PART IV
THE PERSONNEL MANAGEMENT SYSTEM

13
The Personnel Dossier System

Personnel management (renshi guanli ) deals with a wide range of specific tasks such as recruitment, training, appointment, assessment, promotion, transfer, demotion, dismissal, and retirement, in addition to wage and fringe benefit questions. Decisions have to be made and implemented on these issues for each individual cadre; then they are entered into a personnel dossier, a file of written materials on an individual that the regime keeps on every cadre, every party member, and every Communist Youth League member.^[1]

Prior to the CR

Keeping written records is, as Max Weber argued, one of the basic traits of any bureaucracy; all governments normally maintain files on employees. However, the most striking aspect of the Chinese personnel dossier system is the comprehensive scope and political nature of information stored in each file. Cadres have only one personnel dossier, which includes all information on them, although sometimes the regime maintains two versions of a dossier; one is the original (zhengben ), which contains all relevant

[1] "Dossier" in a broad sense refers to a file containing systematically collected materials on any subject. The types of dossiers often referred to by the Chinese news media are personnel, historical, scientific, and enemy dossiers. Different units maintain and control each type of dossier. Access to and secrecy surrounding them varies from one dossier to another. However, when used in cross-checking, dossiers provide the regime with a very powerful control mechanism. It maintains some kind of personnel dossier on everyone except peasants. For English sources on the dossier system, see A. Doak Barnett, Cadres, Bureaucracy, and Political Power in Communist China (New York: Columbia University Press, 1967), 49–50; Melanie Manion, "The Cadre Management System, Post-Mao: The Appointment, Promotion, Transfer, and Removal of Party and State Leaders," China Quarterly , vol. 102, June 1985, 203–33.

material, and the other is a partial duplicate (fuben ), which has only a summary of the materials.^[2]

Obsessed with the cadres' political loyalty, the party-state during Mao's era insisted that

[We] have to examine whether or not one has unlimited loyalty to the party and people and one's political and historical records, ideology, personality, and attitude toward study. [We] have to examine them regularly. Only when we have made a detailed examination of each cadre's political history, political conditions, political quality, ideology, work style, work performance, and ability in a systematic way, can we systematically understand cadres, correctly recruit, and use them.^[3]

During Mao's era the personnel dossier system performed two important functions for the regime. First, it formed the foundation for managing cadres. Through the documentary materials in the dossiers the leadership were able to make rational decisions on recruitment, promotion, assignment, transfer, and so on for each cadre. In this sense, the personnel dossier system constitutes an indispensable part of the broader personnel management system in China. For this purpose, each cadre dossier contains information on: (1) organizational relations—records of party or CYL life, (2) administrative relations—materials on career background, present positions, and educational background, and (3) supply relations—salary grade and other fringe benefits.

Second, the personnel dossier system enabled the regime to maintain tight political control over the cadres.^[4] For this purpose, any information even only tangentially pertinent for checking cadres' loyalty to the regime, to the party, or to any particular faction in power at a given moment was collected. Each of the many campaigns that the regime has carried out since 1949 has generated huge amounts of information for dossiers: the 1941 party rectification yielded a large amount of material that eventually formed the

[2] The level directly above a cadre keeps the original, the cadre's unit, the simplified version. In the past, material in a dossier was not limited to reports about job performance. Wang Faxiong, Renshi Dangan Guanli Gailun (Hubei: Hubei Renmin Chubanshe, 1984), 31.

[3] Guangming Ribao , 14 November 1957.

[4] Ibid.

basis of personnel dossiers; the land reform, about landlords; the Three Antis and Five Antis, facts about people working in commerce and finance; and the antirightist campaign, data on intellectuals. The regime carried out in 1956 the first nationwide investigation of each cadre's file, cross-checking contents against the information provided by the masses and from files seized from the Nationalist Party.

The importance of the personnel dossier system for each cadre cannot be overemphasized. Whether one was recruited, promoted, or demoted, whether one entered college or went abroad for study, whether one survived a political campaign, who would be chosen as a target in a particular political movement—the outcome of all these largely rested on the contents of one's dossier.

Because personnel dossiers are so all-inclusive, they can work to the advantage or disadvantage of each cadre, depending on the information they contain and the users' intentions. The dossier may offer the regime a solid basis for selecting cadres according to the criteria emphasized at a given moment, or it may offer an excuse to persecute anyone the regime may choose to. In 1957, during liberalization, many intellectuals charged that "the party has two record books; one record is a good one, the other is a bad one; when you are needed, the good record is used; when you are a problem to the party, the bad record is used."^[5] The regime did not deny that the dossiers can be two-edged.

Although the cadres have always been concerned about the personnel dossier system, they did not dare to criticize it publicly. Only when political control was relaxed did the system become a focal point of criticism. During the Hundred Flowers campaign, many intellectuals condemned the dossier as "a registrar for life and death," "secret document bags," "benefactor roster" (enming an ), and "ammunition cartridge" (secret reports prepared through dubious channels).^[6] Intellectuals believed that "nobody can be completely liberated [fanshen ] unless the dossier system is abolished."^[7]

The party may have inherited the dossier system from the Soviet

[5] Renmin Ribao , 27 July 1957; Wenhui Bao , 27 October 1957.

[6] Changjiang Ribao , 22 August 1957.

[7] Wenhui Bao , 27 October 1957.

Union, for it has as long a history as the party itself.^[8] However, specific norms for the personnel dossier in terms of format and handling gradually evolved. The first national conference held in 1956 apparently systematized the varying practices and laid down standard procedures for collecting, arranging, and putting materials into the dossiers and maintaining and utilizing them. In the ten years up to the CR, the system was expanded further with the enactment of more specific rules oncadres' dossiers.^[9]

Personnel dossiers are state secrets.^[10] Any mention of them in the media has been general and ambiguous, often only indicating their existence. When someone dies, the regime keeps the dossier for another five years at least, then transfers it to the government dossier bureau.^[11] No individual is supposed to have access to his own or a relative's dossier, and those who work with dossiers are forbidden to reveal their contents. When critics of the regime demanded public access topersonnel dossiers, the regime gave three reasons for their being kept secret. First, an individual's privacy must be protected from other individuals (but not from the state). Second, dossiers include facts that have been verified as well as suspicions, rumors, and other pieces of potentially damaging, but unconfirmed, information that should certainly not be made public. Third, a dossier's confidentiality prevents unnecessary disharmony among the people.^[12]

contents

Each personnel dossier includes official forms, documents, and materials. Most are ones that each individual has provided, and the rest is either evidence collected or opinions rendered by the organization.^[13] The most frequently included forms are one's Summary Career History (jianli piao ), Promotion to Cadre (for newly appointed ones), Cadre Registration (when assigned to a new

[8] A director of the dossier bureau of the State Council started his career as a code operator in 1927. Renmin Ribao , 7 January 1980.

[9] Wang, Renshi Dangan , 75.

[10] Zhengming , no. 3, 1980, 74.

[11] Zhonggong Zhongyang Zhuzhibu Yanjiushi, ed., Dangde Zhuzi Gongzuo Wenda (Beijing: Beijing Renmin Chubanshe, 1984).

[12] Guangxi Ribao , 10 August 1975.

[13] Ibid.

unit), and Application for Party Membership. Supporting materials also appear. The formats of these documents vary, although they all have columns for certain specific kinds of information.^[14]

The contents of most of these forms overlap. As is the case with those distributed by most bureaucracies, spaces are provided for such data as name, age, sex, birthplace, family origin, dates of employment, cultural level, and party membership. Party members must describe when, where, and by whose recommendation they first joined the party, also explaining whether or not the spouse is a party member and what position he or she holds.^[15] A closely detailed explanation can be provided, if necessary. In addition, each cadre must give an account of the economic conditions of their families, in detail, before and after liberation. They also have to specify their family background (jiating chusheng ), for until recently this was regarded as the most crucial factor in personnel management. All classifications are based on what a cadre's father did between 1946 and 1949 (or, for veteran cadres, on when they participated in the revolution). Those who were adopted are required to write the occupations of their legal parents rather than that of the natural parents, although the natural parents' background is also taken into consideration. The justification for this rule is that family upbringing directly influences one's political attitudes.

For family background, the regime provides detailed categories based on economic as well as political criteria for all Chinese except ethnic minorities.^[16] Though the detailed official categories may be thorough, they are not sufficient to cover China's complex reality. Moreover, there has always been a strong incentive to make one's family background look as good as possible. Even Zhang Chunqiao's description of his father changed from "small staff" to "middle peasant," "handicraft worker," and "medical doctor" in different forms.^[17] Detailed explanations of one's family background may be included on the form or they may assume the freer style of autobiography. In addition to writing about their families, each

[14] "Central Document, No. 10, 1977," Zhonggong Yanjiu 14(7–8) (July–August 1980):80, 165, 172, 163.

[15] Ibid., 80, 165, 172, 163.

[16] For further details on the categories, see Richard Klaus, Class Conflict in Chinese Communism (New York: Columbia University Press, 1981).

[17] "Central Document, No. 10, 1977," 167–69.

cadre must also write about their own status, based on what they have done since becoming economically independent.

After the section about family background and personal status, most forms require people to list their "key family members." There are detailed regulations on who constitutes a key family member. Both parents and children are regarded as key family members, whether or not one is living with his parents. But if a woman is married, her parents are relegated to mere "key social relations," and her parents-in-law are considered the key family members.

People must also write about their key social relations. Close friends, classmates, colleagues, and relatives who are not key family members are in this category, whose definition is hazy compared with that of key family members. A cadre has great latitude about whom he lists; but if he leaves someone out—particularly someone who has once been politically active—the organization may question him if suspicion arises.

Most official forms also include specific questions about whether a person or a family member has ever been arrested, jailed, or executed by the CCP. If so, the person filling out the form is unlikely to go far in the party. More information is required about the one who was arrested, jailed, or executed. Did he ever "join a reactionary organization or reactionary military and surreptitious religious organization? What position did he hold? Present relations? Any references?"^[18] Last, theform asks for a brief statement about the compiler's career background, including academic training. Each school attended and each place of employment must be given with references and addresses.

After filling out all the official forms, one must write a detailed autobiography (zizhuan ), which covers the writer's life from the age of seven on, in particular, focusing on such politically relevant activities as demonstrations, publications, and any association memberships. The autobiography is expected to reveal how its author's thought has developed, but it must also include all political relationships. Discussion of friends is not needed except in a political context.^[19]

[18] Ibid., 164.

[19] For an example of autobiography, see Guancha Zhe , January 1980, and Zhongguo Chingnien Bao , 9 February 1980.

The finished autobiography is scrutinized by the small group to which the person belongs as well as by the party committee. If satisfied with the content, the committee will include the autobiography in the dossier. However, everyone is required constantly to supplement his autobiography. For example, if something of political significance happens to a key family member or if a sister emigrates, then this change should be reflected in the personnel dossier. A supplement to the existing dossier may be ordered by the organization when a minor mistake is discovered or when an individual volunteers information out of a guilty conscience.^[20] All these materials are arranged in the order specified by the guidelines of the central organizational department.^[21] Only the department has the authority to put materials into the dossier; individuals cannot collect materials or demand that the department put any particular piece of information into their dossier.^[22]

Maintenance

Although in the past, ultimate authority over personnel matters lay with the party committee, the actual management of the files was handled by the organizational department of the party and the personnel bureau of the government. Staff members working in both organs were usually party members, and quite often they held positions in both organs concurrently.^[23] A party secretary or standing committee member normally supervised personnel work. When the party committee was small, one deputy secretary was in charge of personnel management. Thus, unlike a personnel unit in an American organization, which is considered a staff department, both the party's organizational department and the government's personnel bureau have substantial influence over personnel matters. Consequently, the personnel bureau draws sharp criticism whenever there is any political relaxation.

Following the principle of managing the cadres in a "unified, level-by-level, and field-by-field" fashion (see chapter 14), the party committees at the various levels keep the dossiers on the cadres

[20] Wenhui Bao , 27 October 1957.

[21] Wang, Renshi Dangan , 133–38.

