, where will they find the funds to spread this benevolence? You must think further about this point.

In other words, the emperor was concerned that by allowing low-level officials in Honan to receive yang-lien he would appear to be favoring the officials of one province over the others. Local differences in the application of the reforms were permissible so long as the general structure of fiscal management remained fairly uniform throughout the empire. Honan's surplus could not be used to provide its officials with benefits that other provinces could not match. So long as it was unclear whether other provinces would be able to follow suit, it was not advisable to allow Honan alone to take the reforms this major step further, no matter what the benefits would be in eliminating local corruption and rationalizing local fiscal management.

By the beginning of YC 6 (1728), it was becoming apparent that the reforms were a success and that adequate funds could probably be found to allow every province to extend its yang-lien budget to include all officials. Early in the year, T'ien Wen-ching once again proposed that low-level officials be provided with yang-lien . The real intention of such a move, to eliminate corruption on the lowest levels of the bureaucracy, was made clear by T'ien's concession that officials who were not in direct contact with the people need not be considered for yang-lien . However, such officials as assistant and subassistant prefects and magistrates and registrars, whose role in supervising and investigating local affairs brought them into close daily contact with the people, would be given yang-lien of between 80 and 120 taels.^[94] This time the emperor found T'ien Wen-ching's proposal both fair and appropriate and told him to carry it out.

Concern about the continued acceptance of customary fees despite the institution of yang-lien was instrumental in the decision to extend this new dimension of the reforms beyond Honan prov-

ince. In mid-1728, T'ien Wen-ching was promoted to the rank of governor-general of Honan and Shantung. One of his first acts as governor-general was to expose the continued use of customary fees as a means of funding in Shantung. T'ien's report prompted an edict by the emperor that was to result in the extension of yang-lien to all officials in the provinces as the most effective means to eliminate forever these corrupt and unregulated exchanges of funds.

In his edict of the eighth month of YC 6, the emperor acknowledged that some officials continued to receive lou-kuei regardless of the fact that they had already been granted yang-lien . This was to be deplored. Yung-cheng ordered the governors and governorsgeneral of each province to investigate their subordinates one by one and report on their activities in a secret palace memorial. The emperor still felt that the solution to this problem was to ensure that every official had adequate yang-lien . Therefore, he also ordered the governors and governors-general to find out if there were any officials under them who had not yet been granted yang-lien and to extend it to them. At the same time Yung-cheng publicly endorsed the complete remittance of huo-hao to the financial commissioner and public redistribution of funds from the provincial treasury.^[95]

One of the first provinces to respond to the emperor's edict was Hunan. The edict prompted Hunan to switch from partial to complete remittance of huo-hao according to the Shansi-Honan model. Officials in Hunan also used this opportunity to calculate yang-lien for minor officials within the yamen at all levels of the provincial bureaucracy. Following the principles set down by T'ien Wen-ching, these new grants of yang-lien were made only to officials who came into close contact with the people or who frequently were sent outside the yamen on official missions. Among those included were the secretaries in the financial and judicial commissioners' yamen (li-men, cho-mo ), treasury keepers, jail wardens, registrars, and assistant chou and hsien magistrates.^[96] Hunan's huo-hao revenues were not as large as Honan's; this accounts for the lower rates at which yang-lien was granted to these officials. Nevertheless, the province was able to supplement the cost of these additions to the provincial budget with the return to the public coffers of 8,000 taels in salt fees formerly sent to the grain intendant.^[97]

Not every province was so quick to comply with the intent of the emperor's edict. In many instances, attempts were made to in-

crease the existing yang-lien quotas or provide funds only for higher officials who had not been previously included in the provincial budget. It was in such cases that Prince I and the Board of Revenue began to exercise the initiative that eventually may have led to their greater authority in advising on yang-lien matters. For example, in YC 7 (1729) Chihli Governor-general Yang K'un sent the emperor a recapitulation of the deliberations conducted in that province to provide each official with adequate and fair yang-lien . Prince I and the Board found the amounts agreed upon to be generally appropriate and recommended that the emperor order him to allocate the yang-lien as suggested and turn the original accounting report over to the Board. Although these deliberations provided yang-lien for independent and regular chou subassistant magistrates, no attempt was made to provide yang-lien for officials of the rank of assistant hsien magistrate and below. In view of the fact that almost 80,000 taels still remained in the Chihli treasury after some 220,000 taels were disbursed as yang-lien , the Board recommended that the remaining low officials be provided with yang-lien ranging from several tens of taels to around 100 taels apiece.^[98]

The emperor quickly reacted to the Board recommendation and ordered Yang K'un to reconsider yang-lien for minor officials. Yang consulted the financial commissioner and submitted a yellow register to the emperor listing possible rates of yang-lien for the officials in question. Surprisingly, the emperor responded that the rates seemed somewhat high and that he feared they would be too great a drain on provincial resources. Once again, Yung-cheng took the administration of Honan's finances under T'ien Wenching as a model and told Yang that he might inquire secretly as to the amounts these officials were granted in that province.^[99] In the meantime, Yang K'un was replaced as governor-general by Metropolitan Censor T'ang Chih-yü. T'ang resubmitted a reduced yang-lien schedule that would cost the province slightly over 8,600 taels annually for minor officials, leaving the province a surplus for public expenses of over 30,000 taels.^[100] These new rates were submitted to the emperor in the form of a yellow register and to the Board of Revenue in the form of a clear account (ch'ing-ts'e ), indicating the increasing role being assumed by the Board in yang-lien matters.

Yang-lien for minor officials was such a novel concept that many provinces took years to carry it out. Occasionally delays were due

to disputes over the amount of yang-lien that should be provided. In Kwangtung the provincial leadership readily acceded to the idea of providing yang-lien for these officials. However, differing interpretations of the purpose of salaries led to disagreement over acceptable rates. Both the governor-general and the acting governor regarded these new grants of yang-lien as token payments, along the lines of the emoluments granted officials by the emperor in the form of feng-yin . Therefore, they proposed simply to supply each official with funds from huo-hao equivalent to their feng-yin quota, thus doubling their salaries.^[101] The financial commissioner disagreed with this method, pointing out that yang-lien was not intended merely to pay for an official's personal expenses, but also to cover the costs of yamen administration and missions outside the yamen. If this plan were followed, the officials would have enough to support their families, but nothing more.^[102] The emperor left the decision up to the provincial authorities, who eventually came around to the financial commissioner's point of view. As a result, yang-lien for lower officials was set at between 60 and 80 taels, a decision that added 17,520 taels in new fixed expenses to the already strained Kwangtung provincial budget. The funds for the additional yang-lien grants were provided by surplus revenues from Kwangtung's commercial licensing tax (shui-ch'i hsien-yü yin ).^[103]

In view of the number of people involved, it is not surprising that ranking officials were hesitant to allocate large quotas for minor officials' yang-lien . In Kwangtung alone there were 255 officials for whom yang-lien was added. The largest categories were jail wardens (78) and subdistrict magistrates (126).^[104] This latter figure is of particular interest because it shows that in at least some hsien, not only was there an assistant magistrate, but there must have been more than one subdistrict magistrate below him, bringing the administration far closer to the people than has generally been recognized. By providing such officials with sufficient administrative funds, the government was clearly taking an important step in the direction of eliminating illegal surcharges and extortion.

The control and elimination of corruption at the lowest levels of the administration seems to have been foremost in the minds of provincial officials as they sought to find the funds to furnish minor officials with yang-lien . As each province reported its deliberations to the emperor, several concerns expressed by the Kiangnan governor-general and Anhui governor were echoed re-

peatedly. Anhui had originally failed to provide yang-lien for its minor officials because there was not enough huo-hao in the provincial treasury. It was common knowledge that because these functionaries had no means to pay for their administrative expenses they exacted money from the villagers in their areas (ssu-ch'u li-hsia ) and engaged in the practice of handling litigation for the people without official authorization (shan-shou min-ssu ).^[105] In YC 10 (1732), it was discovered that in Anhui salt fees of approximately 30,000 taels were being sent to high and low yamen in the province. With the approval of the grand secretariat and the emperor, Governor-general Kao Ch'i-cho and Governor-general Li Wei decided to return these funds to the provincial treasury to be used for provincial public expenses. As a result, they were able to allocate 50 tads apiece to the 163 minor officials in closest contact with the people, in the hope of eliminating the illegal methods utilized to support local administration.^[106] In fact, in many provinces it was the cooptation of former lou-kuei and surplus commercial and customs revenues that made the extension of yang-lien and the success of huo-hao kuei-kung possible.

Lou-kuei after the Reforms

In chapter 2 we examined the informal funding network by which local government was financed before the huo-hao kuei-kung reforms. What became of these various items of "saved funds," bribes, customary fees and so on, once officials were provided with regular salaries and public expenses? Contrary to our expectations, not all of these funding sources were declared illegal and eliminated from the provincial budget. Some continued to be collected, but rather than entering private pockets in the form of graft, they were sent to the provincial treasury to "supplement public expenses" (ch'ung-kung ). It was not mere expediency that allowed the Yung-cheng reformers to retain some of the irregular sources of income that had flowed into the informal funding network and reject others. In order to understand how these decisions were made, we must examine the values that lay behind the late-imperial conception of bureaucratic corruption.

Throughout China's imperial history, low taxation was viewed as a fundamental prerequisite of benevolent rule. During the early

Ch'ing this was linked to a vision of relatively static agricultural productivity which reached its ultimate expression in the K'anghsi emperor's ban on future increases in the head tax to match China's growing population (yung-pu chia-fu ).^[107] Although this proscription did not debar the government from extracting additional revenues from increased productivity on the land, especially that resulting from land reclamation,^[108] it did prevent any increases in the legal rate of taxation on lands already entered in the tax rolls. K'ang-hsi recognized the need of local officials for supplementary income, but deemed it safer to allow them to collect it illegally than to condone a rise in the land tax. The risk of diminished central-government control over the behavior of local officials was far outweighed by the possible damage such an action would inflict on the emperor's credibility as a benevolent ruler. At a time when tax resistance was rampant and an alien dynasty was attempting to establish its legitimacy as the rulers of the Chinese people, presentation of the imperial image took precedence over matters of practical governance.

Such a position was possible so long as the "state finances" (kuo-chi ) and "people's livelihood" (min-sheng ) were not too adversely affected. When the illegal diversion of funds itself became a threat to both the kuo-chi and the min-sheng , it became necessary to rectify the existing fiscal system, even if it meant raising taxes and tarnishing the emperor's personal reputation.

