|
|
|
|
Dilemmas of Reform in Jiang Zemin’s China, by Andrew J. Nathan, Zhaohui Hong, and Steven R. Smith (eds.)
4. Institutional Change and Firm Performance
Since the beginning of economic reform in 1978, China’s economy has been growing rapidly; the economy increased approximately fivefold from 1978 to 1995. 1 It is generally accepted that the fundamental cause of this remarkable economic performance has been the institutional change precipitated by reform. But how institutions change and how institutions affect performance in this transition have not been adequately examined.
The significance of studying institutional change and performance goes beyond China. Socialist economies (dominated by public ownership) still account for a large share of gross domestic product (GDP) in the world. In no country and at no time have these socialist economies performed well. 2 Thus one of the most important problems of our times is how to reform socialist economies.
This chapter will first review the theoretical relationship between institutions, institutional change, and economic performance. It will then briefly describe the evolution of China’s economic transition, followed by the presentation of a theory of endogenous institutional change that examines the driving forces behind institutional change, how institutions change, and the consequences of institutional change on the economic performance of firms. It will next focus on the construction of a theory of the determinants of firm performance that evaluates how much institutional factors account for firm performance, and estimates the role and magnitude of each factor in influencing firm performance in China. Our study finds that while economic factors are the most significant, institutional change is also critical, and government intervention is playing a significant role in firm performance in China’s transition. On the basis of these findings, it concludes with a discussion of theoretical contributions to institutional change and firm performance in China’s transition, as well as the policy and managerial implications of our studies.
Institutions, Institutional Change, and Organizational Performance
According to Douglass C. North, 3 institutions are the rules of the game in society that govern the interactions among organizations (and individuals), whereas organizations are the players in the game trying to use the opportunities created by the institutions to maximize their welfare. The government is a also a player in the society just like other players. However, there is one important distinction between the government and the other players: the government can set the rules of the game. Institutions include formal and informal constraints that shape human interactions. Informal constraints are conventions and codes of behaviors developed from culture, whereas formal constraints are designed by a society explicitly to regulate social interactions, such as the constitution and other laws. Institutions affect the performance of an economy by their effect on the costs of exchange and production. As they compete and interact, some organizations will perceive opportunities to increase their returns by altering existing institutions and will push for institutional change. However, it is not always the case that those entrepreneurs in the political and economic organizations that initiated change will achieve what they intended. Indeed, many policies have unintended consequences. Institutions explain why a substantial divergence exists among economies, and are thus the underlying determinants of the long-run performance of economies. Western market economies and socialist (or communist) economies provide a stark contrast in terms of institutions, institutional change, and economic performance. According to North, in a social environment where markets are competitive and there are no increasing returns to institutions, institutions do not matter. 4 If markets are incomplete, transaction costs are very high, poor performance will prevail, and institutions do matter. Western market economies come closer to the former description, and socialist economies to the latter description. None of the former socialist economies that are now undergoing reform had competitive markets, and they lacked incentives for efficiency in economic activities. Consequently, they consistently performed poorly. The contemporary reforms aim at establishing institutions that are conducive to strong economic performance.
Janos Kornai’s seminal analysis of socialist economies provides a theoretical foundation for the institutional approach to studying economic performance in planned and, later, transition economies. 5 According to Kornai, failure of a planned economy is due to the fundamental institutional arrangement whereby the government is the owner of the enterprises. The government has many important objectives that conflict with maximizing profits. These objectives range from providing full employment and maintaining low-cost or free social services to distributing scarce materials and goods, the shortage of which is characteristic of planned economies. Consequently, firms in planned economies have no incentive to make profits.
Since enterprises are indispensable to the government not because they earn profits, but because they provide employment, secure certain outputs, and shoulder many social services, the government cannot shut them down simply because they are unprofitable. Instead, in order to keep the money-losing firms operating, the government must take funds from profitable firms to subsidize the unprofitable firms—like a father who forces the well-to-do brother to help out his poorer siblings. The government also orders state banks to give loans to these failing firms, knowing full well that the latter will never be able to repay the loans. As a result, the budget constraints of the state-owned enterprises are soft. As the managers of state enterprises are aware of the soft budget constraints, they consistently bargain for more resources and low output goals, and they consistently conceal and hoard resources. The situation is further exacerbated by the inability of the government to run effectively the numerous enterprises it owns due to a lack of information. In summary, socialist economic institutions have failed to provide incentives for economic efficiency.
Thus the increasing returns from institutional change in such economies are not hard to perceive, which is the fundamental reason for the economic transition now taking place in more than twenty former socialist countries. China’s transition is particularly interesting; it started as a process of decentralization and is now evolving into a process of rapid privatization.
China’s Economic Transition
In December 1978, two years after the death of Mao Zedong, the Communist Party held the historic Third Plenum of the Eleventh Party Congress, which marked the beginning of far-reaching economic reforms. This process evolved into what is now called the economic transition from a planned to a market economy. The ultimate goal of the transition is to move away from central panning and to build a market economy capable of delivering long-term growth and improved living standards.
Essentially, reform has been a two-pronged decentralization process: decentralization from the central government to the local governments, and decentralization from those governments to the basic economic units—that is, enterprises and households. 6 Decentralization from the central to the local governments gave greater incentives to local governments for revenue creation and economic development and allowed many regions to experiment with new and flexible policies, such as the special economic zones.
The transition also decentralized decisionmaking power from the government to enterprises by giving them greater control and residual claim rights. 7 These include the authority to produce, price, and sell extra products after the quota has been met, the authority to hire and fire within state guidelines, and the authority to determine rewards and punishments based on performance.
Associated with decentralization, a second important development has been the diminishing role of central planning and the emergence of markets. Since the early 1980s, China has gradually freed the pricing of goods. Initially, prices of those goods that were deemed vital to the economy and people’s livelihood—for example, prices of grain and steel—were controlled. Later, prices of most goods were determined by market demand and supply. Needless to say, the transition is far from complete. On the one hand, there is evidence of the emergence of free markets, competition, and the hardening of budget constraints. On the other hand, institutional factors that are legacies of the planning system still exist, such as high entry barriers erected by the government for some industries, interference in market activities, the imposition of restrictions or the granting of privileges to certain categories of firms, and policies that continue to encourage distorted firm behavior.