[22] For instance, testimony by other people certifying one's starting date of revolutionary work is not allowed in the dossier. Ibid., 10.

[23] Barnett, Cadres , 46–65.

over whom they have personnel authority. Generally speaking, the party unit at one level has authority over the dossiers of the high-ranking cadres belonging to the next lower unit. Thus, the dossiers of the highest provincial and municipal leaders are kept in the organizational department of the Central Committee. The provincial organizational departments manage the dossiers of the cadres over whom the provincial authority has jurisdiction. The county party committee is the lowest unit that is allowed to maintain personnel dossiers.^[24] This elaborate division of jurisdiction within the hierarchy is designed to prevent any cadre from keeping his dossier under his own jurisdiction, while making it easy for those who make personnel decisions to have access to the information.

Although it is unclear how authority is divided between the party's organizational departments and the government's personnel bureau, it seems that the division of power between the two depends on the cadre's rank. The organizational department of the Guangdong provincial party committee maintains dossiers for leading cadres (above grade 12) of all organs directly under the provincial government. The personnel bureau of the provincial government keeps dossiers for middle-echelon cadres—those between grades 12 and 18 (between division chief and section chief)—of all the organs directly under the province. According to one informant, the personnel bureau of Guangdong provincial government maintained about 5,000 dossiers at the time of the CR. The cadres below grade 19 have their dossiers kept in the personnel section or the organizational department of the bureau level. The same principle may apply to the units below the provincial level.

At each unit, except probably in small enterprises, dossiers are kept in a specially designated dossier room. One person is accountable for approximately every 1,000 dossiers. To be allowed to work with dossiers, a person receives close scrutiny of his political reliability and party spirit.^[25]

On the dossier's outer envelope are written the person's name, dossier number, sex, unit, birthplace, date of birth, family background, individual status, date of first engaging in revolutionary

[24] Zhonggong Zhongyang Zhuzhibu Yanjiushi, ed., Dangde Zhuzi , 279.

[25] Wang, Renshi Dangan , 57–62, 110–32.

work, status, position, and address. Each dossier is given a serial number, the first two digits specifying the field to which one belongs, and the other digits assigned by seniority. Files are organized according to serial number. In order to facilitate retrieval, each personnel department maintains a directory arranged according to the number of strokes in the last name. In addition to normal updating of dossiers, custodians are required to check the serial numbers once every six months and to index them annually.^[26]

Before the CR, there was a well-established set of rules to access dossiers, and these rules have recently been reinstated.^[27] Two people must carry out any investigation of a political problem in the personnel dossier. In order to have access to the dossier for cadres above section chief, both of the investigators have to be party members, and at least one must be of a higher rank. If the cadre being investigated is below section chief, one investigator can be a nonparty member, although this is very rare. When a party committee wishes to check a dossier held by another unit, it has to write a letter of introduction for its investigators. The letter specifies the investigators' ranks and positions and whose file they are authorized to see. After registering with the persons in charge of the dossier, the investigators leave the letter, which is later added to the dossier. Questions raised about the contents of a dossier are also put into it.

Investigators are allowed to see dossiers of those specified by the letter only in the designated dossier room. No mechanical duplication of a dossier is allowed except for a simplified version of the dossier. When part of a dossier is copied, it is verbatim; no summarizing or paraphrasing is allowed. The person in charge of the dossier must authenticate every page copied and the entire package.

Control over copied materials is strict because they can enter the dossiers of others as supporting evidence. To ensure proper control, the regime authorized each unit maintaining dossiers to set up more detailed regulations.^[28] As for lending dossiers, "as a rule, a dossier cannot be checked out. But under special circumstances, it can be lent with approval [of the party committee]. However, lend-

[26] Ibid.

[27] Ibid., 32–34.

[28] Ibid., 33, 43.

ing should follow strict registration, and those borrowed should be returned within the due date."^[29]

A personnel dossier always follows a person when he is transferred. In the early 1950s the transferred cadre could carry his own dossier in a sealed envelope. This practice allowed many people to change the contents in order to secure a better job. Thereafter, all dossiers were required to be sent out only through "confidential transportation," and now only cadres in organizational departments above the county level may carry them.

Investigation and Assessment

The two most important functions of the party committee were the investigation of the contents of a dossier (shencha ) and the assessment (jianding ) of a cadre's performance in both work and politics. Three occasions precipitate an investigation. First, anything questionable in the autobiography requires further investigation. Second, the reexamination of a cadre's dossiers occurs when the organization receives accusation letters (jiancha xin ). Before the CR, this kind of letter would not automatically have initiated a full-scale investigation unless the charge was serious and the letter was signed by the sender. However, during the CR, any suggestion of wrongdoing, either in an anonymous letter or in an oral accusation, frequently brought about an investigation. The present leadership is ambivalent toward an anonymous letter; on one hand, it admits that fear of reprisals leads people not to sign letters, but, on the other, it warns that malicious people often use the method to lay "false charges."^[30]

The third occasion for investigation is a political campaign. Then, the personnel bureau or organizational department examines dossiers carefully, selecting ones with problems and reporting them to the party committee,^[31] which decides whether a cadre's problem is serious enough to make him a target. The same method was used in the last party rectification campaign of 1984–86, which intended to punish former radicals for what they did almost twenty years before.

[29] Zhonggong Zhongyang Zhuzhibu Yanjiushi, ed., Dangde Zhuzi , 281.

[30] Dangde Shenghuo , no. 8, 1983, 37; no. 8, 1984, 10.

[31] Qingdao Ribao , 14 September 1957.

To find falsification, investigators look for inconsistencies, particularly in the many versions of a cadre's autobiography. Any inconsistency is cause for investigators to suspect that a cadre has deliberately falsified his dossier. Discrepancies are entered in the dossier and dealt with during the next political campaign.

If a case requires further evidence, the party committee may decide to send out its own investigation team. In this case, the regulations stipulate that at least two people (sometimes three if they need to travel a long distance) must be dispatched together. Otherwise, any evidence collected is inadmissible. The investigators, who are usually selected from personnel agencies, ensure the reliability of information collected from witnesses.^[32]

After returning to their respective units, the investigators write their reports, including corroborating testimony from witnesses, and submit them to the party committee. If the discrepancy between the new findings and what is reported in the personnel dossier is not serious, the committee will not make a case immediately, but the entry noting the discrepancy remains. In the next campaign, the party committee will try to verify the materials again.

During the CR, the Red Guards nominally followed the same procedure, quite often using the pretext of investigation for sightseeing trips. Due to an emphasis on "class struggle," suspicion on such trivial matters as class background resulted in sending out numerous investigation teams. As one observer argued, China must have wasted enormous amounts of money and energy on these trips. Moreover, most of the investigators merely collected evidence to support already-formulated conclusions.^[33]

Apart from investigations, the party committee regularly assesses a cadre's work competence and political performance. The assessment can be made: (1) at the end of each year, (2) when the cadre is transferred to a new unit, or (3) at the end of each political movement. During Mao's era, the assessment focused on and emphasized political qualifications rather than functional work competency.^[34]

Standard procedure for making an assessment requires, first,

[32] For a documentary example of this kind of evidence, see "Central Document, No. 10, 1977."

[33] Zhengming , no. 9, 1983, 55–56.

[34] Wenhui Bao , 10 August 1957.

that each individual write his own assessment, detailing salient points and focusing on his own political performance and attitudes, strengths and weaknesses, achievements and mistakes. Second, the self-evaluation is read and discussed in a small group, which consists of colleagues who know one another very well.^[35] Third, on the basis of the self-evaluation and the small group's report, the party committee writes an evaluation that is added to the dossier. Sometimes the evaluation has to be approved by an upper-level party committee. Assessment materials make up a large portion of the dossier.

Those being evaluated have a pretty good idea of the contents of the assessment. They know what they have written and what suggestions were made in the small group discussion. Cadres sometimes have access to the summaries of assessments made by the organizations. They were required to sign their names on the organizational assessment, and they have the right to reserve opinion on that portion with which they disagree. The assessment becomes effective only when the units and those evaluated sign together.^[36] Despite this rule, many cadres complain that "party organizational assessment about cadres is surrounded by mystery."^[37] During the CR many units stopped making assessments, and in places where they continued, coercion replaced objectivity. Recent regulations have revived the old practice of requiring the signature of the person being assessed.

Administrative control over the personnel dossier was well institutionalized in pre-CR China. The dossiers were strictly regulated, and there were at least well-established administrative practices, if not uniform rules, regarding the types and processing of information that went into the dossiers. These guidelines also applied to the maintenance and investigation of dossiers, as well as to rendering organizational judgments on the contents and on each individual cadre's performance. Thus, the dossier provided the party with accurate information on each cadre, enabling it to carry out recruitment and promotion in an orderly and justifiable way.

[35] For a detailed description of the small group operation, see Martin Whyte, Small Groups and Political Rituals in China (Berkeley and Los Angeles: University of California Press, 1974).

[36] Renmin Ribao , 10 August 1957.

[37] Wenhui Bao , 27 October 1957.

The pre-CR personnel management practice, however, had several serious weaknesses, which subsequent politics magnified. First, there was no external mechanism to stop the party from abusing its authority in the area of personnel management. Whatever restraints there were, they were internal ones, whose effectiveness depended on goodwill. For instance, each cadre was granted the right to appeal to the upper echelon of the bureaucracy. But this right was generally ineffective in guaranteeing a fair review opportunity for the cadre because, as many Red Guard newspapers asserted, the upper echelon of the bureaucracy frequently showed more sympathy to the lower level than to an individual cadre.

Second, the heavy reliance on personal testimony—in contrast to the Western judicial practice of attaching importance to material evidence—and the lack of strict rules with regard to the admissibility of evidence and its interpretation offered even more room for abuse. The Chinese practice of interviewing those who knew the investigated person and using their testimony in written form as evidence reflected in part the cultural tradition and in part technical backwardness in dealing with material evidence. The effectiveness of this practice as an investigation method largely depended on an assumed mutual trust and a consensus on what was right and wrong among the people involved in the process. When the CR shattered the consensus on fundamental values, the practice degenerated to producing a large number of false charges and framed-up cases.

Third, the Chinese emphasis on "confession," which was epitomized in the official slogan, "lenience to those confessing, but harsh punishment to those refusing," contained a seed of abuse and excess. The notion that people would honestly confess their mistakes also indicated the utter lack of philosophical distinction between public and private domains, a distinction that led in the United States to the Fifth Amendment, granting the individual the right to refuse self-incrimination. As was the case with witness testimony, confession can work as an effective mechanism for justice only when all the participants share common criteria for right and wrong. When the official criteria changed during the CR, coercion, torture, and other types of physical punishment were widely used to force the purged to "confess" their crimes or to

write accusations against others. A family member, relative, or other acquaintance was subject to the same abuse.^[38]

Fourth, the principle of double jeopardy has never been firmly established in China. Even a case over which the party committee had rendered a final conclusion could be reopened when new evidence—reliable or not—became available. Many people were persecuted several times for the same reason.^[39] As a result, there has been enormous pressure for reversals of past decisions and for the rehabilitation of victims, particularly when policies have changed with new leadership. By discrediting the organizational legitimacy of the party, the CR gave rise to demands to reverse previous decisions. One group of political leaders at the top level could then utilize such demands for their own political gain. As a result, the Chinese people are justifiably worried that if any drastic policy change occurs, what is regarded as virtue today—in Chinese a "red-color dossier"—will turn out to be a liability—a "black-color dossier."

Impact of the CR

From the beginning of the CR, the personnel dossier was the focus of attention, although no one publicly challenged the system per se. When Mao pressured the party into launching the CR, it sent out work teams to school campuses. Largely composed of cadres from outside the campus, the work teams examined the dossiers of each cadre and student and then classified them into rightist, middle roader, or leftist. When the work teams themselves came under public attack for having diverted the main targets from the party leaders to the masses, Mao ordered the materials collected by the work teams (generally known as "black materials") to be burned. However, some work teams hid materials among the personnel dossiers and other confidential party materials for future use. This offered the Red Guards an excuse to seize personnel dossiers under the pretext of searching for "black materials."