If we view the Yung-cheng-period reforms as an effort to readjust the fiscal administration of the empire in order to protect the kuo-chi and min-sheng , the selective elimination and incorporation of elements of the informal funding network becomes more comprehensible. Two types of funds within the informal funding system were strictly banned during the reforms—private levies on the people (ssu-p'ai ) and customary fees exchanged between officials (lou-kuei ). The former were seen as instrumental in depleting the resources of the common people and as a primary cause of arrears in regular tax collections. The latter, by draining the funds available to lower officials, were viewed as a major cause of deficits in provincial treasuries. By fostering informal links of patronage, it was the main obstacle to bureaucratic discipline and morale. Moreover, when these forms of graft interacted, they tended to reinforce each other. Increased demands for customary fees forced lower officials to step up their collections of surcharges from the

people. In order to protect themselves from disciplinary action as a result of the arrears and deficits caused by these illegal diversions of funds, officials were increasingly willing to pay off their superiors in the form of lou-kuei . It was for these reasons that ssu-p'ai and lou-kuei were the primary targets of the reforms, and it was to eliminate the need for such fees and surcharges that rational and legal funding on the local level was established.

Even before the reforms were instituted the emperor issued an unequivocal ban on surcharges and customary fees. Once the reforms were in effect, Yung-cheng continued to warn officials against such exactions. In his edict calling for the extension of yang-lien payments to all officials not previously provided for in provincial budgets, he again decried the collection of customary fees and specifically linked yang-lien to the elimination of these illegal practices.^[109] Even the large-scale general audit of Kiangnan tax arrears undertaken in YC 7 (see chapter 6) was in part a response to irregularities in local tax-collection practices that defied the ban on surcharges and interfered with the well-being of the kuo-chi and min-sheng .

However, a number of items within the informal funding network continued to be collected after the reforms. In fact, our knowledge of many categories of funds within the network depends on reports to the emperor when these funds were returned to the provincial coffers to supplement public expenses. Surplus from the purchase of rice, materials for construction, and so on was considered a legitimate source of public funds. For example, taxes formerly levied in kind to provide local granaries with rice or to fulfill central-government quotas of tribute rice were commuted to cash payments at fixed rates. Inasmuch as the market price necessary to buy the rice for these purposes fluctuated greatly, any profits that accrued to the local government as a result were not viewed as having been taken from the central government or the people. Officials in Fukien were able to generate over 5,000 taels from the surplus "rice price" (mi-chia ).^[110] Similar surpluses, arising from the purchase and sale of stabilization rice, were reported in Shantung.^[111] There, the financial commissioner also reported an average annual surplus of 3,000 taels in tribute-grain taxes that were also returned to the public treasury.^[112] In contrast to windfall profits from the purchase of goods, the accumulation of funds through the use of improperly balanced scales was considered cor-

rupt, because it entailed collecting more than the legal quota from the people. Whenever possible, this practice was eliminated, and in some instances new scales were issued to provincial yamen.^[113]

Salt Fees

The most important of all the informal funds coopted by the provinces during the reforms were salt fees (yen-kuei ). Almost every province depended to some extent on fixed periodic payments made by salt merchants to supplement the revenues in the provincial treasury.^[114] Generally, these fees were paid to individual officials. In addition, in salt-producing provinces such as Yunnan, officials received surplus revenues from productive salt wells and surplus profits from the sale of salt within the province.^[115] Following the emperor's proscription against the acceptance of customary fees, there was considerable confusion as to whether salt fees were to be included in the ban. Many officials, fearing prosecution for the acceptance of salt fees, immediately memorialized the emperor to demonstrate that they had used the funds derived from salt fees for public purposes. For example, Shantung Governor Huang Ping reported that during his six years as a judicial commissioner he had received a total of 30,000 taels in the form of "customary gifts from salt merchants" (yen-shang kuei-li ). During that period he had used these funds to establish charity schools in each of the six prefectures in the province, to institute stipends for students and to pay teachers' salaries, to supply aid to families of deceased officials without the means to return to their home provinces, to aid starving refugees, to help repair bridges, buildings, and roads, to build charity cemeteries, and to help clothe and feed prisoners. In addition, in the spring of KH 60 (1721), when salt bandits appeared in the province, Huang claimed to have used part of his salt fees to send extra soldiers and police to assist in their capture.^[116]

Despite their fears of imperial sanctions, officials did not move as quickly to curtail the collection of salt fees as they had in the case of surcharges levied on the common people. On the one hand, fees from the salt administration, as a source of revenue for high provincial officials, were too important to be eliminated without some consideration. Moreover, fees from salt merchants, and from merchants in general, were not viewed in the same light as fees exacted from the people. Whereas the productivity of the land was

seen as finite, so that additional exactions by local officials would of necessity mean less available income for either the central government or the people, the salt industry was expanding and could accommodate additional demands for revenue without too much harm to the merchants and with virtually no drain from the revenues due the central government. Even more important, the salt merchants from whom these fees were derived obtained their wealth by the emperor's grace in their participation in an official monopoly. Therefore, it was considered natural that they should want to repay this favor by contributing to the well-being of the area in which they did business.

In fact, the main argument offered by officials in favor of the continued acceptance of salt-merchant fees and contributions was that they were given voluntarily and therefore were not to be compared with forced levies on the common people.^[117] The case of a group of Ho-tung salt merchants was typical. In YC 3 (1725), 100,000 new salt licenses were issued in Shensi province. The salt merchants of the region benefited enormously from the opportunity for increased sales, and in YC 6 (1728) they volunteered to contribute 5,000 taels annually to the public expenses of the province. According to the merchants themselves, they made such an offer not only because they profited from the increased issue of salt licenses, but because the dynasty itself had brought such prosperity to the land that the population was increasing and the demand for salt was constantly rising.^[118]

One merchant, Chang Hung-yü, felt that his life had been particularly touched by imperial grace and wished to contribute an additional 2,000 taels annually beyond that being proffered by the merchant group. Chang stated that his grandfather had been the first in his family to go into the business of selling salt. Chang and his father inherited this business from his grandfather. At first they were just small merchants doing business in the countryside. There were no contract merchants at the time to transport and sell the salt. After he had transported the salt from the well, he did not even have exclusive rights over its sale. Now, Chang said, he was in charge of all salt licenses for Honan's T'ang hsien. His license quota was 4,787 yin . As the population of T'ang hsien increased and harvests continued to be reaped in abundance year after year, he was able to sell all his quota and an excess quota (yü-yin ) as well.^[119] Chang offered to repay the government for his own success by contributing part of his profits to the public fund. It is impor-

tant to note that it was only after the Board of Revenue was satisfied that the 7,000 taels these merchants wished to contribute were indeed "excess profits" resulting from the sale of additional salt licenses that they granted the merchants' request.^[120]

From the above example we can see that the government was not unconcerned about the merchants' own welfare. Often, when salt fees continued to provide a source of provincial income, the amount paid by the salt merchants was reduced from the level common under the informal funding system. In the Ho-tung salt administration, more than 100,000 taels had previously been collected from salt merchants both to supplement public expenses and to help pay officials' yang-lien and the wages of clerks, runners, and patrol soldiers. This sum was considered excessive and was lowered under the supervision of the salt controller.^[121] In Hukwang, the level of yen-kuei was reduced from 160,000 taels to slightly over 30,000 taels.^[122] Similar, if not as drastic, reductions were also made in Chihli and Shantung.^[123] Lack of evidence does not mean that reductions were not made elsewhere. However, whether or not the fees were lowered in every province, the method of remittance and management of these funds was probably altered everywhere in accordance with the funding guidelines established during the Yung-cheng reforms. It was these changes that had the greatest impact on local fiscal administration.

Prior to the huo-hao kuei-kung reforms, salt fees were sent by the merchants directly to the officials in whose jurisdiction they did business. Additional fees were also sent by groups of merchants to high provincial officials residing in the provincial capital. After the institution of yang-lien it was felt that officials had adequate funds for their own administrative expenses. Moreover, because the provinces now had public-expense funds in the provincial treasuries, the chou and hsien would no longer need supplemental funds with which to make contributions to provincial construction projects (kung-chüan ). Therefore, salt merchants were instructed to remit their yen-kuei directly to the provincial coffers, either by sending them to the financial commissioner or by submitting them to the provincial salt controller, who would handle their transfer to the provincial capital.^[124] Moreover, the rate at which salt fees were remitted was now fixed and no longer depended on the whim of the official to whom they were sent.

Salt fees and surplus revenues were used for various purposes, some of which were of special benefit to the salt merchants. The

huo-hao kuei-kung reforms encouraged the integration into the regular chain of bureaucratic command services that previously had been provided through the informal funding network. For example, Acting Liang-huai Salt Controller I-la-chi reported that under the old system the salt controller sent agents to the dikes at T'ai-chou and Kao-yu chou and to the north bridge at Yangchow to inspect the boats delivering salt. Each of the three agents was sent lou-kuei by the salt merchants, providing them with an annual income of some 6,000 taels each. According to I-la-chi, these funds all ended up in the pockets of the three agents. These agents being yamen runners who received a fixed wage already adequate to support their families, I-la-chi did not feel it was appropriate for them to pocket fees from the salt merchants as well. Moreover, the embankments and bridges already had official deputies (wei-kuan ) assigned to them. so it was unnecessary to send additional yamen servants to inspect the unloading process.^[125]

I-la-chi discontinued the deputation of the three yamen runners. However, he did not eliminate the lou-kuei customarily sent to them. The salt controller reasoned that the elimination of this 6,000 taels would mean only a small increase in the profits of individual salt merchants, whereas it represented a valuable addition to the operating expenses of the official yamen. Instead he proposed to continue its collection to cover the costs of patrol and seizure of smuggled salt. This would not only serve the interests of the merchants in controlling illegal competition in the sale of salt, but would relieve the merchants of the need to supply the government with irregular levies of supplemental aid.

Under this new system of patrol, officials, runners, and yamen servants would be posted at strategic places along the salttransportation route. A registrar, archivist, and saltwatcher from the salt controller's yamen would be used. Inasmuch as these officers already had salaries, no additional expenses would be incurred through their participation in the patrols. In addition, five low-ranking officers, including four probationary officials (shih-yung kuan ), would also be assigned to this duty. I-la-chi estimated that it would require 400 taels per month, or 5,000 taels annually, to pay additional official salaries and to hire the necessary patrol boats and runners. The remaining 1,000 taels would then be saved and be available if it became necessary to set up additional patrols.^[126] In this way the supervision of salt transport and sale would be systematically integrated into the regular control functions of yamen

officials. At the same time, the funding of these services would be rationalized and it would no longer be necessary to levy additional duties on the salt merchants, thus eliminating the pretext for random extortion by yamen personnel.

Customary fees from salt merchants were also used to pay for the wages and salaries of officials, clerks, and runners within the salt administration. The wages and salaries of officials and runners allowed by quota in the regulations of the Board of Revenue were deducted from the salt gabelle in much the same way that civil officials deducted their wages and salaries from the regular land and head tax. As in the case of the regular administrative apparatus, however, these allowances never covered the full costs of clerical wages, stationery, and supernumerary runners. Therefore, after the reforms the salt administration itself continued to collect a fixed amount of customary fees to cover its own administrative expenses.^[127] These fees were also used for social services such as banquets to honor leading salt merchants and to provide relief for saltproducing households that were victims of flood and other natural disasters.^[128]

Of course, salt fees, once returned to the public coffers, were not used exclusively for services pertaining to the salt administration itself. In many provinces they provided an important supplement to the revenues derived from the land and head tax. In particular, when the decree was issued calling for the allocation of yang-lien for minor officials, many provinces were compelled to turn to salt and customs administration lou-kuei for the needed funds.^[129] Salt fees also constituted an important part of the provincial surplus and were often used for such elastic expenses as the purchase of relief grain and the provision of welfare for famine victims.