The two most important institutional arrangements that affect firm performance in the transition economy are firm ownership and government relationships.
Firm Ownership Structures
Ownership has been recognized as one of the most important institutional arrangements in determining incentive structures, which in turn determine firm performance. 8 There are three types of Chinese enterprises based on the mode through which transactions are coordinated and property rights are embodied. 9 They are state-owned enterprises (SOEs), collectively owned enterprises (COEs), and privately owned enterprises (POEs). 10 Despite the decentralization effort, to a great extent SOEs remain an administrative, not a business, system, with operations planned and controlled by higher-level agencies that provide the necessary inputs and social functions. They still operate under soft rather than hard budget constraints. Consequently, the performance of SOEs continues to deteriorate, but the government cannot simply shut down these poorly performing SOEs because they provide employment and social services on behalf of the government.
COEs carry the ambiguous definition of property rights in socialist economies to the extreme: COEs are the most unclear, fuzzily defined forms of property rights. And yet they have undergone the most dramatic transformation in China’s transition economy. COEs first appeared in the first decade of China’s socialist economy (the 1950s) as a second-class, less attractive alternative to shoulder the burden of SOEs by employing millions of people whom the SOEs simply could not absorb. By the definition of the time, COEs were “collectively owned by the people,” a vague concept without any clear indication of the owner. COEs did not enjoy as many privileges as SOEs in terms of receiving subsidies and scarce resources from the state, but they were subject to all state controls and regulations. Generally, they were poor performers.
The reform granted greater freedoms to COEs, partly because the state could no longer afford to take care of them. Most were “passed down” to be run by local governments. At the same time, fiscal reform gave local governments strong incentives to see that the COEs in their jurisdictions performed well. This newly acquired institutional advantage, namely, operational freedom and local protection, enabled COEs to outperform SOEs and even to some extent POEs. 11
Unlike SOEs and COEs, private firms follow hard budget constraints, and their survival and operational decisions depend on market performance. Private firms require much stronger entrepreneurial incentives, exacting cost-benefit calculations and greater risk taking and innovation than SOEs. 12 However, private enterprises in a transition economy such as that of China are substantially constrained by the lack of clearly defined property rights.
Government Relationship
All enterprises are directly subordinate to (lishu) one of the five levels of government—central, provincial, city, county, or township. The level of government to which an enterprise is subordinate depends on the nature of its operations, location, and scale. SOEs tend to belong to higher levels of government. As one moves down the government hierarchy from state to village, 13 the proportion of SOEs decreases from nearly 100 percent to zero; most public firms at the bottom of the hierarchy are COEs.
Government officials tend to have stronger interests in the public firms as one moves down the hierarchy. Township and village officials are functionally equivalent to a board of directors or shareholders of public firms in the West. Their personal incomes and welfare depend on the performance of the enterprises under their jurisdiction. Decentralization has given strong financial incentives to lower levels of the government hierarchy to ensure that their firms perform well. Thus, in consideration of their own economic benefits, local governments do not tolerate poorly performing enterprises. On the other hand, as one moves higher in the government hierarchy, the fiscal revenue of the government depends less on the revenue from the firms in that jurisdiction and more on taxes. Thus the incentives for higher-level governments to see that their firms perform well remain weak.
Privatization in China 14
It is generally agreed that institutional change is the driving force behind China’s impressive economic performance. Yet little is known about institutional change or the major forces behind such change. Particularly interesting is why China’s reforms, which started as a decentralization process, seemingly unintentionally have led to a transition from public to private ownership.
As noted earlier, North’s study of institutions and institutional change provides an interpretation of this issue. 15 Montinola, Qian, and Weingast’s study of “market-preserving federalism” sheds more light on this issue with particular reference to China. 16 They argued that cross-regional competition has played a central role in the rise of China’s economy over the last two decades. However, how cross-regional competition triggers privatization has yet to be addressed formally.
A Theory of Endogenous Institutional Change
This chapter presents a theory of endogenous institutional change in the context of transition economies. 17 The theory uses a mathematical model whereby two firms play a Bertrand-Nash price game in a product market. 18 It demonstrates how decentralization and cross-regional competition force local governments to privatize the SOEs and COEs under their control. The game uses a backward theorizing process, starting from market competition activities. At the market level, in order to increase their market share, firms compete on price. At the firm level, price competition leads to efforts on the part of managers to lower costs, and the results of these efforts evolve into institutional competition to determine which institutional arrangements provide the greatest incentives for managers to lower costs.
To facilitate this discussion, the following terms must be defined. “Ownership of the firm” is defined by residual claims. 19 “Privatization” is a process of shifting residual claims from the government to enterprise managers. 20
In this game, there are two local governments and two enterprises. The two enterprises were formerly owned by the central government. At the initial stage of the reform, the central government “passed down” the enterprises to the local government, with each local government owning one enterprise. The local governments then obtained the residual claims on after-tax profits and also the right to decide whether or not to shift residual claims to the managers. In other words, the local government now has the autonomy to decide whether or not to privatize the firms. To simplify the analysis, we assume that the manager has complete control of the firm’s business, except the rights of taxation and privatization, and that the manager’s residual claim rights are well preserved. Thus, when the manager holds all residual outputs of the firm, he becomes the de facto owner of the firm.
At the initial stage of the game, the central government delegates ownership to the local governments. The former still maintains authority to set the tax rate and the latter’s share of tax revenue. These parameters are the same for the two localities. At the second stage, the local governments determine the managers’ after-tax profit retention rate in an institutional competition, that is, whether or not to privatize to maximize revenue. At the third stage, given the tax rates and the managers’ incentives, the two managers choose unobservable efforts in a cost-reduction competition. At stage four, the two firms compete with each other in product markets by choosing prices.