As the dossier rooms became less inviolable to the Red Guards,

[38] Shehui Kexueyuan Yanjiu Cankao Ziliao (Sichuan), no. 10, 1983, 20–23; Ming Bao , 21 August 1981; Guangming Ribao , 8 January 1979.

[39] For instance, see Zhonggong Gongchandang Lishi Jiangyi , 3d ed. (Jilin: Jilin Renmin Chubanshe, 1982), 2:182.

party leaders improvised various methods to protect the dossiers. In Guangdong, Zhao Ziyang, then the first provincial party secretary, moved all the dossiers under his jurisdiction to the Guangdong military command. The military kept the Red Guards at bay all through the CR by insisting that personnel dossiers were military secrets. At Zhongshan University, the party committee moved its dossiers to the school library, which was closed to students. However, Red Guards raided the library, seizing and destroying some of the dossiers.^[40] Personnel dossiers for high-level cadres (those above section chief) survived intact, but the dossiers for lower-level cadres sustained some damage.^[41]

In February 1967, the center ordered that mass organizations should not raid or seize personnel dossiers, that seizure of power in dossier rooms should be carried out exclusively by the rooms' original personnel, and that people unsuitable for the work should be transferred according to regular procedures (which required upper-level approval). Realizing the danger of seizing dossiers but suspicious that other factions might obtain access to them, warring factions in many units negotiated to seal the dossier rooms. Denied access to official personnel dossiers, the Red Guards began to build up their own by collecting materials and carrying out investigations.

The mass organization's inability to look at official dossiers, however, put them at a great disadvantage vis-à-vis PLA representatives. The latter often used their privilege of having access to dossiers in order to promote partisan political interests by supporting cadres of their choice, while rejecting cadres supported by the mass organizations. By contrast, the mass organizations had to rely on their own less reliable materials or on the opinions of the PLA representatives for the selection of "revolutionary cadres." In some cases PLA representatives allowed the mass representatives to look into cadres' dossiers in order to promote consensus, although that was a clear violation of pre-CR practice. Once the revolutionary committee was formally established, it assumed jurisdiction over dossiers.

[40] Hai Feng, An Account of the Cultural Revolution in the Canton Area (Hong Kong: Union Research Institute, 1971).

[41] Dangde Shenghuo (Tianjin), no. 18, 1980.

When the CR began "purifying class ranks" in 1968, the main thrust of the mass movement was turned against "hidden class enemies"—those who were suspected of having falsified their class backgrounds and of covering up their past ties with the KMT. The witch-hunt atmosphere was so intense that every Chinese, including the top leaders, became obsessed with the dossiers. Even Zhou Enlai was accused of once having surrendered to the Nationalists, and his case was not closed until Mao personally intervened.^[42]

Radical leaders had their share of blemishes; Jiang Qing worried about the materials concerning her starlet life in the 1930s; Zhang Chunqiao had to protect his wife, who had once been captured by the Japanese, in addition to accounting for his past ties with the Nationalist secret organization Fuxing She ; Yao Wenyuan had to cover up the fact that his father was labeled a rightist in 1957.^[43]

According to its critics, the Gang of Four used every conceivable means to destroy any materials detrimental to themselves, while collecting, distorting, and manufacturing evidence to incriminate their political opponents. They are accused of having used deception, coercion, and torture for their purposes. Subjected to their abuse were not only the persons suspected, but their relatives, colleagues, subordinates, casual acquaintances, and even innocent bystanders who had any knowledge either of the Gang of Four's past or that of their political enemies. For instance, a librarian at Shanghai University discovered by accident some materials on Jiang Qing's 1930 career in the stacks; because of this, he languished in jail.^[44] Even the Red Guards, who accidentally discovered more background on the Gang of Four, were persecuted for having "collected black materials against the central leaders." The story of how Yao Wenyuan handled the "black materials" reveals the radicals' obsession with them. When Yao received materials on Zhang Chunqiao in a sealed envelope, he wrote: "Contents are not opened and are not known. Ask for instructions from Comrade Zhang Chunqiao and Wang Hongwen. Handle according to party regulations." But Yao kept the materials in his house for almost

[42] Dangshi Yanjiu , no. 1, 1980, 14.

[43] Renmin Ribao , 6 July 1977.

[44] "Central Document, No. 10, 1977."

seven years until his arrest. It appears that at that time the only rule governing politics was that one needed materials to purge others. But there were absolutely no rules about how to determine the materials' reliability or how one obtained them. As a result, "materials" could have an explosive effect, regardless of their reliability.

In brief, the series of continuous political campaigns after 1966 largely destroyed the normal administrative procedures of personnel management. Regular authority and procedures for decision making on personnel matters were abandoned. The rule that only party members could have access to dossiers was completely discarded, as many nonparty member rebels were promoted to positions that allowed for handling the dossiers. As a result of the simplification of the administrative structure, the sections in charge of the personnel dossiers were abolished. When the establishment of proper criteria for recruitment, promotion, appointment, and purging became the major focus of contention among the various political factions, materials in the dossier proved to be irrelevant. The scale of personnel changes during the same period was also gigantic. About 18 million new party members were recruited during the CR, whereas about 2.9 million people were purged. About 2 million workers, soldiers, and peasants were promoted to cadre positions while retaining their previous occupation status, the practice known as using peasants as cadres (yinong daigan ) and using workers as cadres (yigong daigan ).^[45] For instance, workers who were promoted to revolutionary committees continued to keep their former grade. As a result, the positions they actually occupied and their grades (which determine salary) did not match.

Were all of these changes recorded in the personnel dossiers? Where were the materials of self-criticism and the records of being accused and investigated kept? What happened to the huge amount of materials collected by the mass organizations? There is not enough information to answer these questions systematically. Fragmented information allows for a few general observations.

Although no specific regulation was issued by the center, the materials collected by the mass organizations were kept separately from personnel dossiers in order to be checked for reliability later.

[45] Renshi Guanli (Beijing: Beijingshi Renshiju, 1985).

However, it is also very likely that many unauthorized materials entered the dossier.^[46]

In the ten years of calamity, when right and wrong were reversed, the cadres' personnel dossiers, which were not perfect, were further destroyed. Merit became a liability, mistakes became merits, and facts were reversed. Many false materials entered the cadres' dossiers. . . . So much unnecessary and false information entered the cadres' dossiers.^[47]

Moreover, any decision rendered by the organization through the organizational process was entered in the dossier, even if it was reversed later. The verdict made by the Ninth Central Committee on Liu Shaoqi probably was in his dossier. Also, numerous self-criticisms made by the highest leaders—for instance, Deng Xiaoping—were probably included in their dossiers.

A more serious problem was the accumulation of unsettled cases. Even after the reconstruction of the party organization, the party committees at the various levels were unwilling or unable to render organizational conclusions on many cases, largely due to the unreliability of the collected materials, the absurdity of charges made, and the lack of guidance from the center during the chaotic ten years of the CR. The personnel dossier system was in disarray when the Gang of Four was ousted.

Reform

Immediately after the Gang of Four's fall, the regime tried to restore the pre-CR practice. First, party committees at various levels regained their exclusive authority over personnel matters. Then, the regime issued several regulations about dossiers, for example, "Regulations on Cadre Dossier Work," "Methods of Handling Cadre Dossiers," and "Regulations for Personnel Working on the Dossier System." Even though the contents of the rules are not known, the section dealing with dossiers in "Questions and Answers on the Party's Organizational Work" reveals that these new regulations are quite similar to what had existed before the CR.^[48] They intended to tighten organizational control over dossiers

[46] Daily Report , 21 October 1981, 1.

[47] Zhibu Shenghuo (Shanghai), no. 18, 1985, 24.

[48] Zhonggong Zhongyang Zhuzhibu Yanjiushi, ed., Dangde Zhuzi , 276–80.

rather than protect the individual's rights by specifying rules with regard to collecting materials, investigating their reliability, placing them in the dossier, and arranging them in a certain order. For instance, now only materials "approved by the organization" can enter the personnel dossier. Any materials needed for evaluating cadres can enter the dossier if they are approved by the organization. Such materials as records of speeches or results of opinion polls and reports, that the organization has not yet examined, can be kept separately in the personnel departments. Moreover, the regime has organized nationwide conferences on personnel dossiers and endeavored to give specialized training to all people dealing with dossiers.^[49]

In addition, reforms initiated since Mao's death have had direct consequences for personnel management. As noted, the emphasis in recruitment and promotion has shifted from political reliability to ability, particularly technical and professional competency, and largely measured in terms of formal education. The removal of class designations such as counterrevolutionary, rightist, and landlord has made class origin and past political activities less relevant to personnel management. On the whole, documentation of actual performance rather than past performance is now more heavily weighed in personnel decisions. Some measures have been taken to protect individual rights in the state constitution, the party constitution, and the party rules and regulations. The new rules grant individuals the right to "put forward their statements, appeals, accusations, and defense at party meetings or to party organizations at a higher level, right up to the party committee." Also, party members have access to any decision made by the party organization on "appraisals, verdicts, and punishments."^[50] The new criminal code considers it a punishable crime to make false charges or accusations.

The second national conference on cadres' dossiers held in 1982 decided to check every dossier, updating the contents to make them accurately reflect the personnel changes after the CR, and, most important, to "cleanse" personnel dossiers by destroying all materials collected since the CR. The Shanghai organizational de-

[49] Wang, Renshi Dangan , 43.

[50] "Guiding Principle for Inner Party Life," Beijing Review , 7 September 1980.

partment assigned ten cadres to straighten out the personnel dossiers.^[51] Despite these efforts, some cadres are still complaining about the inaccuracy of the information stored in their dossiers.

Even after four years since 1980, still some of the cadre dossiers are not yet cleansed. . . . Still some problems exist with the dossier work. For example, some dossier materials reflect only good news not the bad news, or they reflect the bad news, while failing to report the good news. Most information in the dossier reflects the situation at a given moment, their contents are old, failing to reflect the entire history of a cadre, his work achievements and real conditions.^[52]

To rehabilitate victims of the CR and former rightists completely, the regime issued an official certificate of rehabilitation or removed rightist labels. In order to prove the regime's sincerity, on some occasions the party committee has shown the concerned individual the materials to be removed from the personnel dossier and then destroyed them. However, wondering whether or not all incriminating materials have been removed, some individuals have demanded to see the entire contents of their dossiers. The party committee, of course, will not allow that. The cadres' apprehensions appear to be justified because the policy of cleansing the dossier is not categorical. On the contrary, official policy specifies types of materials to remain in the dossier. They are "materials accurately reflecting historical reality and useful for understanding cadres," "materials related to the investigation, reinvestigation, and rehabilitation of a person's historical problems," and "the organization's conclusion and its supporting material." Understandably, this makes some Chinese nervous about what is left in the dossier.

Moreover, the removal of rightist, counterrevolutionary, capitalist, or other class enemy labels does not mean erasing past records. Most official forms used for cadres still include columns for family background and individual status, and in filling the class origin column, each individual is required to report his former class labels—no one is allowed to change his family background or individual status without approval of the organizations—and to add

[51] Zhibu Shenghuo (Shanghai), no. 18, 1985, 1–18.

[52] Daily Report , 28 September 1981, P1.

his present occupation.^[53] This is the case with the children of the former exploiting class. In other words, the regime has not yet stopped soliciting information on class background but simply promises that it will not be used as a criterion for personnel management. Thus, this practice, plus keeping the traits of past political decisions, makes concerned individuals uneasy about the possibility that when the political wind changes, their records can be used against them.

In fact, it seems that the regime continues to rely on information stored in dossiers whenever the necessity to check a cadre's political reliability arises. For instance, whether or not one is "revolutionized" is largely determined on the basis of materials in the dossier. The investigation of the "three types of people" relied heavily on materials stored in dossiers. In addition, new materials are constantly being generated for dossiers.^[54] For instance, every decision made in the party rectification of 1983–86 entered the dossiers—including delayed and denied registration. The sentence that one has made "comparatively serious mistakes during the CR" can enter one's dossier.