Surplus from Customs and Commercial Duties

Determining the contribution of customs and commercial duties to the provincial coffers is far more difficult than assessing the impact of salt fees. By their very nature, customs revenues were more variable than other taxes, and throughout the late-imperial period control of these sources of income presented a challenge to the administrative skills of both the central and local governments. Because of the extreme fluctuations in customs revenues in any given year and at any particular collection point, logic would seem to dictate a policy of full remittance without setting specific quotas.

However, it was this very elasticity that compelled governments in the late-imperial period to set quotas for customs revenues fully recognizing that the quotas had little relationship to actual income.

A number of factors influenced the volume of trade at a particular customs station. Natural phenomena frequently disrupted commercial activity. The customs quota for the Lin-ch'ing station in Shantung was established during the Ming dynasty. However, spring flooding and winter freezing tended to limit the use of the Wen and Wei rivers leading to this customs station. Moreover, in the summer and fall, when ships were able to travel, the sluices were often closed on the route south, causing a backup of merchant ships and bringing transport to a standstill.^[130] Therefore, merchants preferred either to unload their goods from the south and carry them by cart for the remainder of the trip north^[131] or to bypass the Lin-ch'ing stop on the Grand Canal route and pass through Chi-ning and Tung-ch'ang instead.^[132] By the Yung-cheng period, only ships carrying grain from Honan and Shantung on route to Tientsin continued to pass through the Lin-ch'ing station. A particularly good harvest in Shantung and Honan, however, coupled with natural disaster in the chou and hsien on the border between Chihli and Shantung, could stimulate a continuous stream of merchants plying the route through Lin-ch'ing. This was the case during one year in the late Yung-cheng period, with the result that instead of falling short of its quota, the Lin-ch'ing station had surplus customs revenue of more than 40,000 taels.^[133]

In areas where trading was generally brisk, late rains and a poor harvest could reduce merchant activity by half, with disastrous results for individual customs stations.^[134] Even when harvests were good throughout the empire, price differentials between regions could completely upset the flow of the staples trade. The Huai-an customs station relied largely on the bean trade moving south for its revenues. In 1727 and 1728, harvests in Kiangnan were extremely good and prices fell. As a result, merchants could not be induced to come to Kiangnan to trade in beans, and customs quotas at Huai-an could not be met.^[135]

The flow of trade could also be affected by imperial policy, often in ways not anticipated when a decree was promulgated. The Yangtze River trade between Szechuan and Kiangsu relied largely on ships carrying textiles upriver to Szechuan and saffron and rice downriver to Kiangsu. About 50 to 60 percent of the taxes collected at the K'uei pass derived from these items. When the num-

ber of ships carrying cloth and saffron began to decrease, the K'uei superintendent of customs, Mu-ko-te-pu, investigated the matter and found that the reason lay in the emperor's prohibition against carrying Szechuan rice across its borders. According to the testimony of long-distance merchants, in the past all bolts of cloth shipped to Szechuan were collected at Chungking prefecture for sale to the villagers in the surrounding countryside. Rice was the main medium of exchange used to purchase the cloth. Since the prohibition against carrying rice out of the province had become effective, merchants were unwilling to trade for rice but demanded their payment in silver. Consequently, the cloth piled up unsold and less new cloth was moved into Szechuan, resulting in serious losses in customs revenues.^[136]

Given the unpredictability of those revenues, a dual system of remittances was established to guarantee minimum revenues to the central government and to allow enough flexibility to ensure that at least some of the income collected beyond the quota was also sent to coffers in the imperial capital. The regular quota (cheng-e ) was sent to the Board of Revenue quarterly. Collections beyond this quota were accumulated in the course of the year and remitted annually in a lump sum to the Imperial Household treasury.^[137] These excess revenues, called ying-yü , varied from year to year and were not considered in an official's evaluation of performance in office. This arrangement was designed to protect merchants from excessive exactions by officials competing to surpass their predecessor's collection record,^[138] besides providing a source of income for imperial coffers. Because one's career did not depend on accurate accounting of ying-yü income, a large portion of excess customs revenues was never reported to the central government and ended up in the pockets of local officials, runners, and clerks connected with the customs administration.^[139]

In conjunction with the reforms in the collection and distribution of land- and head-tax revenues, the Yung-cheng emperor set about overhauling the administration of customs income as well. Just as part of the intent of the huo-hao reforms was to protect the people from harsh official exactions, the reforms in the customs administration were designed to protect merchants from unjust surcharges. At the same time, distribution of the customs surplus was rationalized to guarantee the fullest possible remittance to the central government while satisfying local claims to supplemental customs income. Part of the requirements for such a

plan were met by the huo-hao kuei-kung reforms themselves. By providing provincial officials with adequate sources of revenues, it was hoped that the demands for lou-kuei from the customs administration would be reduced. Serious efforts were made in many places to reduce or eliminate the customs surcharges levied on long-distance merchants that had previously entered the informal funding network.^[140]

The efforts to regulate customs duties followed the same principles applied to taxes in the agricultural sector: uniformity, public announcement of rates, and improved bookkeeping techniques. In the sixth month of YC 4 (1726), Yung-cheng issued an edict calling on customs officials to collect and remit in full (chin-shou chin-chieh ) all surplus merchant duties at each customs station in the empire.^[141] In order to encourage officials to declare the actual revenue potential at their customs stations, a one-year deadline was allowed for such declarations and a moratorium was placed on repayment of previously undeclared surplus income.^[142]

Two factors stood in the way of central-government control of customs revenues: a complete disregard for the legal rates of collection at individual duty stations and concealment of customs income through duplicate bookkeeping. In order to eliminate the first obstacle, a decree was issued requiring all customs stations to collect duties according to the rates issued by the Board of Revenue. To ensure adherence to these rates, they were to be engraved on blocks and printed so that they might be clearly posted at each collection point, much in the same way that officials had been ordered to post the land-tax rates at key points in each chou and hsien.^[143]

The difficulty in enforcing compliance was apparent immediately. Although the official rates were expressed in terms of rates according to type of cargo, the volume of trade passing through certain customs points, coupled with a lack of staff, had forced some customs stations to collect duties on the basis of the capacity of the boat, without regard to the type of cargo. In the case of the Chiu-chiang customs in Kiangsi, this practice had been in effect as early as SC 2 (1645), when the customs station first opened. Initially there were no regulations governing collection, and the Board was satisfied so long as the quota of 99,000 taels was met. In SC 12 (1655), under the guidance of Customs Superintendent Fei-ta, each boat traveling the route that passed through the port at Chiu-chiang was given a registration number, and these numbers, along

with the boat's specifications, were sent to the Board. However, as the boats stopping at the port proliferated, it became increasingly difficult for the station's staff to measure and investigate the contents of each boat. Although the quota itself had since doubled, customs inspectors could do no more than rely on visual estimates of each boat's capacity, levying duties at a uniform rate of approximately 0.02 tael per picul of cargo.^[144]

The inefficiency of such a system had long been known to both provincial and court officials. In fact, all attempts at regulation had become futile. Boats varied in size, depth, and width. Some were old and some were newly built, and the capacity of these various types could not be judged at a cursory glance.^[145] In Shantung, where duties were also levied on the capacity of the boat, shippers renovated their boats to make the cargo hold deeper and wider, thereby deceiving the official inspectors, who could make only a perfunctory estimate of capacity from onshore.^[146] Moreover, although boats carried a variety of goods of vastly differing value, duties were applied uniformly. These anomalies had been reported to the court many times, but the fear that inadequate staff and confusion in levying customs duties would result in deficits in the regular quota prevented any change in the system of collections from being implemented.^[147]

This resistance to change reflected an aversion to risk shared by both provincial and court officials. Officials preferred to maintain an inefficient method of collections that had at least produced a reliable minimum quota, rather than switch to a more complex system that could produce greater average income for the government but could also result in greater problems in particular years. Instead, the government turned to improvements in bureaucratic procedure to protect merchants from excessive demands by customs staff and to provide the central government with better information about the actual volume of trade passing through each port. This was accomplished by issuing registers (hao-pu ) to each customs station on which were recorded the name of the duty station, the date, and the amount of silver collected as tax from each merchant. In order to avoid falsification of these records, the merchants were to fill out and sign these registers themselves. Moreover, in the manner of the three- and four-stub receipts used in the collection of the land and head tax, the customs inspector's portion of the "receipts in payment" (wan-shui ch'üan-ken ), issued

upon payment of the duties, was also sent along to the Board for comparison with the records in the hao-pu .^[148]

The need for improved accounting is vividly illustrated by the state of customs collection in Fukien. The province had sixteen customs stations in the early Ch'ing. Yet despite regulations requiring the documentation of all customs transactions, at year's end the quota for the whole province would simply be divided among all the stations and false records would be compiled under the auspices of the stations at Nan-t'ai and Hsia-men.^[149] Any surplus above the quota would then presumably be divided among the staff of all sixteen stations, with payments to civil officials in the province being made in the form of customary fees. As a result of the reforms, not only were accounting methods made more stringent, but variations in the rates of collection between and within customs stations in the province were made uniform, in order to simplify collection and ensure that the increase in prices as a result of taxation was shared more equitably by consumers throughout Fukien.

