State Capacity

Definition

State capacity is one of the defining characteristics of any political system (Almond and Powell 1966; Katzenstein 1978; Zysman 1983; Migdal 1988; Ikenberry 1988; Organski and Kugler 1980). By "state capacity," I refer to the ability of a government to administer its territory effectively (Skocpol 1985). In the modern world, the survival and functioning of a political system depends on four basic state capacities: the capacity to mobilize financial resources from the society to pursue what the central policymakers perceive as the "national interest" (extractive capacity); the capacity to guide national socioeconomic development (steering capacity); the capacity to dominate by using symbols and creating consensus (legitimation capacity); and the capacity to dominate by the use or threat of force (coercive capacity).

These four capacities are conceptually distinct but interrelated in practice. For instance, the legitimation of a regime is dependent on its performance. If the state is able to produce and deliver economic and social goods at the level its subjects expect, or at least as its rulers promise, it should have no legitimacy problem. On the other hand, if the state apparatus cannot adequately steer the economic system, this is likely to result in a decline in its legitimacy. With legitimation capacity, the state can effectively steer activities without the necessity of constantly deploying coercion. Without legitimacy, however, the state would find it much more difficult to extract resources from the society, and would have to bear much higher costs for maintaining law and order (Habermas 1975). An overloading of control problems would weaken the state's coercive capacity, which in turn would lower the expected cost of joining the opposition. Of course, the decline of the four capacities rarely occurs simultaneously or in the same sequences in different cases. But their changes do tend to reinforce one another. Once overall state capacity falls below a certain threshold, the regime would be in serious trouble, if not in danger of collapse.

Measurement

State capacity is more easily asserted than measured. Indeed, the concept of overall state capacity as discussed above is too complicated for ready measurement. In this chapter, extractive capacity is therefore selected as a key indicator of overall state capacity. The substitution is based on the assumption that an effective political system should be able to extract resources,

aggregate them, and use them for national purposes; a government that is unable to generate sufficient resources for realizing its policy goals is less effective. Extractive capacity is arguably the most fundamental of state capacities. The availability of resources permits the state to carry out its other tasks. Many empirical studies have shown that the government's political-capacity-as-fiscal-extractive-capacity provides a useful analytic tool for explaining such divergent phenomena as the outcomes of wars among major powers (Organski and Kugler 1980), demographic transitions (Organski, Kugler, Johnson, and Cohen 1984), the probability of developing countries' suspending their external debt-service payments (Snider 1990a), and domestic political violence (Snider 1990b).

To construct an indicator of the state's extractive capacity, we need two points of reference. The first is the absolute value of the actual revenue captured by the public sector. The other is gross domestic product (GDP) or national income, which measures the total volume of the nation's financial resources. The two reference points enable us to define state extractive capacity as the ratio of the former to the latter:

Since extrabudgetary funds are an additional source of public expenditure and investment in China, they should be included when we measure the size of the public sector. Definition II may thus provide us with a better tool for measuring the extent to which the national income is captured by the public sector in China.

The Determinants of State Capacity

As a dependent variable, what are the determinants affecting longitudinal variation in the extractive capacity of the state? What factors influence the executive's ability to allocate resources in pursuit of its policy objectives? In discussing the conditions underlying the effectiveness of the state, emphasis has been laid on "state autonomy." Many believe that the state as a corporate entity is unlikely to be capable of formulating goals or implementing them unless it is somewhat insulated from the surrounding social structure. Here societal resistance is considered the primary obstacle to the effectiveness of state actions (Krasner 1978a).

While there is no doubt that state autonomy is necessary for effective state intervention, it is mistaken, however, to think that autonomy is in itself sufficient for effective state action, because the state is not monolithic. Instead,

it has a complex structure both horizontally in terms of numerous ministries and vertically in terms of different levels of government. Central ministries and local governments, although supposedly the rulers' agencies, do not always share their principals' aims. It is probably more realistic to assume that they have goals other than those imposed by the rulers. If they do, they could be competing extractors. In analyzing political constraints on the extractive capacity of the state, we thus need to distinguish between resistance by private citizens who are directly affected by the extractive policies of the state, and competition among state agents who desire control over the resources already extracted in the name of the state. While resistance occurs mainly at the extraction stage, competition is more likely to appear at the allocation stage (Lamborn 1983).