As the scope of reform in other fields expands, the existing dossier system poses some serious problems. For instance, the system comes in conflict with the current policy of utilizing China's limited resources of technical cadres efficiently by making it easier to transfer them across administrative boundaries and by allowing individuals to seek employment and transfer on their own initiative. At present, no organization will accept anyone without first going through his or her dossier. However, unless one is officially transferred, it is extremely difficult to persuade a unit to send a dossier to the new unit.^[55] Anyone who controls the personnel dossier of a particular cadre actually controls the cadre. When several units compete for a particular person, the unit that has access to his dossier can have the person because the other competitors cannot make a decision on him or her without looking at the dossier.

To sum up, the regime has not yet fundamentally changed the

[53] Zhonggong Zhongyang Zhuzhibu Yanjiushi, ed., Dangde Zhuzi , 287. But other sources indicate that after June 1979 one can write his present occupation in the column (geren chengfen). Dangde Shenghou (Tianjin), no. 20, 1980, 9.

[54] Zhibu Shenghuo (Shanghai), no. 13, 1985, 26.

[55] Zhengming , no. 9, 1983, 55.

practice of collecting detailed information on each cadre and storing it in a personnel dossier, although a serious effort is made to check the potential abuse of materials. When some Chinese demanded the abolition of the dossier system on the ground that "since the basic goal of the regime has changed, there is no need for the dossier system," the official view insisted that "on the contrary, when the party's main task changed, the personnel dossier became more important, and its utility increased."^[56] Nonetheless, it seems that decentralization of personnel decision-making power and transferring it to the government personnel bureau are changing the administrative practice of managing dossiers. When the civil service system is thoroughly implemented, the personnel bureau of the government will manage all personnel dossiers for functional civil servants, leaving only the dossiers of political civil servants with the party's organizational departments. Instead of abolishing the personnel dossier system, the regime seems to intend to strengthen administrative control over files and to use them more rationally (e.g., entering examination scores). Consequently, a cynical Hong Kong observer asks: "China is economically and politically the most backward; but it has the most advanced personnel dossier system in the world. Is it a blessing or a curse for the Chinese?"^[57]

There is absolutely no sign to indicate that the regime is concerned about possible infringements of individual rights through the dossier system. The utter lack of concern with individual privacy was again revealed when Qien Xuesen, a renowned Chinese scientist, suggested a few years ago that the Chinese computerize their personnel dossier system. The ostensive objective was to make the personnel management system rational, enabling the government to select easily the best qualified persons for the most suitable jobs.

By using combinations of centralized but grade-by-grade management, we can simultaneously evaluate a thousand, ten thousand persons, and we can select the best-qualified people from several tens of millions of personnel dossiers. The old method of checking personnel dossiers one by one cannot do that. Already there is a

[56] Wang, Renshi Dangan , 44.

[57] Zhengming , no. 9, 1983, 55–56.

system of examining dossiers by computer and even dossier investigation can be automated. A computer can work several thousand times faster than man. Once the criteria for selection are set up, a computer can automatically select the most suitable persons. Since the dossiers are recorded on magnetic tape, they can be supplemented: work condition, masses' and leaders' opinions of a person, and his health record can enter the dossiers through the regular process.^[58]

Given the present level of science and technology in China, it would be extremely difficult to develop a comprehensive system of classifying and coding all the necessary information on personnel management of cadres in the foreseeable future. However, in some units computerization of dossiers has started. It is chilling even to think of a computer that contains all conceivable physiological, biological, medical, and psychological information on every one of the millions of Chinese people.

[58] "Engineering in the Socialist Manpower Field," Hong Qi , no. 2, 1982. Also for computerizing personnel dossiers, see Wang, Renshi Dangan , 184–85.

14
The Party's Changing Role in Personnel Management

All Leninist parties maintain their authority over cadres in order to control not only the vast bureaucracy but also access to elite status. The CCP regards its monopoly of personnel management as "the basic source of authority that our party needs to fulfill its political mission at each historical stage, [and that] renders the organizational guarantee for the successful completion of a new democratic revolution, a socialist revolution, and a socialist construction."^[1]

The party-state vigorously defends its monopoly over personnel, reacting violently whenever social forces challenge this prerogative. During the Hundred Flowers campaign the intellectuals' demand to share authority over cadres prompted the top leaders to initiate the antirightist campaign. Similarly, it was only when Polish workers called for abolition of the nomenklatura system that the military stepped in to repress the Solidarity movement.^[2]

The phrase "the party manages the cadres" justifies the CCP's monopoly. "State cadres are the party's cadres, and all cadres should be managed according to the party's direction and policies and the principle of unified management." More specifically, the principle implies that the party has the exclusive right to set up the "line, direction, and policies" relating to personnel management, no one can challenge the party's prerogative, and no regional variation can be tolerated. All cadres, whether party members or nonmembers, administrative or technical, are to carry out the party's line and program under the CCP's leadership. Furthermore, the principle also states that only party organs at the various levels

[1] Cao Zhi, ed., Zhonghua Renmin Gongheguo Renshi Zhidu Gaiyao (Beijing: Beijing Dafue Chubanshe, 1985), 7.

[2] Takayuki Ito, "Controversy over Nomenklatura in Poland" (unpublished paper).

can select, assign, and utilize cadres. However, the principle does not tell how the party should exercise its authority.

Past Practice

From its foundation to 1949, the CCP directly managed all cadres under a principle of "unified management" without sharing its prerogative with any other government agency.^[3] At each level, party committees made all major decisions on personnel matters while the party's organizational departments handled all administrative aspects of cadre management. Moreover, the system was highly centralized, with upper-level party committees making most decisions on cadres employed at the lower levels. This centralized system was suitable to the revolutionary war period when the size of the cadre corps was small and the CCP was engaged in a desperate struggle with the KMT.

When China started its first five-year plan in 1953, the regime modified its management of cadres to fit the tasks of economic development and socialist construction. As the cadre corps grew and the types of work they were required to perform multiplied, it became more difficult for the organizational departments to manage all the cadres, much less develop a long-term plan for cultivating specialized cadres.^[4] Therefore, the CCP decided in 1953 to manage cadres "department by department and level by level" (fenbu and fenji ) under the unified management of party committees and organizational departments. All cadres were grouped into ten functional categories (xitong ): (1) culture and education, (2) agriculture, forestry, and water conservation, (3) united front, (4) party and government organs, (5) industry, planning, labor, and statistics, (6) finance, economics, commerce, banking, and grain and supply cooperatives, (7) transportation, telecommunications, and postal service, (8) public security, civil affairs, judiciary, court, and procuratorate, (9) foreign relations, foreign trade, and overseas Chinese, and (10) mass organizations such as labor unions, the CYL, and women's associations.^[5] The various functional depart-

[3] Cao Zhi, ed., Zhonghua Renmin , 7.

[4] Ibid., 7–8.

[5] Wang Jianxin, Yang Shugui, Jin Guoliang, and Yan Zhuanyu, eds., Ganbu Guanli Gailun (Liaoning: Liaoning Dafue Chubanshe, 1984), 44; Cao Zhi, ed., Zhonghua Renmin , 417–18.

ments, established under the Party committees, managed the cadres working in their fields.

Level-by-level (fenji ) management meant that each level managed the cadres of its subordinate units down to two or three levels below. The center was responsible for ministers, vice ministers, directors, and deputy directors of the central government and party organs, as well as for leading cadres of provinces and districts. Provincial party committees were in charge of their own middle-level cadres and leading cadres of districts (diqu ), the municipality, and the county. District party committees handled leading cadres of the county and village (xiang ), in addition to their own middle-level cadres (heads of division and sections). County committees looked after their own middle-level (section and team) cadres and the leadership group of village, town (zhen ), and hamlet (cun ).^[6]

Paralleling the party's organizational departments were personnel bureaus, set up in the government, enterprises, and business units to "assist" (xiezhu ) the organizational departments.^[7] Generally speaking, the personnel bureaus managed low-ranking cadres and workers or took care of the administrative work of personnel management, whereas the organizational departments worked as the administrative arm of party committees. Compared with previous practice, the new system was reportedly more suitable for "investigating, observing, and understanding more deeply the political quality and functional competency of the cadres . . . thus helping to fulfill the first five-year plan successfully."^[8]

In order to implement the state constitution, all cadres appointed by the central government or the State Council were screened by the organizational departments before any formal announcement. The cadres to be elected by the state or party organs had to report to the proper authorities before the election.^[9]

By 1955, the center completed the lists of cadres (nomenklatura ) whom it directly managed. The lists included "all cadres assuming nationally important positions."^[10] The appointment or transfer of any cadre on the list had to be approved (pizhun ) by the center,

[6] Cao Zhi, ed., Zhonghua Renmin , 8.

[7] Ibid.

[8] Wang et al., eds., Ganbu Guanli Gailun , 44.

[9] Cao Zhi, ed., Zhonghua Renmin, 156–63.

[10] Ibid., 143.

which also instructed the local committees and their functional departments to prepare lists of cadres under their jurisdiction, using the principle of "managing fewer and managing better" and considering the importance of the post rather than the incumbent.^[11] The lists included a brief description of the major responsibilities of each post. Thereafter, every year the organizational department must update the list.

For conflicting jurisdictional claims over units under the dual leadership of the central functional department and the local party committee, the center urged the involved parties to negotiate through the management offices—for example, the organizational departments—rather than through responsible persons.

Whenever a new question arose with regard to jurisdiction, the regime ruled on a case-by-case basis. For instance, in March 1953 the center placed the leaders of the Sino-Soviet Friendship Association and the Chinese Writers' Association under the propaganda department. The higher learning institutes were placed under the various central party departments according to the functional areas of the schools; thus, agricultural and forestry colleges fell under the jurisdiction of the central agricultural work department, and cadres of nationality institutes were given to the united front department.^[12]

Thus, the pre-CR personnel management system took into consideration to only a small degree the type of unit the cadres belonged to (fengong ) or the level in the bureaucratic hierarchy they occupied (fenceng ). The system lacked any notion of classifying the cadres according to the type of work they performed (fenlei ). Although such terms as "party, state, and mass organization cadres," "business cadres," and "enterprise cadres" (classified according to the type of organ where the cadres were employed), or "political cadres," "administrative cadres," and "specialist cadres" (classified according to the type of work the cadre performed) were used, these categories had no practical ramifications for managing the cadres. The party committees managed all cadres, using uniform criteria for different types of cadres.

How to classify the cadres according to the type of work they

[11] Ibid., 144.

[12] Ibid., 419–21.

did, how to manage different categories of cadres using different criteria, and how to distribute personnel authority among the different state organs—for example, the party, government, and enterprise—were questions that were never raised until the 1982–87 reforms.

Nonetheless, a modus operandi for making decisions on personnel matters evolved from 1953 to the CR, nominally involving four parallel hierarchies—that is, decision-making and executive organs of the party and state—although the party had the final say.^[13] The system was not designed as a master plan; rather specific regulations dealing with concrete problems were accumulated, thus producing the general structure of the system.

The chaotic CR, however, destroyed whatever regulations and rules that existed at the time and abolished the personnel bureaus in the government units. As may be inferred from table 38, about 6.4 million joined the cadre ranks without going through formal procedures.

Initial Adjustments

After the fall of the Gang of Four, the regime restored the pre-CR system whereby each party committee managed the cadres two levels below it. It reissued cadre lists to be managed by the center.^[14] In addition, the regime clarified the issue of jurisdiction over enterprise and business cadres: the central organizational department would manage leading cadres of the enterprises directly under the center in coordination with the party core group (dangzu ) of the relevant government ministries.^[15] Management of middle-echelon cadres in the enterprises was generally assigned to the enterprise and business unit party committees. For enterprises

[13] Wang et al., eds., Ganbu Guanli Gailun , 54–56.

[14] Cao Zhi, ed., Zhonghua Renmin , 147.