The combination of the huo-hao kuei-kung reforms and the order to eliminate corruption within the customs administration at the same time that efforts were being made to remit all surplus customs revenues in full to the imperial capital could not but cause some consternation and confusion at the local level. This was particularly true in those areas where the reforms were implemented early. As explained in chapter 2, apart from blatant falsification of customs revenues, many customs stations gained extra revenues by charging fees to cover the costs of customs administration and lou-kuei to superior officials. Because lou-kuei to superiors had been eliminated and officials had been provided with yang-lien , it was not considered unreasonable that customs duties now be collected according to the established Board rates. However, when the emperor issued his edict calling for the complete remittance of excess revenues (ying-yü ), many customs officials feared that without collection of surcharges their ying-yü remittances would drop and they would be accused of stealing. Therefore, in some areas officials continued to sanction the collection of duties at rates above those promulgated by the Board.^[150]

Yung-cheng's assurances that his edict on ying-yü was simply intended to guarantee complete remittance of legally collected duties seems to have assuaged some of the fears of officials in charge

of customs revenues. However, as in the case of salt fees, it soon became apparent that some provision had to be made to allow a share of customs revenues to continue to flow into the provincial coffers. The emperor's order to eliminate surcharges and fees within the customs administration was not met with unanimous approval. Some officials felt that additional levies on merchants were justifiable because merchants could afford them and had paid them willingly for as long as could be remembered.^[151] Although all agreed that hiding and pocketing customs revenues was not to be condoned, many provincial officials were concerned that the customs administration itself could not operate without some transfer of funds from the customs surplus to its own coffers. Ch'ang Pin, who had been a customs official himself, pointed out that runners at the Canton customs collected various inspection fees that were used to pay wages, transportation costs, and so on. In an impassioned plea he asked the emperor how the clerks and runners and other staff members of the customs administration could be expected to work on empty stomachs.^[152]

Ch'ang Pin was not alone in his concern. Superintendent of Fukien Customs Chun-t'ai argued that the number of employees in the customs administration went far beyond the statutory quota and that expenses accrued in running this agency were great. Only by allowing the regulated collection of long-standing customary fees for their support could one guard against the more serious problem of embezzlement and deficits.^[153] Kiangsi Governor Pu-lan-t'ai went into even greater detail in describing the expenses incurred by the customs administration in his province, expenses for which, he claimed, it was imperative to skim a certain amount off the revenues remitted to the central government. To cover Board fees and the transport of documents and taxes alone, several thousand taels were needed. These expenses were covered by a hao-hsien on the regular tax as well as a transport fee (chieh-fei ) deducted from the ying-yü collected at each station. To pay the living expenses of yamen runners, a fee called "list silver" (tan-yin ) was charged to each merchant submitting a list of the taxable goods he carried. Moreover, every month, 400 to 700 taels were accumulated because of both differences in weights between the official scales and the market scales and variations in the fineness of silver. This miscellaneous customs revenue (lin-kuan ling-shui ) was turned over to the supervising official at each customs station to pay for his daily living expenses, ink, paper, private secretaries' and clerks' salaries,

and other costs. Any surplus at the year's end was reported and submitted to the governor's yamen.^[154]

In order to meet the needs of customs administration, an arrangement was finally worked out whereby all customs expenses were deducted from the ying-yü before the latter was remitted to the imperial capital.^[155] These funds were used to pay the yang-lien of customs officials and the wages of their clerks, runners, and private secretaries, as well as administrative expenses and the costs of compiling accounts and remitting taxes. Only after all such deductions were made and reported to the emperor in a palace memorial were the remaining "excess revenues" sent to the Board to be turned over to the Imperial Household treasury. Because the ying-yü thus remitted represented not the total surplus of the customs administration, but only the saved portion after expenses, it was often called "saved surplus" (chieh-sheng ying-yü ).

The ying-yü total continued to represent both collections of fees beyond the Board's schedule of duties and increased income due to the growth of the volume of trade. In accordance with the spirit of the reforms, however, the amount of customary fees, too, were published so that merchants would know precisely what was expected of them. Moreover, instead of allowing the customs-administration staff to pocket fees personally, yamen expenses were budgeted from the excess revenues and monthly allowances were issued to support customs employees and curb corruption.^[156]

Despite the proscription against the acceptance of customary fees, customs fees (shui-kuei ), like salt fees, continued to play a role in the finances of the provinces. Following the reforms, fees that were previously paid to high provincial officials were returned to the provincial coffers to help cover the public expenses of the whole province. In many instances the first response of provincial governors to the reforms was to insist that because customs revenues rightly belonged to the central government, all former customary fees should be combined with the cheng-hsiang customs revenues and sent to Peking. In every case where such a request was made, the emperor refused to accept these revenues and authorized their use in the province of origin to supplement public expenses.

Yung-cheng's special edict to Acting Kiangsu Governor Ho T'ien-p'ei explains his reasoning on this matter. Ho reported that in addition to the maritime customs quota and excess revenues collected in Kiangsu, there was a surplus of 5,732 taels that did

not belong to anyone. Rather than cover up the existence of this surplus, Ho reported it to the emperor and proposed that it be sent to the capital. The emperor replied:^[157]

This kind of surplus lou-kuei has never been predictable. Although you have loyally and sincerely memorialized the facts, it is not appropriate for me to permit you to voluntarily send [this money to the capital]. You need only memorialize about funds like this and keep them temporarily on reserve in the provincial treasury in case there are local public expenses or the need for funds to reward [outstanding soldiers or officials]. These funds belong to you to memorialize about and use as you see fit.

Yung-cheng was consistent in his policy of allowing provincial officials to return customs lou-kuei to the provincial coffers. In Kiangsi, customs and salt fees were retained to provide minor officials with yang-lien .^[158] Kwangsi's revenues from the land and head tax were so low that in YC 5 (1727) the emperor issued a special edict granting over 15,000 taels from the province's customs surplus to supplement the costs of providing officials with yang-lien . When it was discovered that an additional 31,400 taels in surplus could be derived from Kwangsi's customs administration, this too was turned over to the governor, with instructions to divide it among the ranked and unranked officials in the province on the basis of the difficulty of their posts.^[159] Evidence also exists for the return of customs surplus to the provincial coffers in Shantung, Fukien, and Fengtien.^[160] In each case the basis for allowing retention of customary fees or surplus levies seems to have been that they were paid willingly by the merchants and did not interfere with the fulfillment of the quota due the central government.^[161]

Revenues from Miscellaneous Commercial Duties

Customs duties were not the only form of commercial taxes that was legitimized as a source of local revenue during the reforms. Although the evidence surrounding these sources of provincial income is less abundant, local commercial duties were also a focus of reform.

As already pointed out in the description of the implementation of huo-hao kuei-kung , taxes on the sale of property, unloading of goods, brokerage licenses and so on played an important part in

the revenues of several provinces. Like the land and head tax and customs and salt duties, local commercial taxes were legally the property of the central government, to be collected in full and remitted in full to the Board of Revenue. However, as we noted in our discussion of informal funding prior to the reforms, these so-called miscellaneous taxes (tsa-shui ) were the most difficult taxes for the central government to control. Because the volume of commercial transactions fluctuated from year to year, magistrates generally remitted only a customarily agreed-upon quota to the central government, retaining the surplus to cover quota shortages in future years and to pay for their own expenses and for contributions to the expenses of superior administrative units.^[162]

Throughout the early Ch'ing, as the volume of trade and transactions in the sale of land and buildings rose, the central government's ability to regulate trade and the remittance of taxes was further circumscribed. One indication of the extraordinary rise in local commercial activity was the increased issuance of brokerage licenses during the early-eighteenth century. Brokers were licensed middlemen who were particularly important in arbitrating prices and bringing together buyers and sellers in local periodic markets.^[163] In order to control the volume of local trade and ensure the payment of taxes and maintenance of a "fair" price structure, the central government set limits on the number of brokers who could be licensed in each province. In Hupei, the Board of Revenue set a quota in YC 4 (1726) establishing the number of licenses the financial commissioner could issue to each chou and hsien to certify brokers in their jurisdictions. Legally, new brokerage households (ya-hu ) could be authorized only by rescinding the license of one broker and substituting another. Nevertheless, according to a report by Governor Wei T'ing-chen, by YC 9 (1731) the number of licenses actually issued by the individual chou and hsien had increased tenfold over the quota set five years earlier.^[164] The revenue derived from the sale of licenses was small. Annual licensing fees ranged from 0.5 tael for a large firm (shang-hang ) to only 0.3 tael for a medium-sized (chung-hang ) one and 0.15 tael for a small firm (hsia-hang ). By selling an extra 1,000 licenses, the province might hope to increase its revenues by at most around 300 taels, hence it is likely that the violation of the quota reflected a real need to increase the number of brokers engaged in handling a growing trade, and was not a ploy to gain additional revenues. This fact was acknowledged by the Board of Revenue when it granted Gov-

ernor Wei's request not to hold subordinates to fixed quotas in issuing licenses and collecting commercial taxes.^[165]

Provinces could indeed generate a substantial surplus from the collection of taxes on commercial transactions. Inasmuch as this excess revenue was the result of legitimate tax collection and not the product of surcharges or monetary manipulation, it was not subject to the same proscriptions applied to lou-kuei . Instead of demanding a greater share of this surplus for the central government coffers, Yung-cheng allowed the provincial governments to retain surplus commercial taxes for their own use.

Although evidence is scanty, it is probable that commercial-tax surplus contributed some income to every province. For example, in YC 8 (1730) the Board of Revenue permitted Kiangsi province to retain part of its unloading tax (lo-ti shui ) and property-deed tax (t'ien-fang ch'i-shui ).^[166] It was in areas where revenues from huo-hao were lowest that income from commercial taxes had the greatest impact on the provincial kung-hsiang . As pointed out in chapter 4, both Kwangsi and Yunnan relied heavily on revenues from mining, salt, and surplus comercial duties.^[167] In Kwangtung, surplus revenues of at least 35,000 taels were collected from the unloading tax and an additional 48,448 taels from the property-deed tax.^[168] By YC 12 (1734), the annual surplus from these two duties alone had risen to more than 120,000 taels. Seventy thousand taels were earmarked for provincial yang-lien and to cover the costs of boat repairs, and the remaining 50,000 taels were stored in the provincial treasury for emergency expenses.^[169] The most dramatic example of the reliance on commercial taxes for the success of the reforms was in Kweichow. There, only 18 percent of the revenues returned to the provincial coffers derived from the land and head tax. A full 59 percent of all provincial revenues emanated from salt and commercial taxes retained by the province for its own use.^[170]

The reliance of these provinces on salt, customs, and miscellaneous duties both for emergency expenses and to supplement the income necessary to allocate yang-lien and kung-fei demonstrates that the role of "surplus" taxes and fees from the commercial sector was far more important than is indicated by statutory quotas. Although their total contribution to local treasury stores was still not as great as that derived from the land and head tax, only a small number of provinces could survive without an infusion of some nonagricultural income. In the land-tax-poor areas beyond

central China and the north China plain, commercial revenues were vital to the success of the huo-hao kuei-kung reforms. Unlike lou-kuei , which burdened the people and destroyed official morale, income from salt, customs, and commercial taxes took advantage of China's growing population and commercial activity to provide an expanding and legitimate source of local funds that neither violated dynastic law nor harmed the livelihood of the masses of common people.

6—
Obstacles to Reform: Low-Level Corruption and the Kiangnan Tax-Clearance Case

The Decision to treat treasury shortages as an institutional, rather than a normative, problem was an important step in the development of the modern Chinese state. The transformation of that which was private (ssu ) into that which properly belongs to the public realm (kung ) suggested a major reinterpretation of the relationship between government and the people. This notion applied not only to corruption, that insidious impulse toward personal aggrandizement at the expense of the dynasty. Essential to the reform movement was the concept of government responsibility for a wide range of public services which fiscal insolvency had previously made impossible. Moreover, the provision of adequate revenues at the lowest levels of the bureaucracy improved official morale by facilitating adherence to Confucian values and codes of administrative conduct.

The reaction of most provincial officials to the new fiscal order was one of considerable excitement and hope. Even the misgivings of magistrates seem to have waned over time. The initial success of huo-hao kuei-kung , however, could not conceal the harbingers of its demise. For the reforms to fulfill their promise as a mechanism for rational local-government finance, one crucial condition had to be met: taxes had to be paid. Despite the reliance of many provinces on nonagricultural income to fill their coffers, huo-hao was

still the primary source of local revenue. As a surcharge on the land and head tax, huo-hao 's role in fiscal administration was inextricably linked to the collection of the dynasty's regular tax quota.