The second through the fourth stages assume that the central government’s initial decision to decentralize and the tax rates are made independent of the local governments (or exogenously determined). Factors affecting competition include transportation costs, trade barriers, regulation, and other costs. Some of these factors are under the control of the central government, and others are under the control of local governments or are determined by technology. To emphasize how competition triggers privatization, we assume these factors to be exogenously given. It is reasonable to assume that total costs are a function of managerial effort. In a competitive market, in order for firms to survive, they must cut costs as much as possible.
The residual share of the manager gives him an incentive to reduce costs in order to increase market share. Thus the manager’s effort increases with his residual share; this means that a firm’s market share increases as the manager’s residual share increases and decreases as its rival’s residual share decreases. This is the fundamental reason why two local governments compete for privatization.
The local government’s revenue consists of two parts: the tax revenue and the profit revenue. The trade-off facing the local government is that a higher share of the residual claim for the manager (namely, privatization) generates a higher profit, but reduces its own (relative) share. The first effect (larger absolute size of revenue due to higher total profits) may be labeled the “incentive effect,” and the second effect (smaller relative size due to a higher share to the manager as a result of privatization), the “distribution effect.”
It is reasonable to assume that the local government cares only for its total revenue, not its share relative to that of the firm. 21 The following propositions are based on the analytical results of the game.
If competition is sufficiently intense, then the local government is motivated to privatize its firm, since the incentive effect dominates the distribution effect. Therefore the local government prefers to obtain the tax revenue since the revenue increase resulting from a bigger profit share to the government is not large enough to offset the tax revenue loss due to weaker managerial incentives.
This implies that a larger share of tax revenue to the local government (as opposed to the share of the central government) promotes a larger private sector. The greater the share of tax revenue to the local government (given the tax rate), the more likely the incentive effect (due to privatization) dominates the distribution effect.
In order for competition to work effectively, regional governments cannot erect trade barriers, and yet they must have autonomy to make other economic decisions to respond to competitive pressure. So far, it has been assumed that regional governments have full autonomy to set after-tax residual shares. When the autonomy of a regional government is restricted by the central government, the privatization process may slow down. Indeed, in the game, if all the rights to set after-tax residual shares are in the hands of the central government, public ownership will prevail; under complete control of the central government, there is simply no need for local governments to motivate their firms to perform.
On the other hand, if two regions perfectly collude, initially there will be no cross-regional competition; each local government will claim one-half of the total profits earned by the two regions and neither government will have any incentives to give any of the residual profits to the managers, and thus the managers will make no effort to reduce costs. However, in general such collusion is not an equilibrium. If one government sets the residual profit of its manager at zero, the other government, in order to beat its rival, will try to give a bit more incentive to its manager. As a result, each local government is tempted to go a little further than its rival toward privatization until equilibrium is restored at full privatization. In other words, when competition in the product market is sufficiently intense, two local governments will compete in terms of the privatization of their firms. They are in a “prisoner’s dilemma.”
In summary, in a game of price competition by firms owned by local governments, production costs are determined by the managers’ efforts. The local government is concerned with its own total revenue, which depends on its market share and profit rate. When competition is sufficiently intense in the product market, the local government will be motivated to shift the residual claim to its manager. As the product market becomes more competitive, the market share and therefore profits are more sensitive to production costs. In order to maintain a minimum market share for survival, the manager must be motivated to work harder to reduce costs. Given that the government cannot directly monitor each manager’s efforts, privatization is the only effective means by which the local government can motivate its manager. In contrast, if the central government sets the after-tax residual share, or if the two local governments perfectly collude to maximize their joint revenue, then public ownership may prevail. The game shows that efficiency improves as a consequence of privatization.
Empirical Evidence and Test of the Theory
Empirical evidence from the past two decades supports this theory. In the past two decades and particularly since the early 1990s, both explicit and implicit privatization have been taking place at an increasing pace. 22 In 1978, at the beginning of the reform, 78 percent of the total industrial output came from SOEs. By 1995, the SOEs’ share had shrunk to only one-third of the total output. 23 Note that these statistics only account for explicit, not implicit, privatization.
The major players behind the privatization process are local governments at various levels. 24 Although not all local governments are undertaking explicit, wholesale privatization programs, almost all local governments are considering privatization in one way or another. 25 For example, a recent survey estimates that more than 70 percent of small SOEs have been fully or partially privatized in Shandong Province. 26 In November 1997, the Beijing municipal government began to auction off many small SOEs. In early 1998, Dalian city government began to auction off not only the failing SOEs, but also the profitable ones. The aim is to “cut off all SOEs from the government.” 27
This theory was tested using newly available industrial census data from 1993 to 1995. These censuses are conducted by the State Statistical Bureau and cover all manufacturing firms subordinate to the township government (the lowest-level governmental body in China) or above. The censuses include between 400,000 to 500,000 firms. In 1995, the output of these firms accounted for 94 percent of the total industrial sales by all firms with independent accounting systems. 28 Firms not included in the census are the very small, often family-run workshops. The data set contains ownership type, level of government control, geographic location, revenue, and other performance and demographic variables. As mentioned earlier, virtually all firms are subordinate to (lishu) governments at different levels (central, provincial, city or prefecture, county, and township). The firms subordinate to a government may be state-, collective-, or privately owned. In general, SOEs are subordinate to county-level governments or above, COEs are subordinate either to county-level governments or above or to township governments and village committees, and private enterprises are primarily subordinate to village committees, or township or county governments.
In this theory, a region’s privatization level is determined by the following factors.
Degree of control over a region’s own public enterprises (SOEs and COEs). A local government can only privatize the public firms under its control. It has no control over firms located in its jurisdiction but reporting directly to a higher level of government (in fact it may not even have any incentive to do so). Under the pressure of cross-regional competition, the higher the degree of control over its public firms, the more the likelihood that the local government will privatize them.
The existing level (history) of privatization in a region. A higher degree of privatization in a region implies greater intraregional competition and greater freedom to privatize, thus the greater the likelihood of further privatization.