[15] The party core group was set up right after 1949 in all government agencies when many administrative heads were nonparty members. Existing separately from the party committee at the same level, it was appointed directly by the higher party committee to which it was accountable. Xinhua Wenzhai , no. 11, 1978, 1–6; Zhongguo Shehui Kexue , no. 6, 1987, 3–22. Consisting of three to five persons and a group secretary, the party core group maintains the list of people over which they have authority. John Burn, "China's Nomenklatura System," Problems of Communism , September–October 1987, 36–51.

subject to dual jurisdictions—the central party department and the local party committee—"the higher-level party committee will be in charge, and the lower-level party committee will help with observation, examination, training, education, and investigation." With regard to appointment, promotion, transfer, reward and punishment, the lower-level committees could make only suggestions.^[16]

With regard to the division of work between the party's functional departments and its organizational departments, the regime instructed each provincial authority to decide in a way that would uphold the principle of "unified management," while making the system flexible by allowing each functional department to manage cadres in its own field.^[17]

Soon the two-level management by the superior party committee proved to be too highly centralized to work effectively. The regime gradually decentralized the personnel authority by transferring some cadres from central to local jurisdictions. In 3 February 1979 the organizational department authorized nine central organs to manage their own deputy heads. On 24 December 1980 a notice issued by the central organizational department transferred to provincial authorities the jurisdiction over district party secretaries, directors of provincial bureaus, and responsible persons of some enterprise and business units. Provincial party committees were required simply to notify the center of their decisions.^[18] The same notice divided the cadres on the center's lists into two categories. Category A included only the leading cadres of the center and provinces.^[19] All those not listed in category A belonged to category B. When appointing or dismissing members of category A, the party functional departments or the party core groups in the government ministries and the other relevant organs first had to seek the opinion of "the central leaders or vice prime ministers in charge of the units" and include their opinions in the report submitted to the central organizational department. For cadres under

[16] Cao Zhi, ed., Zhonghua Renmin , 148.

[17] Ibid.

[18] Ibid.

[19] According to John Burn, this category included about 5,000 of the most senior people throughout the country. For a sample of the list, see Burn, "China's Nomenklatura System."

dual leadership, the dominant unit would first seek the opinion of the other side and then report both opinions to the center.

For the posts included in category A but whose incumbents were selected through election, the list of candidates had to be reported with the information specified in the "Form for Reporting Cadre Appointment and Dismissal" to the center for examination before an election. The results of the election also had to be reported to the center for approval. Cadres on the B list were transferred to provincial authorities, who reported their decisions to the center for record keeping.^[20] Thus, cadres on the B list were in fact transferred to provincial jurisdictions.

Any appointment or dismissal of cadres, which required the State Council's approval, had first to be forwarded to the party center. After reviewing the request, the organizational department notified the ministry of labor and personnel of its decision. Only then would the case be submitted to the State Council to formalize the changes according to the regulation. For provincial leaders, whose appointment and dismissal required approval of the local People's Congress, the relevant authority—that is, the provincial organizational departments—first had to consult with the standing committee of the local People's Congress and then report to the central organizational department or to the State Council. These complicated rules were designed to keep personnel authority in the party organs while following constitutional and other legal formalities.

For cadres belonging to category B, each provincial and municipal party committee as well as the various central party organs, the various ministries, and the various people's organizations had to make nominations in written form, explaining in detail the reason for the appointment or dismissal to the center.

The "Regulation Regarding Reform of the Cadre Management System," adopted on 5 October 1983, finally formalized the scope of cadre management by party committees at the various levels. According to this regulation the center managed only "the first-class leadership" of the provincial organs. It also managed ministers, vice ministers, members of the party faction in government ministries, and directors and deputy directors of bureau-level

[20] Chao Zhi, ed., Zhonghua Renmin , 149.

organs. Control over the rest of the cadres was given to provincial authorities.^[21] On 14 July 1984 the organizational department issued more instructions in order to clarify the previous decisions and to solve the problems arising in the process of implementing the decisions. Thereafter, provincial and prefectural party committees prepared the list of cadres they controlled while others were sent down one level below.

The new system is called "managing according to ranks and having each echelon responsible" (fenji quanli, cengceng fuze ). This system decentralized the personnel management authority to a certain extent. Each level manages the one directly below it, instead of two levels below. The organizational department issued a "Notice Regarding the Revised Cadre Post List Under the Center's Authority," which retained only one-third of the cadres on the previous list. The rest were transferred to a lower level. Each state organ reprepared the list for which it was responsible. The new lists, according to an official claim, enable each central ministry and local party committee more effectively to "supervise, check, and discover the talents" for the positions for which they are responsible. The party committees survey personnel changes in the positions listed under their jurisdiction twice a year and report them to the organizational department.

However, overlapping claims on cadres continue, particularly among government ministries, which exercise functional leadership (yewu lingdao ) over their subordinate enterprise and business units scattered nationwide, and local party committees, which have "leadership relations" with the same units. The functional leadership is often called "line" (tiaotiao ), and the leadership that the local party committee exercises over the enterprise and business units is known as "area" (kuaikuai ). The division of work in the area of personnel management between tiaotiao and kuaikuai is particularly unclear and gives rise to situations where both pass the buck and both check each other.

Generally, there are three types of arrangements for the claims of the tiaotiao and kuaikuai . The first type is dominated by the central ministries: their party core groups are responsible for appointments, transfers, examinations, investigations, assess-

[21] Huaqiao Ribao , 14 July 1987, 1.

ments, promotions, and dismissals. Local party committees assist the work of the various central departments, assuming major responsibility only for the political, ideological, and functional education of the cadres.^[22] The local party committees also supervise the administrative side of personnel management as well as such party functions as convening party conferences, electing party committees, and examining (shencha ) candidates for committee membership.

The second type is dominated by local party committees: they are mainly responsible for personnel management, whereas the upper organs provide functional leadership and make suggestions only on personnel matters. The third type involves the conflicting claims between local party committees and enterprise and business units under the local authority. For this case the involved parties are supposed to divide their work through negotiation and bargaining.

In addition to the limited decentralization of personnel authority along the party hierarchy, the regime also tried to rationalize the methods of managing personnel matters.

Recruitment

All cadres in China are "appointed" (renming ) by superior organs (shangji ), and this method has not changed in any meaningful way, despite the frequent misuse of authority by leading cadres.^[23] Instead, the regime has endeavored to improve its operation by emphasizing careful screening by the organizational departments and collective decision making by party committees. Officially, any important decision about cadres should involve "democratic nomination, consultation with a wide range of opinions, selection of candidates with care, investigation by organizational departments, collective discussion by party committees, reports to upper echelons, rechecking by organizational departments, and final approval by party committees one level higher."^[24] In addition, the

[22] Wang et al., eds., Ganbu Guanli Gailun , pp. 48–49; Cao Zhi, ed., Zhonghua Renmin , 422–23.

[23] Shehui Kexue Pinglun , 5 September 1985.

[24] Renmin Ribao , 2 February 1986.

regime has experimented with several other methods to improve the effectiveness of the appointment system.

One remedy is to make appointments only after soliciting recommendations from the masses and after careful organizational investigation (kaocha ).^[25] The regime now encourages the masses, low-ranking cadres, and retired cadres to recommend qualified persons for specific, cadre positions. Even if fully utilized, this method will not change the appointment system; it only broadens the pool of candidates from which party committees will make final selections. So far it seems that only retired cadres effectively use their "right to recommend" to influence the final selection for their benefit.^[26]

Another remedy is a kind of election, primarily used to select administrative heads for small enterprises and business units. For instance, Guangdong province reportedly used an "election" to select the managers of the Jianmen Beverage Factory and the Friendship Store of Shenzhen, as well as the department heads of the Central-South Institute.^[27] Viewed as a means of "finding competent persons" rather than of electing representatives, the "election" resembles an opinion poll because the outcome must be approved by a proper party committee. Despite its experimental nature, some old cadres are openly critical of elections on the grounds that they aggravate factionalism, and "talent [rencai ] will not necessarily get the most votes."^[28] It is, therefore, very unlikely that "elections" will become an important mode of selecting cadres.

For recruiting specialist cadres, some units use an "invitation system," in which each hiring unit and individual applicant bargain, negotiate, and make a contract that contains specific terms on tenure, remuneration, and type of work to be done.^[29] The regime tolerates a limited labor market for specialist cadres because of the need to utilize scarce manpower effectively and to allow peripheral regions and rural areas to attract much-needed specialists. To criti-

[25] Ibid., 12 October 1982.

[26] Ibid., 7 August 1985.

[27] Guangzhou Yanjiu , no. 3, 1985, 41–42.

[28] Renmin Ribao , 21 March 1985.

[29] Guangdongsheng Renshiju, ed., Shenzhen Ganbu Renshi Zhidu Gaige (Beijing: Laoding Renshi Chubanshe, 1984).

cism that the system undermines state planning and brings back the "capitalist labor market," proponents respond by stressing that specialists, once hired, become "co-owners" of the means of production and that the system induces specialists to improve their knowledge and to raise their productivity. By offering attractive terms, Guansu province managed to draw about 8,800 specialists from outside the province, and a poor county in Henan province invited about 700 outside technical personnel.^[30]

In July 1984, the center designated three provinces, two municipalities, and three scientific organizations to experiment with the invitation system (pingren ).^[31] Shaanxi province reportedly used this method to select the leading cadres of six provincial organs.^[32] The invitation method is widely used for choosing village and township-level cadres.^[33] Hubei province reportedly recruited 15,000 people for leadership positions at that level.^[34]

In some places, an examination is used to recruit people for low-level functionary positions in such sectors as banking, taxes, industry, commerce, prison work, and legal fields.^[35] Initially, tests were administered to only a fixed number of candidates designated by the recruiters, but the regime promises to change the method to "open invitation, self-application, firmly upholding test results, evaluation of virtue, intelligence and physical conditions, and selection of the best." The city of Shenyang has used the "open invitation and test" method to appoint leading cadres of its science and technology commission, measurement bureau, and tourist corporation.^[36] By 1987, Zhejiang province reportedly recruited 13,900 people through the examination system.^[37] Seventy-five percent of the newly recruited or promoted were from units other than where they assumed leadership positions.^[38]

[30] Jiaoyanjiu Cankao Ziliao , 1 February 1986; Guowuyuan Bangongding Diaocha Yanjiushi, ed., Zongguo Xingzheng Guanlixue Chutan (Beijing: Jingji Kexue Chubanshe, 1984), 55–81.

[31] Renmin Ribao , 29 March 1985.

[32] Shaanxi Ribao , 6 February 1985.

[33] Altogether about 60,000 were invited. Renmin Ribao , 15 October 1984.

[34] Ibid., 1 June 1985; Chiangjiang Ribao , 1 June 1985; 23 June 1985.

[35] Renmin Ribao , 8 November 1980.

[36] Jianzhou Shiyuan Xuebao , 10 April 1985; Liaowang , no. 39, 1984, 5.

[37] Beijing Review , 31 August 1987, 8.

[38] For instance, a local school teacher of economics became the head of Ningpo city's price bureau. Ibid.

Evaluation

The merit-based personnel management system requires impartial and rigorous appraisal of performance, the system which many Chinese scholars believe to be the key to solving the problems of the "iron rice bowl [and] eating from a big pot." The regular assessment that the CR had ended has been restored by the regime and is now conducted once every two years.^[39]

However, the criteria and procedures for assessing performance have not yet been systematized or standardized. Currently, cadres are assessed in four areas: virtue, ability, diligence, and achievement. The meaning of these four terms is ambiguous, and the effort to clarify them produces only more numerous subcategories and indicators. For instance, one Chinese writer divides virtue into four subsets: personality, moral and ethical standards, professional ethics, and political attitude. Each of these subcategories is further divided into three indicators. As a result, the evaluation of virtue requires the examination of twelve indicators.^[40] Similarly, ability, diligence, and achievement are divided into several dozen indicators. When more than fifty indicators are graded in terms of four categories—excellent, fine, fair, and poor—the evaluation becomes too complicated and confusing.^[41] Shanghai reportedly developed forty standards with which to evaluate each cadre, and the scores on each criterion obtained by surveying superiors, subordinates, and colleagues were coded and stored in a computer.^[42]

These multiple categories allow an inordinate amount of discretionary power to evaluators.^[43] Therefore, leading cadres frequently rely on simple criteria such as age, education, and their impression of the candidate. Despite an official emphasis on "pioneering spirit," cadres with such spirit often offend the powerful and "cannot stay on the job for a long time," but "those with ordinary ability can manage very well." The best way for the Chinese bureaucracy to manage is to take a cautious middle position on every issue: "Swimming at the upper end of the stream receives punishment,

[39] Cao Zhi, ed., Zhonghua Renmin , 174.