The huo-hao kuei-kung reforms attacked the cause of deficits at the level of the bureaucracy, but the Ch'ing state had little power to affect the forces that undermined its finances outside the official hierarchy. The same informal networks of funding that had allowed the bureaucracy to operate before the Yung-cheng reforms had also nurtured the growth of corruption within local society. If free rein had not been given to yamen runners, clerks, and members of the local elite to wring additional revenues from the taxpaying population, it would have been impossible for magistrates to fuel the informal funding system that huo-hao kuei-kung replaced. Once these local arrangements were established, it was far more difficult to wipe them out.

Tax arrears, evasion, and corruption below the level of the bureaucracy continued to be part of the fabric of rural life. Nowhere were they a more serious problem than in the three Kiangnan provinces of Kiangsu, Kiangsi, and Anhui.

Measures to Control Arrears

As early as SC 15 (1658), the Kiangnan region was the focus of an imperially sponsored investigation of arrears. This early inquiry into popular tax resistance was prompted by Governor-general Lang T'ing-tso's report that accumulated unpaid taxes during the period from 1651 to 1656 had reached over 4 million taels.^[1] Soon after the death of the Shun-chih emperor, the Kiangnan region was again the focus of an imperial tax investigation, in the form of the famous Tsou-hsiao an of 1661.^[2] Despite the harsh punishment meted out to the Kiangnan gentry as a warning against tax evasion, these three provinces continued to be plagued by arrears throughout the K'ang-hsi reign. Although the people of this region were the beneficiaries of numerous imperial tax remissions issued during this period, the emperor felt compelled to single out Kiangsu taxpayers for a special exemption from unpaid taxes in 1712.^[3] This act of imperial grace was followed by two more remissions of arrears granted by the Yung-cheng emperor during his first year on the throne.^[4] Nevertheless, by YC 3 (1725), the bureaucracy of

Kiangsu was still trying to work out a formula for the repayment of outstanding taxes totalling 8,810,000 taels, almost three times the province's annual tax obligations.^[5]

Some effort was made to alleviate the cause of Kiangnan's arrears by lowering tax quotas in those areas subject to Ming fu-liang . During the first few years of the Yung-cheng period, however, the primary focus of official energies here, as elsewhere in the empire, was on the investigation and clearance of deficit cases. The analysis of people's arrears continued to be simplistic. The high level of min-ch'ien was largely attributed to the machinations of local magistrates, who were accused of falsifying their accounts in order to transfer newly paid taxes to cover old deficits. During late YC 4 and early YC 5, at least twenty-one chou and hsien magistrates in Kiangsi were removed from office on charges of embezzlement or accumulation of deficits, much of which was concealed in their accounts in the guise of people's arrears.^[6]

In part, the failure of provincial officials to rectify this problem lay in the difficulty of detecting arrears malpractices. As Acting Kiangsi Governor Yü Chu pointed out at the time, any internal investigation was almost useless. As soon as word got out that a prefect was about to carry out a p'an-ch'a of his subordinates' accounts, the chou and hsien magistrates in his jurisdiction would alter the liu-shui hung-pu in which the taxes collected each day were recorded. They would then alter the payment receipts (ch'uanken ) stored in the local fiscal secretary's office (hu-fang ) to match these falsifications. Only taxes already sent by the chou and hsien to the financial commissioner would be recorded as paid. The rest the magistrate would enter as min-ch'ien . According to Yü, even a thorough investigation at the village level would bring to light no misconduct, because the clerks in charge of tax collection in the villages (ching-ch'eng ) were in collusion with the yamen personnel.^[7]

Given the appalling state of fiscal administration in the Kiangnan region, it is no wonder that the Yung-cheng emperor chose it as the focus of his own investigation of people's arrears. By YC 6 (1728), the huo-hao kuei-kung reforms were being implemented with considerable success almost everywhere in the empire but here. In fact, it was the chaos in Kiangnan's tax administration that initially led the emperor to prohibit these three provinces from participating in the reforms. In that same year, Yung-cheng issued an edict to all high provincial officials that called attention to the possibility that the Achilles heel of Chinese fiscal admin-

istration now lay not at the bureaucratic level, but below, where the huo-hao kuei-kung reforms did not penetrate. Without immediate attention, the repercussions of this low-level corruption could cause the destruction of all that the emperor and his reforming administration had striven to accomplish.

Yung-cheng identified clerical engrossment (hsü-li chung-pao ) as the single most important factor in the spread of low-level corruption. Clerks had long been targets of official disdain and were frequently used as scapegoats for malfunctions in local administration. However, the emperor was not simply indulging in empty rhetoric. As he had always been during the implementation of his reforms, Yung-cheng was concerned to discover the manifestations and techniques of corruption. Only then, through structural changes in tax administration or improved administration of existing laws, could officials devise the means to root out the cause of this blight on the body politic.

The emperor outlined several methods that might be employed by clerks to embezzle the people's taxes. Among these were proxy remittance (pao-lan ), altering tax receipts (hsi-kai ch'uan-p'iao ), opening and stealing from the tax envelopes deposited in government tax chests, and charging fees to conceal the names of households owing back taxes. Yung-cheng felt that such practices persisted because the magistrates rarely took the time to inform the people of their individual tax liabilities and the amount of taxes outstanding under their names. Magistrates may not have intended to become conspirators in tax fraud. In order to guarantee the functioning of the old informal funding network, however, they had simply ignored government regulations designed to enable the individual to take full responsibility for his own tax payments. As a result, what the officials considered tax arrears the people might regard as taxes paid in full, while clerks persisted in their treachery, deceiving both.^[8]

As a first step in remedying clerical fraud, Yung-cheng ordered the chou and hsien to have the shu-shou^[9] in each hsiang and li submit a list of all taxes paid and in arrears within his jurisdiction. The magistrates were then to check these lists against their own records. They were instructed that if there were no discrepancies they were to apply their official seal to the lists and have them posted in the li or hsiang concerned so that every household in arrears would be informed of its exact liabilities. If the amounts posted did not agree with a taxpayer's own estimation of what he

had paid, he could bring his receipts to the yamen and register an accusation of clerical engrossment.^[10] In this way, the emperor hoped to prevent clerks from engaging in what he called wanton banditry in the guise of official business. It should be noted that once again the emperor placed emphasis on the direct relationship between the individual taxpayer and the government and on providing the taxpayer with full and public information to enable him to protect his interests against rapacious yamen personnel.^[11]

The chou and hsien magistrates themselves were not without culpability in the persistence of arrears. Because the accumulated arrears in the provinces were so large that the people would have difficulty paying them off, Yung-cheng had granted his subjects the privilege of paying their debt in installments. The period allowed for repayment averaged about ten years, depending on the amount owed and the distance of one's home from the capital. To the emperor this was an expression of his love for the people and should have given the people ample time to pay their debts to the government with ease. However, Yung-cheng had heard that unscrupulous magistrates in some jurisdictions had set up unofficial registers to record the amount of arrears collected each year. If the people paid more than the exact quota set for a particular year's installment, the magistrate remitted the quota and embezzled the rest. Because the amount called for was sent to the financial commissioner, no further investigation of the magistrate's accounts was required. Meanwhile, people who in their public-spiritedness had paid more than was necessary that year would be pressed for payment of their arrears as if they had never been paid off. It was to the people's despair over this practice that the emperor traced the rise in tax resistance and delayed payments. Again, the solution he proposed centered on the public posting of installment quotas and the amounts already paid by each household.^[12] Moreover, "good people" who paid more than the installment, rather than being penalized, were to be rewarded and their actions reported to the magistrate's superiors.

The emperor's edict on arrears set off a flurry of new tax investigations, the largest of which were carried out in Hukwang, Chekiang, and Kiangnan.^[13] However, it was in Kiangnan that the investigation of arrears developed into a full-scale tax-clearance case (ch'ing-ch'a ) that would last three years and involve not only all the officials of the provincial administration, but special agents of the central government as well.

The Origins of the Kiangnan Ch'ing-Ch'a

Following the emperor's edict calling for the investigation of arrears, Governor Chang Tan-lin submitted his own evaluation of the situation in Kiangsu. Most striking was the volume of taxes the people of this province owed to the government. The size of the Kiangsu debt was not simply a function of its high tax quota; it also reflected the exceptionally slow progress of arrears repayment. For example, in YC 4 (1726) only slightly over 11 percent of the outstanding taxes due that year had been remitted to the magistrate's yamen.^[14] In view of the liberal repayment policy established by the central government, such delinquency could only be the result of flaws in the operation of local tax administration.

According to Chang, the most important impediment to efficient tax collection in his province was the inadequacy of chou and hsien tax registers. The large size of their jurisdictions and the broad range of duties for which an individual magistrate was responsible made it difficult to verify the accuracy of the records on which tax collection was based. Omissions and falsifications in the registration of land could be found in every subordinate unit in the region. Sometimes several tens of households were merged under the name of one household. At other times the property of one wealthy household was registered under scores of different hu designations. In the former case the party really responsible for tax payment was impossible to detect. In the latter, arrears in large amounts were hidden by dispersing them among a myriad of fictional landowners. As a result, Kiangsu's taxes were rarely paid in full.

Although the inaccurate registration of land served the interests of Kiangsu's more prosperous inhabitants, the paucity of reliable tax data forced magistrates to relegate responsibility for fulfillment of government tax quotas to yamen and village clerks. Proxy remittance (pao-lan ) was therefore a common practice in Kiangsu. The imposition of a third party between the taxpayer and the government contributed to both arrears and their persistent default. Clerks and runners invariably pocketed a portion of the taxes consigned to them, recording them in the yamen accounts as people's arrears. In order to conceal their own embezzlement, sub-bureaucratic personnel were often instrumental in encouraging popular resistance to arrears repayment and in inciting antitax demonstrations.^[15]

Governor Chang's suspicions that the common people were not responsible for much of the tax indebtedness in his province was rekindled once the new drive to recover arrears was under way. Almost immediately, petitions began pouring in to the yamen of Kiangsu's magistrates, expressing concern over the effect that the investigation would have on the region's poor. According to reports transmitted to the governor's office, the petitioners claimed that most of the outstanding taxes in Kiangsu were owed on land that had long since been sold by needy proprietors. Utilization of the land had passed to new owners, but the liability for past arrears remained with the household that had originally incurred them. Worried that these people would not be able to pay the government what they owed, the petitioners requested that all arrears be divided equally among the present landholding population.

Similarities in the content of requests coming in from all over the province raised doubts as to the spontaneous nature of the petition campaign. Inquiries into the identity of these "public-spirited" subjects convinced the governor that it was in fact households who controlled the pao-lan networks that were behind the move to redistribute the debt. Inasmuch as these households themselves were responsible for a large part of the unpaid tax quota, a redistribution of accumulated arrears would serve to transfer their own liabilities onto the shoulders of others. However, Chang had a more important reason for rejecting the plan. Even if all the households in Kiangsu complied with the reapportionment of existing arrears, such a move would do nothing to root out the causes of pao-lan and tax evasion. Within a few years the problem of unpaid taxes would have to be confronted again. Only a thorough ch'ing-ch'a could rectify the system of tax collection and eliminate the source of abuses once and for all.