Having intentionally destroyed all competing societal centers of power, the communist state is no doubt much more insulated from societal pressures than its counterparts in the West or in the Third World. Authoritarian statism does not, however, correspond to a univocal strengthening of the state. To enforce authoritarian rule, the state has to construct an elaborate set of agencies throughout the nation. These agencies may over time develop their own preferences. The unique preferences these agencies assume create in them centrifugal tendencies. These state agencies could become "power centers" competing with central rulers and threatening the coherence and the stability of the state as a corporate whole. The principal weak point in the state socialist system is thus more likely to lie "within the regime itself, within the apparatus of the state" than "outside it in its relations with civil society" (Schmitter 1975).

Distance usually makes it much more difficult for national leaders to monitor local governments' activity than that of central ministries. This study will therefore focus on subnational governments.^[1] The extent to which local governments are able to compete with the rulers over resources, and thereby weaken overall state capacity, depends on two variables: their preferences and the resources already at their disposal. The desire to compete depends on how far local governments' preferences deviate from the center's. When local governments have no preferences of their own, or their preferences largely converge with those of the rulers, we should expect subnational governments to act as central decision makers' deputies, implementing

[1] For the sake of convenience, this study treats the central government as a homogeneous entity, represented by what Levi calls "the rulers" (1988), what Zysman calls "the national political executive" (1983), what Krasner calls "central decision makers" (1978b), and what Chinese call "the center" (zhongyang ). It needs to be emphasized that, composed as it is of hundreds of agencies and thousands of bureaucrats, the central government as such is also internally fragmented; and that conflicts between different central agencies, like those between the central government and local governments, could compromise the ability of the central government to pursue coherent policies.

whatever orders they receive from the center. Local governments would not seek to capture resources for themselves unless they have interests that diverge from their superiors'. However, it is one thing to have a desire to compete, while it is another to have the ability to do so. The ability to compete is a function of local governments' control over relevant power resources and their ability to influence central decision making. Local governments could improve their situations by employing the resources already under their control to gain control over other (and more) resources.

The distinctive sense of identity and the independent source of resources tend to reinforce each other. The growing consciousness of self-interest is likely to enhance local governments' desire to grasp more resources into their own hands; with more resources under their control, subnational governments are inclined to develop new independent preferences. The process of mutual reinforcement could result in a centrifugal tendency among local governments.

Since the state is not a monolithic entity, we need to modify definitions I and II. These definitions assess the extractive capacity of the state by breaking the state into its components, assessing the capacity of each part, and then summing the capacity of all components for the total capacity of the state. But, as often happens, the value of a whole is sometimes less than the sum of the values of its parts. When local governments pursue goals running contrary to the center's interest, for instance, the financial resources at their disposal will undermine rather than enhance the overall capacity of the state.

I am not suggesting that local governments always use their share of budgetary income and extrabudgetary funds under their control to pursue interests at odds with the center's. Nor should we assume that every yuan extracted in the name of the state serves to strengthen state capacity. It should be clear by now that when I speak of state capacity, I am referring to the aggregate capacity of the state as a corporate whole, which is definitely weaker than what definitions I or II suggest. How much weaker? There is no way to give a precise answer. In fact, we do not need a precise answer so long as we are able to gauge a longitudinal trend. For these reasons, I give the following three definitions to approximate the real extractive capacity of the central state. Definitions III, IV, and V are based on a truism: at least, central planners are able to employ the center's share of budgetary revenues at will to pursue what they believe to be the national interests.

Definitions III, IV, and V are indispensable supplements to definitions I and II. When the preferences of the central and local governments are convergent, definitions I and II should be able to serve as a good indicator of state extractive capacity. However, definitions III, IV, and V better measure state extractive capacity when the preferences of the central and local governments diverge.