[40] Wang et al., eds., Ganbu Guanli Gailun , 189.

[41] The "Provisional Regulations of the Civil Servant" prescribes the same procedure for evaluation. Liaowang , 17 June 1985, 26–27.

[42] Renmin Ribao , 19 September 1984.

[43] Wang et al., eds., Ganbu Guanli Gailun , 187–94.

swimming at the lower end of the stream maintains the present position, and swimming in the middle is blessed."

As part of a planned civil service system, the regime experimented with a new evaluation system in three different units. One was a year-end assessment of political cadres in Zhejiang in 1987, and it plans to test the same method in eighteen provinces and municipalities in 1988. A guideline issued on 8 July 1987 for assessing county-level cadres followed the existing personnel management procedure. The authority to assess the county secretary and magistrate is in the hands of a special committee organized by the district-level party secretary or administrative heads.^[44] For a year-end assessment of administrative cadres, the regime plans to test a new method in Huangbu district and the commerce bureau of Shanghai. Also, some central government ministries are slated to experiment with the new evaluation method.^[45] The "Provisional Regulations of the Civil Servant" specifies that any evaluation of cadres of division (chu ) level or above should include the opinion of the masses as collected through an "opinion poll."

Responsibility System

Since Deng Xiaoping's speech in March 1983, which stressed the need to specify tasks, people, quantity, quality, and time needed to complete every project, the center has issued a series of directives and organized numerous conferences to exchange experiences to establish a system for defining responsibilities.^[46] By 1984, about 20 percent of the provincial organs and about 50 percent of the district- and municipal-level organs in twenty-three provinces had allegedly established some kind of responsibility system.^[47] By 1987 about 93 percent of districts and municipalities and 94 percent of county governments had established some kind of responsibility system.^[48]

Understandably, defining the authority, responsibility, and task for each of 27 million cadres is a mammoth undertaking in China

[44] Zhuzhi Renshixue Yanjiu , no. 2, 1988, 36–38.

[45] Renmin Ribao , 11 December 1987.

[46] Lilun Zhanxian (Anhui), 15 January 1984, 13; Sichuan Ribao , 14 March 1984.

[47] Lilun Zhanxian , 5 January 1984, 4–5.

[48] Zhongguo Renshi Guanli , no. 9, 1987, 20.

where even systematic and effective tables of organization are nonexistent. Instead of issuing general guidelines on how to set up the responsibility system, the center encourages each unit to develop its own system on the basis of its concrete conditions. Most systems reported in the newspapers include one or another attempt to delineate authority, responsibility, and the main task for each office and post, thus tying performance to some kind of reward-and-punishment scheme. Liaoyuan municipality divided its entire workload into "routine work and important work," assigning twenty points for the first and eighty points for the latter. One county categorized its entire workload into three types: specific tasks assigned by the county party committees, the task of developing spiritual civilization, and the objective of improving work style. Fulfillment of the first task deserves sixty points, and the completion of either of the other two is worth forty points.^[49]

Broadly speaking, there are four types of responsibility systems. For leading cadres whose job involves many different kinds of work, "position, responsibility, and tasks" are taken into account during assessment.^[50] Jiangyan district of Wuhan municipality, with 2,800 cadres in sixty-three units, has reportedly defined the responsibility, authority, and task for each unit and each individual. Now, each cadre keeps his "work record" which is used as a basis for assessment.^[51]

The method of contracting out specific tasks with a fixed number of people, a deadline, and points to earn is used for the leading cadres of rural areas. "Managers and other cadres of enterprises sign contracts on such specific economic indicators as production value, profit, quality, and sales, etc., and when all these targets are met, they are rewarded."^[52] This responsibility system based on contracting out for technology is used for technical persons in functional fields and basic research units.

It remains to be seen whether a written job description for each position will improve the efficiency of the Chinese bureaucracy. Although this practice marks a sharp break from the Maoist practice of stressing internal remedies based on ideological incentives,

[49] Xin Changzheng , no. 9, 1984, 30–32.

[50] Shehui Kexue Yuan (Liaoning), no. 33, 1984, 5.

[51] Renmin Ribao , 11 November 1984.

[52] Gongren Ribao , 6 June 1981.

it seems very unlikely that the regime's present one-sided emphasis on external remedies, based on quantifiable achievement records, will work. Moreover, the responsibility system can produce the desired results only when it is combined with a rational incentive system that links performance to promotion and salary increments.

To summarize, the regime initially tried to rationalize the personnel management system by first strengthening the authority of higher levels and then decentralizing, while restoring the party committee's authority over personnel matters. But all these attempts were made in a piecemeal fashion with the knowledge that the party would directly manage all cadres. Consequently, the personnel management authority still remains too centralized and very rigid—"recruiting persons are separated from managing persons and managing persons from managing affairs." It continues to be a "closed-door system": each tiaotiao and kuaikuai sets up its own system, and any horizontal coordination between the two is extremely difficult. The system, however, cannot be rationalized without resolving the broader question of how to separate the party from the government and other functional units.

Separating the Party from the Government

In post-Mao China, virtually all Chinese leaders have agreed that the past practice wherein party committees directly managed administrative and economic work should be changed. In fact, the regime has already corrected for the excess of Mao's era. The Maoist practice of having one person concurrently holding leading positions in the party and the government has come to an end.^[53] Many functional departments of the party and its core groups in government organs were abolished. The offices of such government heads as minister, governor, and magistrate have been substantially strengthened: they are less dependent on party committees at the same level, and their capacity to develop rational policy has been substantially enhanced with the increase of specialists among their staff and the establishment of numerous research

[53] Renmin Ribao , 26 November 1987.

offices. Within the state the People's Congress system has been restored, and the court system apparently enjoys more independence now than at any other time in its history.

With regard to the party committee's functions at the lower level, numerous articles published in the official media suggest the following tasks: (1) supervising the implementation of the party line and policy, (2) leading ideological and political work, (3) managing the party's internal affairs, (4) making decisions on "important matters," and (5) managing the personnel matters of cadres.^[54]

First, no one disagrees with the idea that ideological work belongs under the party committee's jurisdiction, but whether the committee should limit its work to ideological work or not is open to debate. This is especially true when the significance and relevance of this work to other functional tasks is diminishing as the regime subordinates politics to economic development, while failing either to define the meaning of "socialism" or to develop a coherent official ideology.

Limiting the party's main task to its own internal work, such as managing and educating its members, is not problematic. But a persistent question is whether the CCP, a Leninist ruling party committed to a one-party dictatorship, can afford to restrict its activities to such a narrowly defined range.

The problem with the term "important work" (generally defined as "affecting the fate of the entire people") is that it does not specify how and who will determine which matters are important in a specific case.^[55] The highly centralized bureaucratic structure in China does not leave many "important matters" for the county party committees to decide. Thus, it is not surprising that after extensive debate on what constitutes "important matters," a county party secretary and magistrate decided to consider all issues jointly.

Authorizing party committees to supervise government and enterprise units in the implementation of official decisions is not controversial. But it does not resolve the technical questions of what supervision means in concrete terms, how to supervise the operation of government agencies and enterprise units without taking over their tasks, and how much and what kind of power party

[54] Shaanxi Ribao , 26 October 1983.

[55] Hebei Ribao , 21 October 1982.

committees need for supervision. Lower-level party secretaries insist that "without substantial authority, there is no way to supervise and guarantee that work will be done correctly."^[56] Managers are also unhappy with the supervision: "If they trust secretaries and do not trust managers, it is better to make the secretaries managers." When a manager is higher than his secretary in education, there is no reason to supervise.^[57] In addition, the formula raises other complicated questions. For instance, should a party secretary be held responsible for the profit and loss of an enterprise? If the answer is no, does that mean that they are not entitled to a bonus when the enterprise makes a profit?^[58] On the issue of supervising authority, conservatives and reformers split. Reformers reject the idea of giving supervisory authority to party committees because it could constrain the manager responsibility system.^[59]

The official guideline on how to separate the party from the government at the lower level is not only ambiguous but contradictory. While criticizing party committees' involvement in functional work, top party leaders still insist on the principle of "the party leading everything." On one hand, party committees are instructed not to manage economic issues directly. On the other, superior party committees urge lower-level party committees to "spend 70 percent of their time on economic matters."

The issue of separating the party from the government and enterprises requires the redistribution of political power, thus directly touching on the political interests of party secretaries at the middle and lower levels, most of whom were recruited after 1949 from peasant and worker activists and demobilized soldiers. Therefore, party cadres exploit the ambiguity in the official policy in order to resist the official effort to separate.^[60] They often resort to defensive tactics such as foot-dragging and rearguarding. Some old party cadres are more straightforward: "If the first party secretary does not concurrently hold the first position of the government, that

[56] Lilun Yu Shijian (Shenyang), no. 13, 1985, 34–35; Tianjinshi 1984 nian Shehui Kexue Keti Diaoyan Chengguo Xuanbian (Tianjin Renmin Chubanshe, 1985), 11:411–27.

[57] Xuexi Yu Yanjiu , no. 11, 1985, 15–17.

[58] Hubei Caizhengxueyuan Bao , no. 4, 1985, 41–42.

[59] Lilun Tantao , no. 2, 1987, 76.

[60] Zhengzhi Yu Xingzheng Yanjiu , November 1985, 18–32.

means that the party will not have power, and the idea of the party leading everything becomes an empty slogan." Others play on the fear of "dispersionism and independence [of the government] from the party."^[61]

Results in County Governments

Numerous investigation reports prepared by Chinese scholars after their fieldwork at the county level indicate that party committees still dominate administrative organs. Usually, the county party committees and their government counterparts jointly issue orders and convene meetings. County party committees still discuss a wide range of issues from economics to party affairs, spending most of their time on economic issues. Although they are supposed to meet only once a month, they usually meet more frequently. On the other hand, "the management committee" of the magistrates—which is supposed to meet weekly—rarely convenes because "there is not much left for the government to decide, and the management office of the government is redundant."^[62]

In addition, magistrates, who are usually deputy secretaries subordinate to the first party secretary in the party hierarchy, now act more like representatives of the party committee rather than heads of the government unit. The party core group (dangzu ) rather than the directors continue to exercise the real power in county government bureaus. Consequently, "the party secretary presently does the work of the magistrate, and the magistrate does the work of the director of county bureaus."

Party committees continue to dominate not only the executive organs but also the People's Congress and its standing committees, which theoretically are representative organs of the entire people. Leaders of the standing committees (changwei ) of the People's Congress are organized into party cells, which are subordinate to party committees. Most standing committee members are semiretired party cadres. As a result, many Chinese satirize the county standing committee of the People's Congress (which they call "the nurs-

[61] Jiefang Ribao , 30 October 1983.

[62] Jingji Yanjiu Cankao Ziliao , 25 April 1986, 32–39.

ing home") by saying it is "big in three items and empty in one item": its constitutional right is big, the sign board of its office is big, the age of its members is big, but its power is empty.^[63] Party committees, not the People's Congress, prepare economic plans. And if the standing committee rejects the plan, the party committees can overrule it.^[64] Until recently, the party's disciplinary committees investigated government corruption. According to a Chinese expression, "the county party committee writes the scenario, the government performs, the People's Congress watches, the political consultative conference comments, and the disciplinary committee judges."^[65]

The turf fight is particularly fierce over economic issues, which make up 60 percent of government tasks. At the moment, whether or not a county party secretary makes a decision on economic matters largely depends on the personality of the secretaries and managers. Whenever a secretary is capable and enjoys high prestige, even magistrates look to him to make all major decisions. Thus, it seems that appointing able persons to secretaryships, ironically, tends to hinder the separation of the party from the government.