The theory predicts that all of these factors contribute to privatization by facilitating competition. The proportion of revenue contributed by the private sector to a county in 1995 measures the privatization level. The degree of access to transportation is measured by two variables: one is the ratio of coastline length to land area; and the other is the ratio of railway length to land area; both are at the provincial level. The degree of privatization in the surrounding regions is measured by the proportion of revenue contributed by the private sector to the neighboring counties in 1993. Neighboring counties include all counties in the same prefecture as the county being evaluated (excluding the county being evaluated). The level of control over a region’s public firms is measured by the proportion of a county’s revenue being contributed by SOEs subordinate to governments at the county level or lower in 1994. The privatization history of a region is the proportion of revenue being contributed by the private sector to the county under evaluation in 1993. The unit of analysis is the county. Information has been aggregated from all firms extant in 1993 (446,265), 1994 (485,052), and 1995 (450,223) at the county level, and the indexes of privatization and level of control for all 2,002 counties has been calculated based on the industrial census. A multiple regression analysis is used to evaluate the hypothesized causal relationship. If the theory is true, then all the factors (1) to (3) will simultaneously exert positive effects on a region’s privatization level. Given the comprehensive coverage of our data, the test is rigorous.
The result of the regression analysis shows that factors (1) to (3), as predicted, positively affect a region’s privatization in a highly significant way. For example, the regression model shows that a 10 percent higher private share in the neighboring counties triggers a 1.5 percent higher private share in the county under consideration two years later. The statistics show that the chance that the model is wrong is negligible: 0.01 percent. These results provide convincing support for the theory.
The discussion and the test in this section show that the theoretical predictions are consistent with what has actually occurred. The economies of the coastal regions are more privatized than those of the inland regions because the former enjoy not only lower transportation costs (thus facilitating cross-regional competition) but, more importantly, greater autonomy. Similarly, SOEs in the northeast and southwest tend to be less privatized since their high concentration in these regions makes competition less intense. Sectors with simple or standard contracts are more privatized than sectors with complex or specific contracts because the former involve lower enforcement costs and hence face stronger competition than the latter. The former include labor-intensive industries, such as textiles and consumer electronics. The latter generally include capital-intensive and contract-intensive industries, such as machine tools, banking, and insurance. For example, in 1985 the output of SOEs accounted for 17 percent and 64 percent of the total output of the garment and machine tools industries, respectively. By 1995, the SOEs’ share of the garment industry had shrunk to 6.9 percent, while the SOEs’ share in the machine tools industries was still 40 percent. 29 Privatization of town and village enterprises proceeds more rapidly than that of SOEs because TVEs operate in more competitive markets and their governments (townships) have no leverage to shield them from outside competition.
Institutional Versus Economic Effects 30
The preceding section attempts to provide an understanding of change in China’s transition and the consequences of such institutional change. This section examines the question of how institutional factors affect firm performance. The factors that determine firm performance such as profitability are a central problem both for industrial organizational economics and strategic management research. Scholars in both fields, for example, are concerned with understanding what factors are responsible for persistent unequal returns across industries and firms. 31
Both fields recognize that there are two groups of factors that explain variations in firm performance: industry-specific factors and firm-specific characteristics. Industry-specific factors include the concentration level of an industry (consolidated versus fragmented); minimum capital requirements (which is a form of entry barrier, e.g., the aircraft industry requires a much higher minimum investment than the garment industry); or product differentiation (also an entry barrier, e.g., if the existing products are highly distinctive, it will be difficult for new producers to attract customers who remain loyal to the existing products). Firm-specific characteristics include firm resources (capital, technology, and human resources), capabilities (marketing, management, or production skills), and strategies.
Industrial organization economists and strategic management scholars differ regarding the roles each set of factors plays in firm performance. Industrial organization economists argue that industry structures, such as concentration and entry barriers, lead to variation in firm performance. They focus on interindustry differences as the main source of varying performance. Firm-specific characteristics are regarded as less important in determining firm performance. Scholars of strategic management, on the other hand, emphasize the choice orientation of a firm’s behavior and argue that firm characteristics, such as firm resources, capabilities, and strategies, are essential in determining firm performance. They point out that even in weak industries, there are stars that consistently outperform the others, and in high-profit industries there are failures due to mistakes on the part of firms.
In order to evaluate quantitatively the role of industry- and firm-specific factors on performance variations, scholars in both fields have conducted performance variance decomposition analyses. They find that industry factors explain a relatively small proportion of the variations, 13–19 percent; a much larger proportion of profitability variation, 36–47 percent, is attributed to firm-specific characteristics. The remainder is due to estimation errors. 32
The above studies of firm performance are all based on Western experiences. To what extent are these studies applicable to firm performance in a transition economy? How much do industry and other economic factors explain performance variance across firms in China? In a transition economy such as China’s, are there any determining factors unique to the transition from a planned economy to a market economy?
Studies based on Western economies concentrate on economic factors that determine firm performance. However, in a transition economy such as China’s, institutions do matter in terms of economic performance. The institutional legacy of planning hinders an efficient market, and so there are increasing returns to institutions (which is the motivation behind institutional change). Thus, firm performance is jointly determined by both the newly emerging market forces (economic factors) and by the legacy of economic planning (or government interventions—institutional factors).
Decomposing Firm Performance Variance in China
Guided by theoretical arguments, a firm performance variance decomposition analysis can be done. Firm performance is measured by return on assets. In addition to industry and firm factors, two institutional factors were considered: ownership and government relationship, the latter of which should be regarded as the most important (see the above detailed discussion on these two factors). The time period of our analysis was from 1992 to 1995, because China was changing very rapidly during that period. With the time factor (one year) in the analysis, it was possible to evaluate the dynamic effects of institutions, industry, and firms. The results of the analysis show that the dynamic institutional effect accounts for 33 percent of firm performance variance. The dynamic industry effect accounts for only 6 percent, and the dynamic firm-specific factor accounts for 24 percent; the remainder is the variance unexplained by these factors. As expected, performance variance across firms in China’s transition is different from that in the West. Institutional factors are most important in explaining variance in firm performance.