Chang Tan-lin was familiar with the methods used to carry out a ch'ing-ch'a because his cousin, Chang Tan-jang, had just assisted in such an investigation in nearby Chekiang. Armed with his cousin's advice, the governor memorialized the emperor to carry out a full-scale tax audit in Kiangsu as well. What distinguished Chang's proposed ch'ing-ch'a from a routine investigation of provincial fiscal administration was the role assigned to officials outside the provincial bureaucracy. Normally it was the duty of assistant and subassistant prefects to inspect the tax records of subordinate chou and hsien. Ostensibly to avoid overburdening these provincial officials, Chang requested that the Board of Civil Appoint-

ments and the top officials in the province be authorized to select officials not yet assigned to regular posts to take on this responsibility. Because Suchow, Sung-chiang, and Ch'ang-chou prefectures had the greatest deficits and arrears, the governor suggested that four officials be assigned to each. The number of special investigators sent to the other prefectures in the province could be fewer, depending on the size of their debt and the problems anticipated in their recovery. Altogether, Chang estimated that twenty-four "expectant" officials would be necessary to investigate the tax-collection records and procedures of all the prefectures in the province and assist in the compilation of new, accurate tax registers to avoid future malfeasance. In addition, the governor proposed that once the task was completed these men be kept on in the province to help carry out a much-needed land survey to guarantee that the government's information on landholdings was indeed up-to-date.^[16] Chang Tan-lin was transferred to Kiangsi before he had a chance to test his plan in Kiangsu, but his proposal was quickly taken up by the new governor, a young Manchu bannerman named Yin-chi-shan.

The task facing Yin-chi-shan was even more difficult than his predecessor had anticipated. Not only was pressure being placed on the provincial government to shift the burden of arrears to innocent taxpayers, but the emperor's call for stricter prompting of payment of old debts soon gave rise to overt resistance to government tax collectors in the countryside. As early as the second month of YC 6 the magistrate of Chia-ting hsien had encountered resistance to the payment of old arrears. In this case, "evil people" had called the inhabitants of one village together by beating a gong and incited them to cut down the suspension bridge in order to deny the tax collector access to the village.^[17] The reaction to the tightening of sanctions against tax defaulters in Wu-hsi hsien was even more hostile. In the sixth month of the same year, Wu-hsi magistrate Chiang Ch'ang-jung escalated his campaign to collect past taxes by making the heads of all households with arrears wear the cangue. Soon after this order was issued, a group of "evil people" rioted in the hsien yamen and released the prisoners being held in the hsien jail. As a final act of defiance, they even burned the government's boats to prevent the hsien police from pursuing them.^[18]

Not all magistrates were as diligent as those of Chia-ting and Wu-hsi. Some feared that the provincial government's determina-

tion to retrieve old tax debts would reflect poorly on their own career evaluations. Thus, in spite of good fall harvests throughout the province, the government began to receive reports of crop devastation due to floods from a number of hsien, including An-tung, T'an-yuan, and Shu-yang. Subsequent investigations disclosed only minor damage from rains in a few isolated low-lying villages.^[19] Although the governor did order the financial commissioner to provide flood relief for the few who were actually affected, it was clear that the magistrates in these hsien, unable to collect their full quota of arrears, had exaggerated reports of natural disaster in the hope of receiving an imperial tax exemption for their jurisdictions. The governor felt that they too were under the influence of locally powerful "evil people."

Who were these "evil people"? As is so often the case in official communications, the perpetrators of corruption and lawlessness were described in such vague terms as to include almost anyone, save perhaps "simple, ignorant peasants" and the local officials themselves. This ambiguity was in part deliberate, for the roots of corruption extended so deeply into the Kiangnan tax administration that without a thorough and honest investigation, the officials of the province could not know exactly who the "evil people" were or how they carried out their clever deceptions. However, even before the investigation had begun, an increasing number of officials had come to believe that the gentry, the mainstay of the imperial state, could not escape culpability in the sabotage of local fiscal management.

Chang Tan-lin had not been the only one to suspect that the chronic arrears in the region were not entirely an outgrowth of popular tax resistance. Wang Chin, an aging member of the gentry from Chiang-ning prefecture in Kiangsu, reported to the emperor that the fault lay not in the chou and hsien magistrates' lack of skill in collecting taxes, but in the abuses of gentry and members of large households (shen-chin ta-hu ). During his term as a magistrate in both Hunan and Chekiang, Wang found that the "little people" who did not own much land tended to pay their taxes promptly. It was the gentry and large households, with vast holdings, who attempted to evade taxation by registering their land under fictitious names. However, without the connivance of clerks and runners, complete concealment would not have been possible.^[20] Such "evil people" delayed remittance of their taxes in the expectation that an imperial tax remission would eventually re-

lieve them of any obligation to repay arrears. From the ample evidence of such imperial acts of grace we can see that they were not wrong in assuming that the worse arrears became, the greater the likelihood that a remission would be granted.

Fa-hai, who was governor of Chekiang during the first half of the Yung-cheng reign, had already urged the emperor to investigate the illegal activities of Kiangnan's local degree-holders. He also attributed arrears to proxy remittance and tax resistance by gentry and local "ruffians," as well as embezzlement by yamen and village clerks.^[21] Fa-hai felt that few officials understood the situation and praised Financial Commissioner Oertai and the Fan-t'u and Wu-hsi magistrates as the only members of the Kiangnan bureaucracy who wholeheartedly attacked the problem of arrears and neither pandered to the will of powerful local families nor acted with excessive leniency in order to win over the people's affections.

Some of the officials who were eventually charged with carrying out the Kiangnan ch'ing-ch'a were also sensitive to the fact that other kinds of malfeasance were often disguised as people's arrears. This was particularly true of those officials already in Kiangnan, engaged in the earlier investigation of government deficits. Among these men was the provincial censor, I-la-chi. In order to uncover the actual amount of deficits in areas like Suchow and Sung-chiang, I-la-chi and his fellow investigators had ordered each prefecture, chou, and hsien to compile an account of people's arrears. Among the items listed as min-ch'ien , I-la-chi had discovered many that could be traced to embezzlement by officials and their staff. However, he also contended that the vast majority of genuine arrears belonged not to the common people, but to the gentry. Moreover, of these gentry offenders, most were members of families that owned extensive fields. Tax delays by this group were clearly not due to financial distress, but were the result of deliberate tax resistance (yu-i k'ang-ch'ien ).^[22]

It is little wonder then that Governor Yin-chi-shan came to share his predecessor's suspicions of the so-called spokesmen of the people who clamored for the redistribution of arrears of taxes. Even after the latter's rejection of their proposal, pressure for such a move did not abate. A joint petition was sent to Yin-chi-shan by inhabitants of the chou and hsien of Sung-chiang prefecture, and requests from individual chou and hsien continued to pour in as well. Groups of literati and commoners even carne to the gover-

nor's yamen to enjoin him to submit a memorial on the matter. The persistence of these groups only served to confirm in the governor's mind what earlier investigators had suggested, that very few of the arrears on record could really be blamed on poor people who did not have the means to pay. "The remainder," Yin-chishan insisted, "are the result of tax resistance by large households (ta-hu chih k'ang-wan ), proxy remittance by village clerks (t'u-shu chih pao-lan ), and embezzlement by yamen runners and clerks who tamper with the accounts, dividing embezzled taxes among many households and falsely registering land ownership (hua-fen kuei-chi, i-ts'e kai-ming )."^[23] The claim that all these arrears were wu-cho , without anyone to whom responsibility could be assigned, would hold true only if they arose from the impoverishment of former landed households. If they were the result of personal aggrandizement by powerful members of the local community, then responsibility could be assigned and the funds recovered.

The outcome of Yin-chi-shan's own inquiry into the identity of the petitioners was similar to that of Chang Tan-lin. Many turned out to be taxpayers with large accumulated arrears, who stood to profit from the redistribution of unpaid debts.^[24] Others, whom the governor labeled local ruffians, evil gentry, wicked runners, and traitorous clerks, hoped to halt any investigation that might interfere with their control of the tax-collection process. Although a large number of local officials were seduced by the simplicity of the petitioners' plan to eliminate arrears, particularly as some of the people's debts were decades old, the governor's mind was made up. A ch'ing-ch'a was essential.

The Ch'ing-Ch'a Begins

The Kiangnan ch'ing-ch'a , as it finally developed under Yin-chishan's supervision, came to include at least seventy officials especially chosen on the basis of their integrity and experience in tax matters (ching-yü ch'ien-liang ).^[25] Particular emphasis was placed on separating the investigation from the normal functions of the bureaucracy in order to guarantee that those engaged in the review of local tax practices would be free of any ties that might prejudice their findings and would be able to devote all of their time to the task at hand. Three levels of officials from outside Kiangnan's official hierarchy were assigned to the investigation. At the top, the

emperor appointed two imperial commissioners to superintend the ch'ing-ch'a of taxes (ch'in-ming tsung-li ch'ing-ch'a ch'ienliang ), Board of Works Vice-president Ma-erh-t'ai and Censor An Hsiu-te.^[26] Below them were nine imperially appointed branch investigators (fen-ch'a ta-ch'en ), amongst whom the various jurisdictions in Kiangnan were divided.^[27] These fen-ch'a officials came mainly from the ranks of circuit intendants and were in charge of the everyday supervision of the investigation at the local level.^[28] At the bottom of the ch'ing-ch'a hierarchy were the hsieh-ch'a officials. These men were chosen from the ranks of minor officials, expectant magistrates, and assistant prefects and subprefects.^[29] Officers showing promise in this capacity could hope to receive promotion to more permanent positions in the regular bureaucracy.

Although the greatest volume of arrears originated in Kiangsu province, the ch'ing-ch'a was designed from the start to deal with Kiangsi and Anhui as well. Officials assigned to Chekiang were also consulted on occasion, because it was recognized that many of the problems affecting Kiangsu were prevalent throughout the region.^[30] Thus, although most of the extant documentation refers specifically to Kiangsu, we can assume that the general procedures and many of the findings were typical of Kiangnan as a whole. Because the investigation covered so large an area, centralized supervision was tempered by a recognition of the need to grant inspectors flexibility in their actions at the local level.^[31] Moreover, high-level provincial and ch'ing-ch'a officials divided up the territory in order to facilitate closer supervision of fen-ch'a officials and coordination of all investigative efforts. All those assigned to the ch'ing-ch'a were instructed by the emperor to meet often in order to share information and help each other with the development of efficient auditing techniques.

During the initial stages of the ch'ing-ch'a , the main concern was not to recover arrears, but to determine their cause and designate responsibility for future repayment. The emperor even issued a two-year injunction against the collection of back taxes, after which time all debts were to be cleared and the investigators were to certify the accuracy of all tax registers in the chou and hsien.^[32] It soon became apparent that even this would require more than the allotted time and manpower. Therefore, the commissioners concentrated first on clearing up old arrears and investigating corrupt practices. Verification of tax registers was begun, but it was recognized that a complete cadastral survey would be necessary and

could not be carried out with the existing complement of ch'ing-ch'a personnel.