Since there is no separation, the question of how to share personnel management authority is premature. County party committees have not relinquished their prerogative over personnel management. According to an investigation by Chinese scholars, county party committees manage all the important leading cadres of first-class organs (ministries [bu ], commissions [wei ], bureaus [ju ], and villages [xiang ]) and second-class organs (sections [tuan ] and certain enterprises), or they handle all the leading cadres of the first-class organs while allowing the personnel bureau under the magistrates to manage the cadres of the second-class organs. Believing that managing cadres is their most important responsibility, county party committees spend most of their time on personnel matters.^[66] Although its authority over certain government cadres is constitutionally guaranteed, the People's Congress remains a rubber stamp.

[63] Zhengzhi Yu Xingzheng Yanjiu , November 1985, 18–32.

[64] Jingji Yanjiu Cankao Ziliao , 25 April 1986, 32–39.

[65] Makesi Zhuyi Yanjiu , no. 4, 1986, 7.

[66] Jingji Yanjiu Cankao Ziliao , 25 April 1986, 32–34.

Results in Enterprises

The situation in enterprise units is much better than on the county administrative level. Despite the strong opposition of party cadres, the manager responsibility system has been gradually implemented in most state-owned factories in the past few years, thereby substantially weakening the party secretaries' power.

However, the manager responsibility system has not yet completely resolved the controversy over personnel management authority, despite bitter complaints from managers that the existing system constitutes the most serious obstacle to the efficient management of enterprises, more serious than the shortage of raw materials.^[67]

Reformers advocate fenggong guanli , that is, authorizing managers to manage administrative cadres, while limiting the party committee's authority regarding party cadres in order to make the managers' authority and responsibility coincide and to enable them to exercise "uniform leadership with regard to production, management, and administrative management."^[68] Viewing the proposal as a serious threat to the party's prerogative over personnel matters, the conservatives raised objections at every stage of the development of the manager responsibility system.

The battle over the issue is fought on three specific questions. The first is to what extent party committees will share personnel authority with managers. The second is the managers' right to "form cabinets." The third concerns jurisdiction over middle-level administrative cadres.

The conservatives do not oppose the idea of giving some personnel authority to managers because they are well aware of the shortcomings of the past, but they still insist on the principle of "the party managing cadres" and have tried to keep as much authority as possible in the party committees. The central organizational department usually sides with the secretaries by reiterating that the party committee is responsible for the "recruitment, alloca-

[67] Zhongguo Jingji Tizhi Gaige Yanjiusuo Conghe Diaochazu, ed., Gaige: Women Mianlin de Diaozhan Yu Xuanzhe (Beijing: Zhongguo Jingji Chubanshe, 1986), 137.

[68] Zhengzhixue Yanjiu Tongxung , no. 4, 1983, 7–13.

tion, evaluation, and supervision of cadres."^[69] Pervasive cadre corruption strengthens the conservatives' position that party committees—as collective decision-making bodies under the tighter supervision of the upper echelons—rather than the one-person manager are more suitable for safeguarding the state's interests and for screening the political quality of cadres, preventing those "lacking loyalty to the revolutionary business" from being promoted.^[70] By contrast, reformers insist that "the party managing cadres" does not mean that party committees manage everything directly; rather it means that under the principle of the party's "unified management," some concrete work can be delegated to government and enterprise units.

Once the regime decided to give managers personnel authority, the next question is to what extent the party committee should be allowed to influence a manager in exercising his authority over personnel matters. The reformers want to prevent the party committee from interfering with the managers' authority, but they cannot state their position in such a blunt way, lest it be construed as an open challenge to the CCP's authority. Instead, they argue that managers should use their authority wisely, paying attention to the opinions of the masses and the party committees. By contrast, conservatives try to guarantee the party committees' influence in personnel matters. One compromise requires managers to consult the "management committee" on personnel matters. Since the committee usually includes members of the party committee, as well as the disciplinary committee and labor unions, both of which are under the party committee, this option allows the secretaries to continue to exert influence over personnel matters. In fact, whenever the party committee objects to the choices made by managers, there is no way for the manager to prevail. The party secretaries still retain veto power.

Pushed by reformers, the regime publicized the idea of allowing managers to "form their own cabinets" for a while. Several differ-

[69] Guojia Jingwei Jiceng Zhengzhi Gongzu Bangongshi, ed., Shixing Changzhang Fuze Zhi Hou Qiye Dangwei Ruhe Gongzhu (Beijing: Jiefang Zhengzhixueyuan Chubanshe, 1985), 37–49. See also Gaige Qiye Lingdao Zhidu Shixing Changzhang Fuzezhi (Beijing: Nengyuan Chubanshe, 1986), 34–42.

[70] Guojia Jingwei Jiceng Zhengzhi Gongzu Bangongshi, ed., Shiping Changzhang , 21–36.

ent methods of "forming cabinets" were practiced between 1984 and 1985. The manager either nominated or appointed deputy managers as well as all heads of administrative positions, who in turn nominated their own deputies and ordinary cadres. Or managers appointed deputy managers, who in turn appointed administrative heads, who selected their own deputies and ordinary staff. Another method required the approval of the organizational department of the upper echelon and the workers' congress of the enterprise to finalize the appointment.^[71]

The procedure for forming a cabinet included the following steps. First, the manager, the party secretary, and the chairman of the relevant labor union discussed the matter in advance and set up the method of forming the cabinet. Second, they formulated the criteria to be used for cadre selection. Third, the manager solicited comments from various sectors, including ordinary workers. Although the party secretary and labor unions were involved in the process, it was the manager who made the final decision. In this sense, the system marked a departure from past practice.

To the conservatives, allowing the manager to form his own cabinet clearly violated the principle of the "party managing cadres" and precipitated factionalism.^[72] Moreover, the practice threatened the personnel authority exercised by the party committee one level higher as specified in the nomenklatura system. Despite the reformers' vigorous defense, which sometimes twisted logic, they apparently lost the battle on this particular issue. The regime finally decided not to use the term "forming a cabinet."^[73] Apparently, the central organizational department believed that the practice of managers choosing their own deputies posed too serious a threat to the entire personnel management system.

Over the third question—the manager's authority over middle-level administrative cadres—the reformers apparently obtained what they wanted.^[74] Whether or not it was due to Zhao's insistence, the regime announced revised regulations, which, among other things, specified the manager's authority over middle-

[71] Jingji Gongzuo Zhe Xuexi Ziliao , 1985, 24, 8–31.

[72] Ibid.

[73] Xuanchuan Shouce (Beijing), July 1985.

[74] Zhengdang Yu Jiandang (Liaoning), 1 June 1986, 30.

echelon cadres, although it allowed the party secretary to make suggestions.^[75]

To summarize, it seems that a compromise has been reached: although they failed to obtain the right to form their own cabinets, managers have the authority to manage middle-level administrative cadres. However, the controversy over personnel authority is likely to continue because party committees are not clearly excluded from managing even middle-level cadres. It is officially recognized that party committees are responsible "for observing cadres, examining the political qualifications of candidates, organizing discussion meetings, and carrying out public opinion polls in order to provide accurate information to managers."^[76] Managers are granted authority over certain cadres, but their authority has to be used in consultation with the party committees: "On the matter of using persons, managers should rely on the party committee, should rely on the masses. . . . Managers should listen to the opinion of the masses, and then decide."^[77] Consequently, many party committees and their organizational departments refuse to transfer personnel authority to managers, who bitterly complain. Despite resistance, the general trend is to weaken party leaders' authority in enterprises. One indication of this trend is the fact that jurisdiction over party committees in enterprises is being transferred from committees of government bureaus (tiaotiao ) to territorial committees, a transfer that helps limit the work of the enterprise party committee to managing party members.^[78]

Planned Civil Service System and its Implications

The Chinese personnel management system was originally copied from the Soviet Union. However, although the USSR has modified its system, China's has remained relatively unchanged.^[79] Because any debate on the system is bound to question the party's lead-

[75] Zhongguo Renmin Gongheguo Gongbu , no. 1, 30 January 1987, 9–12.

[76] Guojia Jingwei Jiceng Zhengzhi Gongzu Bangongshi, ed., Shixing Changzhang , 47.

[77] Zhengdang Yu Jiandang (Liaoning), 1 June 1986, 30.

[78] Zhengzhi Tizhi Gaige (Shanghai: Shanghai Renmin Chubanshe, 1988), 43.

[79] Zhengdang Yu Jiandang (Liaoning), no. 3, 1984; 30 June 1984, 1–4.

ership, the public did not demand reform until 1986, when the regime itself publicly recognized the need for structural reforms. To many Chinese, changing the personnel management system was the main goal of political reform: "Whether or not the structural reforms can be carried out and their results consolidated is largely dependent on cadre reforms."^[80] The need to utilize scarce scientific and technical cadres efficiently further justified demands for a change in the Maoist system.

At the Thirteenth Party Congress newly elected General Secretary Zhao Ziyang announced the regime's plan gradually to introduce a civil service system. According to this plan, a mere 4.2 million out of the existing 29 million cadres—only those who were employed in government agencies—will be classified as civil servants.^[81] In other words, cadres employed in both the judicial and the legislative branches of state organs as well as those working in the party, enterprises, and business units will be excluded from the civil service system.

All leading cadres who legally constitute "governments," ranging from the center to county levels, and whose appointments require the approval of a People's Congress will be classified as "political" (zhengwu ) civil servants. Thus, this category will include the premier, state councilors, and ministers of the central government, governors and vice governors, mayors and deputy mayors, magistrates and deputy magistrates, and some bureau directors.^[82] Regarded as "policymakers," political civil servants will be elected for fixed terms as specified in the existing constitution and in the organizational laws of local governments; therefore, they lose "life tenure," although there are no laws limiting the numbers of terms they can serve.

The rest are functional (ye wu ) civil servants, "the implementors of policy," who are further classified as either "administrative" or "specialized technical." Their selection will be made through "open, equal, and competitive" examinations. Candidates for examinations are limited to graduates of "higher educational institutions" who are twenty to thirty years old and who "support the

[80] Zhengzhixue Yanjiu Tongxun , no. 4, 30 December 1983, 4–7.

[81] Renmin Ribao , 31 October 1988.

[82] Tan Jian, ed., Guojia Gongwuyan Shouce (Beijing: Shehui Kexue Wenxian Chubanshe, 1988), 7.

leadership of the CCP [and] love the socialist motherland." Possessing "certain political conditions" is a prerequisite, but many writers insist that party membership is not required. Specialists will be selected from among those possessing certificates demonstrating specialized technical training. All those who pass the examination are required to go through one year of training at the basic level. Only after a successful training period can they become regular civil servants.^[83]

Moreover, these civil servants are to be promoted grade by grade. To assume leading positions, one has to meet additional requirements: section chiefs have to have an educational level higher than senior high school and at least four years of work experience; division chiefs (chuzhang ) need a college or specialized middle school education and seven years of work experience; bureau directors require a college education and ten years' work experience; and ministers need a college education and at least fifteen years of work experience.^[84] In addition, before assuming these positions, civil servants must receive further training at administration institutes that the government is setting up at various levels.

A provisional regulation guarantees job security by ensuring stable working conditions, reasonable pay, and protection from arbitrary dismissal. Besides the standard rights and duties listed in similar laws of other countries, the regulation also stresses the career civil servants' right to "criticize and make suggestions to their leaders" and the duty to "carry out the party's and the state's line, directives, and policies."^[85]

The regime plans first to experiment with such a system in several central government ministries, provinces, and municipalities and then to implement it in all the state organs.^[86] The first step in establishing the system is to classify positions, which will determine the task, responsibility, and pay scale of each cadre. At the moment, the regime estimates that it will take five years for these five units to set up "the frame" of the civil service system and an additional ten years to implement the system nationwide.^[87]

[83] "Temporary Regulation of State Civil Service System."

[84] Ibid.

[85] Ibid.

[86] China Daily , 20 July 1988.

[87] Renmin Ribao , 31 October 1988.