Estimating the Effects of Government Interventions and Economic Factors on Firm Performance
The performance variance decomposition analysis demonstrates that institutions matter for firm performance. But it does not reveal the theoretical relationship between each determining factor and firm performance. For example, it does not show whether or not subordination to a lower-level government helps firm performance. Based on the argument that both government intervention and market forces jointly determine firm behavior and performance, and on the general relationship between institutions and firm performance, the theoretical relationships between these factors and firm performance can be developed.
Once the theoretical relationships were developed, an empirical test using China’s 1995 industrial census data was conducted. The testing method was multiple regression. From the 1995 census of 450,233 firms, a 10 percent random sample was used to test the hypotheses.
In the empirical test, performance is measured by two variables: sales revenue per employee (a measure of productivity), and returns on assets (a measure of profitability). Using these two performance measures as dependent variables, two multiple regression models were estimated independently. 33
Government Influences on Firm Performance
The dukedom effect: local protectionism. As part of the economic reform, the central government has delegated much responsibility and authority to lower-level governments. This process provides strong incentives to local governments to help the firms under their jurisdiction.
How does a local government help the firms under its jurisdiction? One obvious measure is local protectionism, or the so-called dukedom economy as it is referred to in China. A good example of the dukedom economy is the automobile industry. By its nature, the automobile industry is highly consolidated because economies of scale are vital for success. Due to high tariffs and other restrictions on imports as well as high demand for automobiles, automobiles are very expensive and highly profitable. Many provinces claim the automobile industry to be a “pillar industry” with a high priority for protection. These provinces set up their own automobile plants and ban the sales and use of automobiles “imported” from other provinces. At present, there are over two hundred assembly plants, some with an annual production as low as a hundred units. One analyst has referred to China’s automobile industry as “the most fragmented in the world.” 34 Other examples of local protectionism are the tobacco and alcoholic beverage industries.
This local protectionism causes the industry to fragment geographically and the geographic barriers keep profits artificially high. The regression model shows that the lower an industry’s geographic concentration, the higher the firm’s profit. 35
Effects due to local support. In China, governments have three sources of revenue: taxes, fees, and the profits of the public firms under their jurisdiction. In general, the higher the level of the government, the greater reliance on taxes for revenue and the less reliance on profits from its firms. The central government relies primarily on taxes for revenue. The central government also receives some profits from the firms it oversees, although these are insignificant compared to the taxes. The main tax sources for local governments are various firm taxes. In addition, local governments rely on the profits of the public firms under their jurisdiction. As one proceeds lower down the government hierarchy, the contribution of firms—in both taxes and profits—becomes more important. As a result, local governments are motivated to ensure that firms under their jurisdiction perform well. In general, the lower the level of government, the more dependent it is on, and the greater its economic interest in, the performance of the firms in its jurisdiction.
Because of their strong financial interest, local governments give preferential treatment to local firms rather than to “nonlocal” firms. Regression analysis finds that because of such local support, firms reporting to the lowest level of government tend to perform best, both in terms of profits and in terms of productivity.
Contribution to the local economy. To ensure revenue, local governments may establish various operation-related fees for firms that operate in the community. These fees are normally set in proportion to the size of the operation. In other words, the more revenue a firm generates, the more the firm will be charged. These charges are key sources of income for many local governments.
In addition, firms contribute to the local community in other ways. They provide jobs for the community and, in many cases, they run social services such as hospitals, schools, and even bus lines—not only for their workers but also for the community at large. (Some local firms even support the daily operation of the local courts!) Given their importance to the local economy, large firms (in proportion to other firms in the same locality) can bargain and receive preferential treatment from the local government. The higher the revenue of a firm in comparison to average firm revenue in the local economy, the better it performs both in terms of profit and of productivity.
Soft budget constraints. Perhaps the best-known government support to firms in planned economies is the “soft budget constraint.” This includes loans at reduced interest rates and flexible terms, reduced and ex ante negotiated taxes, and other subsidies.
The practice of soft budget constraints is so common and detrimental in transition economies that many researchers argue that only by reducing such subsidies (a “hardening of the soft budget constraints”) will firm performance in transition economies be improved. 36 The rationale is quite simple: if a firm can spend what does not belong to it and is not responsible for its budget constraints, then the firm will not perform well. However, measuring the effects of soft budget constraints on firm performance is not easy, 37 because most subsidies are hidden.
The effect of soft subsidies in taxes on firm performance can be measured by using the variation in county tax rates. Tax rates vary greatly, not only across counties but also across firms within counties, because to a great extent county governments can control tax rates and set different tax rates for different firms; this is a typical form of soft subsidies from governments to firms.
A greater variation in tax rates implies a greater degree of soft budget constraints imposed on firms in a county. This in turn stimulates firms to seek greater tax subsidies, thus providing them with fewer incentives to perform in the market. Rather, they spend their resources on (or under) the negotiation table with government officials. In addition, firms are motivated to reduce their tax exposures by reporting lower performance. As a result, the higher the variation in the tax rate at the county level, the lower a firm’s performance (profitability and productivity).
The influence of market forces on firm performance. Based on past studies of firm performance in market economies, the most important factors affecting firm performance are industry concentration, the size of capital investment, product differentiation, and industry risks. These studies have shown that profitability is positively correlated with the above factors. 38
This study examines the effects of industry concentration, size of capital investment, and industry risk on firm performance. (Due to the lack of relevant data, such as advertising intensity, the effect of product differentiation on firm performance cannot be assessed.) Since the effects of these variables on firm performance are well established in industrial economics literature, these will also apply to China after controlling for the effects of government intervention on firm performance.
Industry concentration. Higher market concentration in an industry reduces competition and encourages collusion in that industry. As a result, firms will enjoy higher profits. On the other hand, reduced competition causes firms to be complacent and they will be less productive. In the regression model, the higher the market concentration in the industry level, the higher the profitability and the lower the productivity of firms in that industry.