The first task of officials sent to the chou and hsien was to determine what portion of arrears really belonged to the people and what portion could be categorized as officials' embezzlement (kuan-ch'in ) or runner and clerk engrossment (i-shih ).^[33] In order to encourage cooperation with the investigators, the emperor issued an edict suspending punishment of any embezzler or tax farmer who confessed his crime and agreed to pay back the funds stolen from the government and the people (chun pao-lan ch'in-shih chih-jen tzu-shou mien-tsui ).^[34] Repayment in installments was permitted in order to avoid forcing the offenders to resort once again to extortion of the common people. In addition, Yung-cheng expanded his previous suspension of arrears collections to include all taxes in the region. In this way, the emperor hoped that the collection of regular taxes would not interfere with the investigations that were under way.^[35] The curtailment of compulsory debt repayment did not mean that the government ceased its recovery of back taxes. In fact, tax chests were set up in each chou and hsien and the people were encouraged to deposit in them as much as they were able toward paying off their own arrears. At the same time, a bureau was established in each chou and hsien to investigate the sources of arrears and to compile a "clear account" of all unpaid taxes, classified by household. Once again, there was widespread publicity and the posting of the emperor's instructions in every village so as to let the people know what their obligations were.^[36] The ch'ing-ch'a officials also encouraged households with arrears to go to the bureaux in person to verify the amounts under their names and to correct any mistakes in the records,^[37] presumably by presenting receipts for taxes already paid. In the process of compiling "clear accounts," local officials were also enjoined to determine the correct name and residence of each taxpaying household to facilitate a house-by-house investigation of tax liabilities at a later date.^[38] By midyear it was the opinion of most officials involved in the ch'ing-ch'a that to a large extent the arrears were really the product of clerical engrossment, the most difficult category of debts to uncover. For the moment, however, it was hoped that by verifying people's arrears and encouraging officials to confess to embezzlement, funds belonging to the third category would automatically become visible (shui-lo shih-ch'u ).

Despite these elaborate preparations, the investigation of ar-

rears did not go as smoothly as anticipated. In some cases the choice between testing the government's ability to carry out its investigation and avoiding punishment by the confession of one's crimes did result in offenders in all three categories coming forward. Wen Erh-sun, who was in charge of arrears investigations in T'ai-ts'ang chou, reported a fair response by the middle of YC 7 (1729). Arrears in Ch'ung-ming hsien were so small that the task of recovering them was left entirely in the hands of the magistrate. In the remaining chou and hsien, Wen reported the following results:^[39]

However, these preliminary responses accounted for only a portion of all arrears and were not accompanied by large voluntary repayments. It remained to be seen whether those who confessed to responsibility for arrears would reimburse the government, and whether the majority of unaccounted-for arrears could be tracked down.

Trouble Within the Ranks

In this endeavor, the ch'ing-ch'a apparatus encountered immediate obstacles. One was at the highest level of the investigative hierarchy. The other penetrated to the lowest levels of the provincial subbureaucracy. For such a large-scale ch'ing-ch'a to succeed, it was necessary to have the complete cooperation of local officials, and for those in supervisory positions to consult and assist each other in providing information and establishing rules for the inspection and retrieval of funds. However, signs of rivalry and outright dishonesty appeared within the ranks of tsung-li and fen-ch'a officials as soon as the ch'ing-ch'a had begun.

Upon his arrival in Kiangnan, Ma-erh-t'ai found an air of hostility growing among ch'ing-ch'a officials already in the province.

Although all actions were to be decided on in concert, Ma-erh-t'ai reported that Financial Commissioner Chao Hsiang-kuei had taken all responsibility for directing the investigation upon himself, refusing to discuss proposals with the Board officials sent to supervise the ch'ing-ch'a or with the branch investigators. Resentment over this usurpation of power was so strong that the special investigators simply ignored Chao's orders and abandoned any effort at joint planning. This created serious problems for subordinates involved in the audit of taxes, who were constantly faced with conflicting instructions. According to Ma-erh-t'ai, petty infighting of this kind had set the investigation back at least six months.^[40]

Ma-erh-t'ai's report to the emperor confirmed the rumblings that had already been emanating from the ranks of the fen-ch'a officials themselves. Soon after his assignment to the superintendency of the Kiangnan ch'ing-ch'a , Wang Chi was appointed acting governor at Suchow. When Censor I-la-chi was ordered to assist in the investigation of debts in Kiangsu as well as those in Anhui, Governor-general Fan Shih-i instructed him to coordinate his actions with Governors Yin-chi-shan and Wang Chi. They, in turn, would handle all communications with the Board of Revenue. However, because Yin-chi-shan was temporarily acting superintendent of river conservancy and engaged in a crucial construction project, the work of the ch'ing-ch'a was left entirely to Wang Chi. According to I-la-chi, Wang did all he could to subvert his position and authority. The acting Suchow governor failed to include I-la-chi in the drafting of notices regarding the audit of people's arrears and even drafted memorials to which I-la-chi's name was appended without consulting I-la-chi on their contents. Moreover, when I-la-chi's name was included on official papers, his post and rank were not mentioned.^[41] It was only after Board of Civil Appointments Vice-president P'eng Wei-hsin showed him a particularly stern imperial rescript appended to one of these memorials that I-la-chi even learned of Wang's obstructionist activities.

I-la-chi felt that his role as an imperially appointed special investigator was being challenged by a conspiracy within the regular provincial bureaucracy. Although the task of identifying deficits and arrears belonged to I-la-chi and the other ch'ing-ch'a officials and their staffs, recovery of funds was the duty of regular local officials. Every day that recovery was delayed increased the risk that the property of a guilty party would be spent or hidden so that arrears would never be repaid. Although 600,000 taels had al-

ready been retrieved from over 170 cases, I-la-chi calculated that there were still more than 600 cases in the financial commissioners' files pertaining to embezzlement by officials alone. Total revenues still unaccounted for came to more than 5,300,000 taels. Yet, despite direct imperial orders to hasten the recovery of back taxes, Financial Commissioner Chao Hsiang-kuei, with the support of Governor Wang Chi, had given the matter only perfunctory attention (pu-kuo i-fan chi-wan ).^[42] Inasmuch as the governor and financial commissioner were unwilling to devote any energy to the matter, it was only natural that officials in charge of recovering these funds treated their assignments as a mere formality.

I-la-chi feared a total breakdown in communications between local officials and imperially commissioned investigators, which would destroy any hope of the ch'ing-ch'a's success. In a display of frustration not uncommon in secret communications to the emperor, I-la-chi placed the blame on his own incompetence as an official.^[43]

I am really useless in Kiangsu. My Chinese is inadequate. Therefore, the memorials I write cannot be clear and detailed. I respectfully beg the emperor to extend his grace and send down an edict bringing me back to the capital so that I may kneel and memorialize the emperor on the situation I witnessed in the provinces and the various impediments [to my freedom of action]. Then send me to the Board for deliberation [of my case] and punishment as a warning to those who are ordered to carry out a mission and prove to have no ability. I will close my eyes and depart this world in peace.

Submission of such an appeal, whining and childish as it may seem, was a technique employed fairly often by officials caught in a bureaucratic bind. Although I-la-chi did risk being called back to the capital, he probably would not have written such a memorial if he had not believed that the emperor had confidence in him. If such was the case, the memorial would have the effect of forcing the emperor to take some action to confirm I-la-chi's position.

I-la-chi's boldness may have derived from his knowledge that he was not alone in his suspicions of Chao Hsiang-kuei and Wang Chi. At the same time that he submitted his memorial, an even more scathing indictment of the two leaders of the provincial bureaucracy was delivered by the T'ai-ts'ang fen-ch'a official Wen

Erh-sun.^[44] Wen was also under the impression that all tsung-li and fen-ch'a officials were supposed to meet and consult on the implementation of the ch'ing-ch'a . When he arrived in Kiangnan, however, he was told that an argument was going on between Governor Wang Chi and the fen-ch'a official in charge of Huai-an prefecture, Ch'ien Chao-yuan. As a result, no joint meetings of investigative personnel were held and the fen-ch'a officials simply went off to their assigned areas. When Wen visited the governor, he admonished him on this issue but did not receive any satisfaction. Later, Wen met with the tsung-li officials sent by the emperor and was told that without consulting them, Financial Commissioner Chao had issued a detailed set of orders to all prefects and chou and hsien magistrates. Wen was as concerned as I-la-chi that communication not break down between the ch'ing-ch'a officials and the local bureaucracy. The tsung-li officials having already ordered their subordinate hsieh-ch'a officials to compile registers of arrears (ch'ien-ts'e ), Wen suggested that when these were completed and household-by-household investigations were begun, the magistrates be required to come to the capital to consult with the special investigators.

Although Wang Chi agreed to support Wen's efforts to foster cooperation between hsieh-ch'a officials and the local bureaucracy, his own cooperation with investigators on the provincial level was not forthcoming. Rather than aid Wen Erh-sun in familiarizing himself with the situation in Kiangsu, Wang left the chore to P'eng Wei-hsin and I-la-chi, who were only vaguely acquainted with local conditions. After reaching T'ai-ts'ang, Wang heard that there had been a Board communiqué ordering ch'ing-ch'a officials to ferret out deficits as well as arrears. Not only was a copy of this document never sent to Wen, but when he went to Suchow in person and pressed Wang Chi and Chao Hsiang-kuei on the matter, they kept him waiting ten days before refusing to show him the order in question. In the end Wen could get the information he needed only by transcribing a copy of the document in P'eng Weihsin's yamen and then going to I-la-chi's office to obtain the relevant figures for T'ai-ts'ang chou.

Wen suspected that the behavior of the governor and financial commissioner was motivated by a desire to protect their friends in the officialdom. For example, among the cases being handled by Wen was that of former Grand Secretary Wang Yen, who, along

with his son, uncle, and nephew, owed the government 46,000 taels in back taxes. Because the records of the case had not yet arrived in T'ai-ts'ang, Wen was unable to take action to recover the funds. To make matters worse, Wen discovered that the hsieh-ch'a official, Sun Chien-lung, sent by Financial Commissioner Chao to assist Wen in inspecting T'ai-ts'ang's granaries and treasuries, was a disciple of Wang Yen. Wen strenuously pushed for Sun's transfer, but to no avail.^[45] In the end, investigations by ch'ing-ch'a officials uncovered evidence of the private use of government funds by both Financial Commissioner Chao and Hsü Yung-yu, an official who had held posts as both a hsien magistrate and a prefect in the province. Wang Chi was also cited for misuse of his power as governor and for nepotism in appointing his cousin Ning Shu-yeh and his family disciple Sun Chien-lung to ch'ing-ch'a posts. These two officials were removed from office and an investigation was begun to determine which of the other officers recommended by Wang and Hsü deserved dismissal.^[46]

The inquiry into Chao Hsiang-kuei's and Wang Chi's activities in office brought to light other crimes as well. Both were found guilty of embezzlement and collusion with yamen personnel to collect customary fees (lou-kuei ) in the form of gate fees extracted from lower provincial officials. This type of lou-kuei was vigorously attacked after the implementation of huo-hao kuei-kung because it drained local resources and fostered a corrupt dependency relationship between officials at different levels of the bureaucracy. Both the Chen-yang magistrate, T'ao Shih-ch'i, and the acting Hsin-yang magistrate, Chu Kuei-yang, reported that every time they had business with Wang they were forced to pay ten to twelve taels before being allowed past the yamen entrance. A local jail warden stated that he had been forced to pay identical sums to members of Wang's staff on at least thirty different occasions.^[47] When we consider that this amounted to over 300 taels, the impact on the jail warden's own ability to deal honestly with the local people is obvious.