This provisional regulation fails to resolve the most crucial obstacle to creating an efficient personnel management system—the party domination over the government. On this question the provisional regulation itself contains many ambiguous and contradictory points. For instance, it does not grant political neutrality, even to career civil servants. On the contrary, they are specifically required to demonstrate a certain level of "socialist consciousness" and to follow the political leadership of the Communist Party.^[88] However, many Chinese scholars argue that although Chinese civil servants are required to be loyal to the party, the regulation protects them from future policy fluctuations resulting from leadership changes or power struggles among the top leaders. Insofar as the career civil servant faithfully carries out any policy directed by the party, he will not be held responsible for its correctness or incorrectness. This will be an improvement over the situation during the CR when many low-level cadres were purged because they had implemented "revisionist policies" handed down to them through legitimate channels of hierarchy.

Another ambiguity in the regulation and official writings concerns the party committee's authority over personnel matters; on the one hand, it will give up its prerogative over cadres, but, on the other hand, it will continue to exert a substantial amount of authority over political civil servants and "important cadres."^[89] According to Zhao Ziyang, "the party center and the local party committee will nominate political civil servants to the People's Congress at various levels and will supervise and manage political civil servants with party membership."^[90]

A particularly troublesome aspect of the regulation is the role of the party's organizational departments. Although core groups in government agencies were abolished, the central organizational department still keeps a party core group in the personnel ministry. Furthermore, it is likely that as long as the party maintains organizational departments along the administrative hierarchy, the turf fighting between the party and the government over even the functional civil servants will continue. If the provisional regulation

[88] Tan Jian, ed., Guojia Gongwuyan Shouce , 10.

[89] Nie Gaowu, Li Yichou, and Wang Zhangtian, eds., Dangzheng Fenkai Lilun Tanlu (Beijing: Chunqiu Chubanshe, 1988), 17.

[90] Su Yudong, ed., Guojia Gongwuyuan Zhidu Jianghua (Beijing: Laodong Renshi Chubanshe, 1988), 13.

is thoroughly implemented, the party organizational departments will have nothing to do except manage their own cadres and the "political civil servants," which altogether would number less than a thousand. The party may claim jurisdiction over party member officials, as indicated by Zhao Ziyang. If so, an interesting question would be whether or not the party's claim would be limited to the "organizational life" of party member officials, while the government manages their "work relations." The party may argue that unless it has authority over employment conditions of party member cadres, it will not be able properly to supervise their "organizational life." To make matters more complicated, each cadre's personnel dossier includes materials on his party membership, as well as his work relations. Unless the dossier is broken down into two parts, each of which is managed by either the party or the government, however, the organizational departments would have an excuse to avoid relinquishing their authority over personnel matters.

Having agreed to share its power over personnel with other appropriate authorities—be they government leaders, managers, or administrative heads of business units—the party is extremely reluctant to give up its influence over cadre management completely. Now, instead of directly managing cadres, the party is helping other appropriate authorities make the right decisions by recommending qualified persons and supervising their handling of personnel matters. This change has taken place at a time when the party is relinquishing direct control over economic resources. Any weakening of the party's control over the economy makes it more difficult to resolve the persistent question of how to change the structure of personnel decision making in such a way that the new system can maintain a balance between two conflicting goals: ensuring the party's control while managing the cadres efficiently, thus making the system flexible but not creating chaos, and preserving a national planning perspective while allowing lower units to be autonomous even in personnel management.

As noted, the result of the official effort to separate the party from the government has so far been rather limited. In structural terms, the regime completely separated the party from the government and other functional units. But the official formula, in which the party "ensures and supervises" the work of administrative heads, has failed to clarify ambiguities in terms of power distribu-

tion. The most important reason is that the party has not yet resolved important theoretical as well as practical questions. The theoretical questions are what role basic-level party committees will assume and how they will justify their continuing existence when the Leninist Party has declared an end to class struggle and has adopted efficiency as the ultimate criterion for evaluating organizational performance. The practical question is how to maintain the Leninist Party's prerogatives while allowing autonomy in government agencies and economic enterprises in order to promote economic development.

A possible option followed by many Eastern European countries is to abolish the party committee system in functional areas. As many liberal reformers argue, these are "unnecessary political organizations detrimental to efficient economic management."^[91] But the Leninist Party in China is not yet ready to do so. The other option is for factory managers to hold concurrently party secretary positions. This seems to be taking place in many enterprises. In some enterprises selected for experimental purposes, managers take charge of even ideological and political works, thereby depriving party secretaries of using their authority over these works to interfere with managers' decisions. This experiment is obviously designed to weaken the party secretaries' power and eventually to abolish the party committee system in economic enterprises.^[92] However, at present in most enterprise and business units, the dual structure of the administrative and functional hierarchy on the one hand and the party hierarchy on the other still exists.

The continuing existence of this dual structure raises another puzzling question: what qualifications and career backgrounds are necessary to be appointed to the party and other administrative positions? Different career experiences and personal backgrounds among party secretaries and managers can aggravate conflicts.^[93] But, at the same time, using the same criteria for selecting party secretaries and administrative heads and managers makes the dual system only redundant without any tangible benefits.

Available data on this question indicate that career differentia-

[91] Guojia Jingwei Jiceng Zhengzhi Gongzu Bangongshi, ed., Shixing Changzhang , 35–36.

[92] Fazhi Ribao , 26 September 1988.

[93] Lilun Tansu , no. 2, 1987, 31–36.

tion between the two positions is narrowing. According to a survey of 900 enterprises conducted by the Institute for the Reform of the Chinese Economic Structure in 1985, when the initial readjustment of the enterprise cadres was completed, managers were found to be slightly younger than their counterparts in party committees. The average age of managers is forty-five, whereas that of secretaries is forty-seven.^[94] As shown in table 56, managers are better educated; indeed, they are the best educated; the second best educated are deputy managers; third deputy secretaries; and least are the secretaries and advisers. These differences are quite small. Since the educational level of cadres is closely correlated with age—the younger are generally better educated—the differences in education between the two groups may reflect age rather than job differences.

Most managers and deputy managers are promoted from technical, functional, or lower-level managerial positions: 85 percent of them have experience in technical, 72 percent in functional, and 71 percent in basic-level managerial work. By contrast, secretaries and deputy secretaries are largely selected from administrative ranks (54 percent) and from among demobilized soliders (71 percent).^[95]

[94] Zhongguo Jingji Tizhi Gaige Yanjiusuo Conghe Diaochazu, ed., Gaige: Women , 270–305.

[95] Ibid.

Despite career differences, there is a powerful trend that may mitigate the conflict between managers and secretaries in the long term; that is, the career backgrounds of party secretaries are changing. Not only has a new type of enterprise cadre emerged, but the desire to obtain professional knowledge is pervasive even among party secretaries. When enterprise leaders were asked about the type of qualifications that enterprise cadres should have, many cited functional over political knowledge. Enterprise leaders were also keenly aware of their own shortcomings, that is, they lacked technical expertise.^[96]

Another mitigating factor for possible conflict between party and administrative cadres is that the actual leadership type is also changing from what the Chinese call the "personality and virtue type" (pingde xing ) to the "ability type" (nengli xing ). As noted in chapter 11, bureaucratic technocrats are replacing revolutionary cadres even in the highest political positions. The official ideology is in disarray as a result of economic reforms. New types of party members are being recruited. The regime has been co-opting experts in decision-making bodies, thus providing tangible incentives for the transformation of party secretaries. Many informants report that able party cadres are eager to move into government positions. The younger generation tends to embrace the values and outlook of the technocrats, not the revolutionary cadres. Thus, it seems that revolutionary cadres cannot find successors like themselves.

Despite many uncertainties, after thirty years of revolutionary turmoil, which entailed the recruitment of cadres on the basis of political qualifications, China has finally come to emphasize ability by recruiting government officials through open and competitive examinations. The fact that the regime is planning to introduce a civil service system indicates its willingness to end the Maoist practice of selecting officials through mass campaigns and on the basis of class backgrounds, and instead to develop rational, efficient, and competent administrative bureaucrats. The term "civil servants" itself implies that government officials are "public servants" who will manage public affairs as the guardian of public interests, whereas the term "cadres" refers to those with certain leadership

[96] Ibid.

abilities necessary to lead the masses in revolutionary struggles for social change. The term "civil servants" also connotes a merit-based recruitment, whereas the cadre's role requires political skills and ideological consciousness. The introduction of a planned civil service system also represents the regime's intention to separate government employees from party cadres, thus ending the previous practice of managing the party and the government cadres together (dangzheng ganbu ) in some manner and allowing the government to manage its employees. In fact, the provisional regulation stipulates that the "personnel ministry and bureau," will manage the civil servants [performing functional works] "according to laws."^[97] The recruitment of government officials through the examination system also rules out the possibility of any input from the populace along the mass line model emphasized during Mao's era.

Apparently, the regime is preparing a separate set of regulations regarding party cadres, enterprise unit cadres, and business unit cadres. Although the contents of these regulations are not known, the planned regulations for party cadres will probably stress election or some other method that allows rank-and-file party members to select their leaders. For the selection of enterprise and business unit cadres, the planned regulations will probably try to blend different methods, such as "appointment, examination, invitation, and election." However, it is extremely unlikely that the regulations will explicitly deny the party any voice in personnel management. As long as the concept of public ownership continues, the party will consider itself justified in retaining some authority over personnel management.

The impact of introducing a civil service system is not limited to the party-state's bureaucracy. As many studies of Western European bureaucracies have demonstrated, a merit-based recruitment of administrative elites tends to offer undue advantage to the middle class.^[98] As the competition for government employment shifts in emphasis to ability, educational institutions will become the springboard for entering official positions. China's new education-

[97] "Temporary Regulation of State Civil Service System."

[98] Ezra N. Suleiman, Politics, Power, and Bureaucracy in France (Princeton: Princeton University Press, 1974), 72–99.

al policy of attaching priority to academic achievement and spending large amounts of the state budget for key schools will accelerate the stratification of educational institutions. Middle-class students have the advantage of entering better high schools and then highly competitive colleges, from which they can land cadre positions.^[99] Actually, middle-class values are more congruent with the role of technocrats than that of revolutionary cadres.

In this way, the state bureaucracy will lose any claim to representing a broad range of social classes, but it can still be responsive to society.^[100] This nondemocratic characteristic will, however, be compensated when the regime steps up its effort to professionalize state employment. The rigorous training of civil servants, the development of an esprit de corps, and a trend toward professionalization will make the administrative elite act more as guardians of the public interest than as representatives of their original middle-class backgrounds.

As the evolution of Western European civil service systems demonstrates, creating an efficient state bureaucracy is the first step toward reducing the patrimonial power of a monarchy. In Prussia, the absolute monarchy created a civil bureaucracy largely staffed by the landed aristocracy. As the bureaucracy gained power and status, joining the civil bureaucracy offered upward mobility for the newly emerging bourgeoisie. Meanwhile, by claiming to represent the public interest, the civil bureaucracy gradually became independent of the monarchy. As the monarchy collapsed, it was replaced by elected politicians, who, on behalf of the people, supervised the administrative bureaucrats, while the administrative structure began resembling Weber's ideal type of bureaucracy, one staffed by specialists and neutral to the political process.^[101]

In China, the party-state created a large corps of cadres as a tool for revolutionary change. During Mao's era, the party maintained tight control over this group, preventing them from developing into an administrative instrument. Now, the regime is trying to transform the cadre corps into administrative bureaucrats by granting them more autonomy. In this sense, the introduction of a civil

[99] Ibid., 79–92.

[100] Ibid., 386.

[101] Hans Rosenberg, Bureaucracy, Aristocracy, and Autocracy (Boston: Beacon Press, 1958).

service system—even though at the moment the civil servants' political neutrality is not yet recognized—may be the first step toward separating the party from the administrative hierarchy.

However, such questions as what controls the party will maintain over the administrative bureaucrats and to whom the latter will be held accountable still persist. Since in most non-Communist countries, democratically elected politicians supervise the administrative bureaucrats, one may predict that the party leaders at various levels will act likewise. In other words, the party leaders will become "political managers" and "powerbrokers."