Size of capital investment. The scale of capital investment is often seen as a barrier to potential entrants: the higher the capital requirements, the higher the entry barriers. Such barriers in turn ensure that existing firms earn higher profits. In addition, greater investment generally implies large-scale production, which usually correlates with high productivity. As expected, the higher the capital requirement in an industry, the greater the performance of firms in that industry.
In summary, analysis confirms the hypotheses about the effects of government intervention in shaping firm behavior and performance. After two decades of economic reform, the legacy of the planned economy (that is, government intervention) is still felt when firm performance is examined. At the same time, however, market forces also exert an effect on firm performance.
Discussion and Conclusions
This chapter has reviewed a theory of endogenous institutional change and a theory of the determinants of firm performance in the context of transition economies in general and of China in particular. Both theories are strongly supported by empirical tests based on the most comprehensive data sets available: industrial censuses from China.
On Institutional Change and Privatization
Privatization of public enterprises can take place as a consequence of cross-regional competition. Chinese economic reform did not initially intend to privatize public enterprises. But the decentralization policy eventually triggered privatization through cross-regional competition. The recent Fifteenth Party Congress’s promotion of joint-stock systems to bail out the vast majority of failing SOEs is widely viewed as implicit privatization. But explicit privatization has thus far not been adopted as central government policy, mostly for ideological reasons. However, competition is far more powerful than ideology. Regardless of whether or not the central government draws up a blueprint for full privatization, both theory and reality reveal that the privatization process will continue and accelerate with its own logic and vigor.
The Chinese experience demonstrates that the “invisible hand” is powerful not only in allocating resources, but also in creating institutions. Once there is decentralization, market competition may precipitate a self-enforcing process of privatization. As a result, decentralization may induce even more privatization than a deliberate privatization policy if the latter is not accompanied by sufficient decentralization. This is a major lesson that other transition and emerging economies may draw from China’s experience.
Nevertheless, privatization calls for a sound legal system to protect property rights. In particular, de facto ownership by managers must eventually become de jure ownership. Commercial laws are needed to support contracts between privatized enterprises. Two major problems stand out in the current system of commercial law. First, there are no clear and detailed rules to protect private property. To facilitate efficient private investments, detailed civil codes are needed to protect private property under different contingencies. Second, cross-regional commercial disputes are presently settled in local courts controlled by local governments. To mitigate local protectionism and to facilitate competition, either local courts need to become independent from local government control or major cross-regional commercial disputes should be dealt with in the national courts.
On Determinants of Firm Performance
Analysis in decomposing variance in firm performance shows that in a transition economy such as China’s, in addition to economic factors such as industry and firm factors found in market economies, institutional factors measured by ownership structure and government relationship also affect firm performance. Furthermore, these institutional factors exert a larger effect than either the industry effect or the firm effect.
The determinants of firm performance in a transition economy consist of both market factors and government intervention. This chapter establishes the theoretical relationship between industry structure and firm performance, and it identifies and measures the effect of government interventions, especially that of the soft budget constraints. This relationship has been duly tested with empirical data.
Three major findings are worthy of further elaboration. First, in a transition economy such as China’s, both government interventions and market forces are powerful in shaping firm behavior. This is important for future theoretical frameworks on transition economies.
The second important finding concerns the complexity of government forces in a transition economy. Local government is very effective in determining firm performance. However, “local governments as industrial firms” 39 is a double-edged sword and ultimately may be a major obstacle to further transition. At best, it is merely an expedient allowing these public firms to operate better than the “iron rice bowl” system of the old planned economy. Theories of political economy have convincingly demonstrated that government should be the “referee” of the economic game in a society, not a player in the game. When government is both the referee and a player, the rules of the game will be unfair and corruption will inevitably occur, in the long run hindering economic growth and firm performance. 40 Fortunately this sentiment is increasingly accepted by reformers in China. At the recent Fifteenth Party Congress, the government explicitly recognized this problem and stressed its determination to solve it. 41 Shunde, a region in China’s most affluent province of Guangdong, demonstrates such a shift in government roles. Shunde is known for its investment environment and market institutions. Its evolution from “government as corporation” to “government as government” can be clearly seen from the governing philosophy of its mayor, Feng Runsheng, whose observations follow:
I used to be a chief representative of all factories in Shunde. You may call me the biggest CEO of all city enterprises . . . Now, the government and enterprises are separate. As a mayor, I have only two responsibilities: to develop the city’s infrastructure and services, and to maintain law and order . . . All investors—domestic and foreign—can easily learn and understand our policies and regulations. We welcome everyone to invest, but we will not break the rules for anyone. 42
The third finding is revealing for the study of institutional change in transition economies. Earlier studies have shown that firm ownership (such as state-owned, privately owned, or foreign-owned enterprises) is very important in determining performance in China’s transition. 43 But this study shows that the role of ownership is not very important in and of itself in determining performance, after factoring in government interventions. In view of the heated discussions in all transition economies in general and in China in particular about the role of privatization, this finding is likely to stimulate further debate on the issue. This study suggests that in China, where clearly demarcated property rights are lacking and explicit privatization still faces strong public opposition, the firm—together with the local government—has found a convoluted way to stimulate performance: giving more incentives to the decisionmaker of the firm by making the local government the de facto owner.
Managerial and Policy Implications
For managers operating in or considering entering China, the salience of government interventions on firm performance implies that developing and maintaining a good relationship with the local government (not just the provincial or national government) is of vital importance.
For public policymakers, the relationship between the local government and the firm needs to be restructured so that the former is the referee and the latter the player in the economic game. The government needs to reduce its intervention and to allow the market mechanism to operate freely in order to boost performance. Unless new evidence proves otherwise, thus far both theoretical as well as empirical studies have shown that even with good intentions, most government intervention hinders rather than helps firm performance.