Both Governor Wang and Financial Commissioner Chao were removed from office, eliminating an important stumbling block in the way of local official cooperation with the ch'ing-ch'a . However, this was not the only impediment to be overcome. The more covert and dispersed obstructionist activities of local clerks, runners, and gentry would prove far more difficult to counteract.

Subbureaucratic Subversion

The elaborate methods utilized by yamen functionaries and members of the local elite to subvert the collection of arrears had been established long before the ch'ing-ch'a apparatus descended on Kiangnan, and the question of whether even so intense an investigation could deal with them was not an idle one. Some clerical malfeasance took the form of overt embezzlement and was relatively easy to detect, once avenues of communication were opened up between the government and the taxpayers. For example, in one hsien in Sung-chiang prefecture an expectant subprefect named Chu had unpaid taxes totaling over 900 taels. As part of the ch'ing-ch'a , the hsien magistrate issued a summons for him to be investigated (ch'u-p'iao t'i-chiu ). Fearing punishment, Chu immediately volunteered to pay his arrears in full, whereupon the magistrate sent a hsien clerk named Chin Yung-lin to compel him to do so speedily.

This took place in the sixth month, before the regular tax collection season. Chin Yung-lin took advantage of the fact that the locked tax chests had not yet been set up to swindle Chu out of his 900 taels. It was only after the magistrate reported to the prefect for the second time on Chu's intention to pay his debt that he discovered that the funds had already been pocketed by Chin Yung-lin. The magistrate, fearful that his own reputation would be tarnished if the money was not remitted to the central government, pressed Chu for payment of his arrears again (ch'ung-wan ). In a rage, Chu went to Chin Yung-lin and tried to force him to return the money, but could not get him to turn over a single tael. Under normal circumstances, Chu might never have retrieved his stolen remittance, but because the province was in the midst of an investigation of just such malpractices, he was able to obtain immediate redress at a higher level. The fen-ch'a official Feng Ching-hsia had Chin Yung-lin arrested and obtained a confession not only to the theft of Chu's money, but to that of over 2,000 taels from other taxpayers as well. The ch'ing-ch'a superintendent, Ma-erh-t'ai, had Chin turned over to the new financial commissioner for further interrogation and also ordered the investigation of all local officials guilty of "failure to investigate" (shih-ch'a ) the misconduct of their subordinates, with a view toward future impeachments.^[48]

In fact, the ch'ing-ch'a officials appear to have been quite successful in uncovering cases of overt embezzlement by individual

members of the local bureaucracy. During the second half of YC 7, I-la-chi reported on the progress of the investigation of such cases in the prefectures of Suchow, Sung-chiang, Ch'ang-chou, and T'aits'ang-chou.^[49] Under the classification of official embezzlement (kuan-ch'in ), it was discovered that as the magistrate of Li-shui hsien, Chao Hsiang-kuei had stolen several thousand piculs of rice which he disguised as having been used for famine relief. In K'unshan hsien, Wang Su-wei and others had embezzled more than 2,000 taels obtained from the sale of firewood on government land (kuan-ti ch'ai-chia ). Former Hua-ting magistrate Lin K'uang pocketed 8,600 taels which he falsely reported as arrears, and another 4,200 taels were embezzled by the Wu-chin magistrate, Ch'en Shen, using the same method. All of these officials confessed to their crimes under the pressure of the ch'ing-ch'a investigation.

A number of cases of runner, clerk, and gentry embezzlement were also discovered through the investigative efforts of the ch'ing-ch'a officials. A few examples may be cited here. In Lou hsien, a tax-collection clerk was accused of "cheating the people out of their taxes" (k'uang-p'ien liang-hu yin ).^[50] A chien-sheng named Ch'in Pen-heng in Wu-hsi was found to have contracted to pay people's taxes, embezzling their remittances by issuing false tax notification receipts (chia-ch'uan pao-ch'in ch'ien-liang ). Ch'in was deprived of his chien-sheng status and his case was turned over to the financial and judicial commissioners for investigation and sentencing.^[51] A similar case was uncovered by a hsieh-ch'a officer in cooperation with the Shang-hai hsien magistrate. Here, a chien-sheng named Fan was in league with the local tithing chief (chia-shou ) to engage in proxy remittance and the registration of landholdings under fictitious names (hua-fen kuei-hu ).^[52] In P'ei chou and Sui-ning hsien, hsieh-ch'a officers reported that the head yamen clerks had conspired with the local li-shou to steal government funds. Investigation into areas exempted from arrears payments because of flood damage to crops disclosed that in at least four villages no reduction in harvests had occurred. The collection of arrears had continued and the revenues were pocketed by the clerks themselves.^[53]

These isolated successes began to pale in the face of revelations regarding the depth to which clerical malfeasance had penetrated local tax administration. The most shocking discovery occurred by accident. One of the most effective and hardworking of the fen-ch'a officials was a man named Feng Ching-hsia. It was under

Feng's supervision that many of the above crimes were discovered. In mid-YC 7, information reached Feng's office that a ch'ü clerk in Wu-hsi hsien named Li Ming-ju had embezzled a large amount of tax revenue. Feng ordered him to produce his accounts. Quite unexpectedly, a commoner named Yang Shih-yü, who had been hired by Li Ming-ju to do his paperwork, came forward with all the private records (ssu-shou ti-pu ) compiled during Li's fifteen years in charge of tax collection in T'ien-shang ch'ü . In all they came to 200 volumes, which Feng had Yang Shih-yü deliver to Suchow. In contrast to the registers in the magistrate's yamen, the clerk's accounts contained a complete record of the taxpayers in the area, the land they owned, the taxes due on each holding, and how much was paid and in arrears. However, after two days of careful scrutiny, Feng and Yang discovered that Li Ming-ju was able to manipulate the tax-collection process by an inconspicuous system of secret marks placed beside each entry.^[54] In effect, this code provided the clerk with a duplicate set of books so that information about his own activities could be concealed from the magistrate. Only by decoding these accounts could anyone begin to unravel the mystery of who really owed the government money and how the embezzlement of taxes had been accomplished.^[55] When Feng asked Yang Shih-yü if the other ch'ü clerks also used such books, Yang replied that not only did they exist in the fifteen ch'ü in Wuhsi, but there was not a prefecture, chou, or hsien in the region in which ti-pu were not compiled. Clerks would even sell these registers, presumably to incoming clerks, in order to enable them to continue the corrupt practices of their predecessors.^[56] The names applied to these registers varied and the codes were not the same, but ti-pu were an indispensable tool for the clerk wishing to profit from his post.

Armed with this information from Feng Ching-hsia, I-la-chi met with Ma-erh-t'ai and decided to call a meeting of all nearby ch'ing-ch'a officials to be held in the governor's yamen. The urgency of confiscating all such books was impressed upon them, and written instructions to that effect were sent out to all other local and ch'ing-ch'a officials. Before long, ti-pu were confiscated in Suchow, Sung-chiang, and Ch'ang-chou prefectures, and the hsieh-ch'a set to work decoding the secret markings, volume by volume, so that embezzled items and manipulation of the tax rolls could be exposed.^[57] By the tenth month of YC 7, Ma-erh-t'ai reported that 60 to 70 percent of the ti-pu in Suchow, Sung-chiang, Chen-chiang, and Chiang-ning prefectures had been seized.^[58] As

each case of embezzlement and arrears was disclosed, the information was turned over to the fen-ch'a officials, who carried out surprise inspections of the areas involved in order to verify the facts.^[59]

The discovery of the duplicate accounts led to a new influx of petitions from gentry, clerks, and commoners who pledged to pay up their taxes, but begged to have their areas excused from compiling accounts and carrying out a thorough ch'ing-ch'a .^[60] Some officials feared that the large number of confessions to embezzlement received from yamen runners might be part of a cover-up in which runners were being bribed to take the blame for arrears in the hope that once the government was reimbursed the investigations would cease.^[61] If anyone had such hopes, he was soon disappointed, for from the start, the intention of the emperor was to disclose the origins of corruption and arrears at the local level and rectify the irregularities that made them possible. Actual reimbursement of arrears was a secondary issue at this time. Again and again the emperor reminded his emissaries in Kiangnan that the "important thing in this ch'ing-ch'a of accumulated arrears is to ferret out abuses. You must not devote yourselves exclusively to the collection of taxes."^[62] Therefore, Ma-erh-t'ai refused these requests and reminded the Kiangnan public that the purpose of the ch'ing-ch'a was to remove the burdens on the people and thoroughly clean up Kiangnan's tax administration. This was to be accomplished by isolating those households that were falsely registered and discovering the real parties responsible for each mou of land so that accurate registers could be compiled.^[63] Because most of those who were now volunteering to pay up probably had something to hide, Ma-erh-t'ai was adamant that their tax obligations receive special attention.

With the help of the duplicate registers, confessions by clerks, runners, officials, and gentry, and interviews with ordinary villagers, the ch'ing-ch'a officials were able to begin to assemble a profile of abuses in Kiangnan's tax administration. Following the guidelines established by the emperor, these were divided into three categories: official embezzlement, clerk and runner engrossment, and evasion of taxation by the taxpayers themselves.

The Anatomy of Tax Corruption

The fabrication of "people's arrears" by chou and hsien magistrates was the most easily detected of the malpractices examined

by the ch'ing-ch'a investigators. Many of the methods used had already been discovered during the deficit inquiries that accompanied the huo-hao kuei-kung reforms. Chou and hsien officials with shortages in their own treasuries and granaries often falsified the liu-shui ch'uan-ken in their yamen in order to transfer taxes received from the people into their own coffers.^[64] Although this was commonly done at the time that an official turned over his accounts to his successor, or in anticipation of a spot investigation by his superior, natural disasters also prompted officials to embezzle taxpayer remittances. In some instances, a magistrate who had obtained a tax deferment for his jurisdiction would conceal the emperor's benevolence from the local population. Because the government did not expect any remittances at this time, all taxes paid could be pocketed by the magistrate and recorded in his yamen accounts as "yet to be paid." At other times the magistrate chose not to risk punishment for failure to announce a tax remission, but pocketed all taxes received prior to the release of the announcement of the imperial grace period. In both instances the taxpayers would consider their taxes to have been paid, hence the embezzlement amounts would eventually show up in the yamen accounts as people's arrears due to tax resistance.^[65]