From “isms” to “ownership”: Solution to the SOE problem. In the early days of reform, there were heated debates as to whether the reform would lead to capitalism or socialism (xingshe xingzi). The conservatives claimed that the introduction of free-market mechanisms represented a “march toward capitalism.” Deng Xiaoping declared that it was not important to discuss the “isms”; rather, China needed to focus on whatever route would raise living standards. Deng’s political astuteness effectively contained the ideological debate and led to the further development of free markets.
Now the reform has reached a crucial turning point whereby the partial introduction of free-market mechanisms is insufficient; at the foundation of free markets must be a system of clearly defined and well-protected private property rights. However, under Deng, privatization was not possible. 44 The end of the Deng era allowed an opportunity for the restructuring of the property rights system. The current leaders are fully aware of this. 45 The debate today is over “public versus private ownership” (xinggong xingsi). The evolution of the debate from “isms” to “ownership” shows that the transition is at a crossroads. The stakes are high: SOEs account for more than one-half of China’s total industrial assets, but only produce one-third of its output. 46 Most of the inefficient SOEs are losing money because of ill-defined property rights, mounting debts, and resultant social burdens. The only salvation appears to be privatization.
Endnotes
Note 1: China National Statistics Bureau, Zhongguo tongji nianjian, 1996 [China Statistical Yearbook, 1996] (Beijing: Zhonggou tongji chubanshe), p. 42. Back.
Note 2: World Bank, Bureaucrats in Business (Washington, DC: World Bank, 1997). Back.
Note 3: Douglass C. North, Institutions, Institutional Change and Economic Performance (Cambridge: Cambridge University Press, 1990). Back.
Note 4: North, Institutions, Institutional Change and Economic Performance, p. 95. Back.
Note 5: Janos Kornai, Economics of Shortage (Amsterdam: North-Holland, 1980). Back.
Note 6: Gabriella Montinola, Yingyi Qian, and Barry R. Weingast, “Federalism, Chinese Style: The Political Basis for Economic Success in China,” World Politics 48 (October 1995): 50–81. Back.
Note 7: Residual claim rights are the rights to claim residual output beyond the agreed-upon quantity. Economists have shown that in order to design an efficient (output maximization) incentive scheme, it is necessary to ensure that the production decisionmaker is the residual claimant to output. See Hal R. Varian, Intermediate Microeconomics (New York: W. W. Norton, 1996), p. 644. Back.
Note 8: North, Institutions, Institutional Change and Economic Performance; Shaomin Li, “Success in China’s Industrial Market: An Institutional and Environmental Approach,” Journal of International Marketing 1 (1998): 56–80. Back.
Note 9: Max Boisot and John Child, “From Fiefs to Clans and Network Capitalism: Explaining China’s Emerging Economic Order,” Administrative Science Quarterly 41 (1996): 600–628; Victor Nee, “Organizational Dynamics of Market Transition: Hybrid Forms, Property Rights and Mixed Economy in China,” Administrative Science Quarterly 37 (1992): 1–27; Andrew G. Walder, “Local Governments as Industrial Firms: An Organizational Analysis of China’s Transitional Economy,” American Journal of Sociology 101 (1995): 263–301. Back.
Note 10: Private ownership is broadly defined to include all non-SOEs and non-COEs, which are primarily market-driven and have hard budget constraints. They include privately owed companies, joint ventures, limited companies, and joint-stock companies. Back.
Note 11: See Nee, “Organizational Dynamics of Market Transition,” and Li, “Success in China’s Industrial Market.” The advantage of COEs appears to be fading and their performance potential seems to have reached its limit. The reasons for this include: (1) the fuzzy ownership structure (i.e., who is the owner?); and (2) the lack of managerial ability to cope with rapid expansion and market change. See Craig S. Smith, “Factory Towns—Municipal-Run Firms Helped Build China; Now, They’re Faltering,” Wall Street Journal, October 8, 1997, p. A1. Back.
Note 12: Nee, “Organizational Dynamics of Market Transition.” Back.
Note 13: Although a village is not a government, it is a public organization that can own COEs. Back.
Note 14: This section summarizes the study by Shaomin Li, Shuhe Li, and Weiying Zhang regarding privatization in China’s economic transition. For details, see Shaomin Li, Shuhe Li, and Weiying Zhang, “Competition and Institutional Change: Privatization in China” (Hong Kong: City University of Hong Kong, Working Paper, 1997). Back.
Note 15: North, Institutions, Institutional Change and Economic Performance. Back.
Note 16: Montinola, Qian, and Weingast, “Federalism, Chinese Style.” Back.
Note 17: By “endogenous” we mean that the institutional change is caused by economic reform. Back.
Note 18: In a Bertrand-Nash price game, the only criterion for buyers’ decisions is price; thus sellers compete solely on price. As a result of this competition, the equilibrium price will converge to the cost of the sellers. Back.
Note 19: Traditionally, residual rights have defined ownership. Grossman and Hart define ownership as control rights over assets (see Stanford Grossman and Olive D. Hart, “Cost and Benefits of Ownership: A Theory of Vertical and Lateral Integration,” Journal of Political Economy 94 [1986]: 691–719). Presently, economists recognize that both residual claims and control rights are indispensable to ownership. Here we omit control rights, not because they are irrelevant but because of technical intractability. Nevertheless, we suppose that our results apply to control rights as well. Back.
Note 20: Since our model is timeless, “the residual” is best interpreted as the present value of all future residual flows when one applies the theory to reality. That is, privatization should be understood as a permanent transfer of residual claims from government to private hands. A short-term contract between government and management, such as the “contract management responsibility system” practiced in China, can be seen as a partial privatization of state-owned enterprises. Back.
Note 21: One may think that the local government also cares for its market share per se since market share is positively related to local employment levels and often to chances for promotion of local officials in the governmental hierarchy. However, our basic conclusion does not change, but rather is strengthened if market share itself is part of the local government’s objective function. Back.
Note 22: Implicit privatization refers to cases in which assets and profits formally belong to governments, but de facto are held by managers and bureaucrats. Typical characteristics of implicit privatization include hiding assets and profits in private accounts, making investments in so-called subsidiary companies with little control by the governments, setting up companies in foreign countries, and transferring state assets to private or quasi-private companies. Back.
Note 23: China National Statistics Bureau, China Statistical Yearbook, 1996, p. 403. Back.
Note 24: For instance, in 1992, Quanzhou municipal government of Fujian Province sold all of its SOEs to a Hong Kong businessman, who also bought dozens of other SOEs in other regions and successfully packed them for public offering in the Hong Kong and New York financial markets. From 1992 to 1994, Zhucheng municipal government of Shandong Province privatized all of its 272 government-owned enterprises. A similar process of privatization has been undertaken in many other areas such as in Shunde of Guangdong Province, Yibin and Deyang of Sichuan Province, Shuzhou of Shanxi Province, and Xishui and Xianfan of Hubei Province. See Hu Shuli, “Zhongce xianxiang: guanyu yinzi gaizao de jiexi he sikao” [“The Phenomenon of ‘China Strategy’: Analysis and Thinking of Restructuring State Enterprises Through Foreign Investors”], Gaige (Reform Journal) 3 (1994): 74–85; and see Wang Yanzhong and Xu Heping, Zhucheng qiye gaige tantao [Studies of Zhucheng’s Enterprise Reform] (Beijing: Jingji guanli chubanshe, 1996). Back.
Note 25: An anecdote among economists is indicative. When an economist told a provincial governor, “I heard that the central government has created regulations forbidding local governments from selling SOEs,” the governor responded, “Do you know anyone who wants to buy?” Back.
Note 26: China Reform Foundation, Xianshi de xuanze [Reality’s Choice] (Shanghai: Shanghai yuandong Chubanshe, 1997), p. 35. Back.
Note 27: “Jingji Banxiaoshi” [“30-Minute Economy”), Zhongyang Dianshitai 2 (China Central Television 2), April 1–2, 1998. Back.
Note 28: China National Statistics Bureau, Zhongguo gongye renkou tongji, 1995 [Chinese Industrial Census, 1995] (Beijing: Zhonggou tongji ju, 1995), and Zhongguo tongji nianjian, 1996 [China Statistical Yearbook, 1996] (Beijing: Zhonggou tongji ju, 1996), p. 417. Back.
Note 29: China National Statistics Bureau, Zhongguo tongji nianjian, 1986 [China Statistical Yearbook 1986] (Beijing: Zhonggou tongji chubanshe), pp. 242–243; China Statistical Yearbook 1996, pp. 414, 418. Back.
Note 30: This section summaries findings from the following studies: Shaomin Li and David K. Tse, “Do Market Forces Matter in a Transitional Economy? Effects of Economic Factors and Government Intervention on Firm Performance in China” (Hong Kong: City University of Hong Kong, Working Paper, 1997); Shaomin Li, “Success in China’s Industrial Market”; and Seung Ho Park, Shaomin Li, and David K. Tse, “Determinants of Firm Performance in a Transition Economy: Institutional vs. Economic Effects in China” (Paper presented at the 1997 Meeting of the Academy of International Business, Monterrey, Mexico, 1997). Back.
Note 31: Richard Schmalensee, “Do Markets Differ Much?” American Economic Review 75 (1985): 349–365; Richard P. Rumelt, “How Much Does Industry Matter?” Strategic Management Journal 12 (1991): 167–185. Back.
Note 32: Schmalensee, “Do Markets Differ Much?”; Gary S. Hansen and Biger Wernerfelf, “Determinants of Firm Performance: The Relative Importance of Economic and Organizational Factors,” Strategic Management Journal 10 (1989): 399–411; Rumelt, “How Much Does Industry Matter?”; Thomas C. Powell, “How Much Does Industry Matter? An Alternative Empirical Test,” Strategic Management Journal 17 (1996): 653–664; and Anita M. McGahan and Michael E. Porter, “How Much Does Industry Matter, Really?” Strategic Management Journal 18 (1997): 15–30. Back.
Note 33: The profitability regression model is based on the Cobb-Douglas production function. The profitability regression model is estimated by using a linear model. Back.
Note 34: Michael J. Dunne, “The Race Is On,” Chinese Business Review (March–April, 1994): 16-23. Back.
Note 35: A Herfindahl concentration index is used, which is the sum of the ratio of a location’s market share squared. Donald A. Hay and Derek J. Morris, Industrial Economics: Theory and Evidence (Oxford: Oxford University Press, 1986), p. 104. Back.
Note 36: Justin Y. Lin, Fang Cai, and Zhou Li, “Qiye gaige de hexin shi chuangzao gongping jingzheng de huanjing” [“Creating an Environment for Fair Competition Is the Core of Enterprise Reform”], in Zhongguo guoyou qiye gaige [Reform of State Owned Enterprises in China], ed. Xu Dianqing and Wen Guanzhong (Beijing: Zhongguo jingji chubanshe, 1996), pp 49–89; Orjan Sjoberg and Zhang Gang, “Soft Budget Constraints in Chinese Township Enterprises” (Stockholm: Stockholm School of Economics, 1996). Back.
Note 37: Sjoberg and Zhang, “Soft Budget Constraints.” Back.
Note 38: Schmalensee, “Do Markets Matter Much?”; J. Khalilzadeh-Shirazi, “Market Structure and Price-Cost Margins in U.K. Manufacturing Industries,” Review of Economics and Statistics 56 (1974): 67–76; J. L. Bothwell and T. E. Keeler, “Profits, Market Structure and Portfolio Risk,” in Essays on Industrial Organization in Honor of Joe S. Bain, ed. R. T. Masson and P. D. Qualls (Cambridge, MA: Ballinger, 1976), pp. 71–88; Hay and Morris, Industrial Economics. Back.
Note 39: Walder, “Local Governments as Industrial Firms.” Back.
Note 40: Smith, “Factory Towns.” Back.
Note 41: Jiang Zemin, Gaoju Deng Xiaoping lilun weida qizhi, ba jianshe you Zhongguo tese shehui zhuyi shiye quanmian tuixiang ershiyi shiji [Hold High the Great Banner of Deng Xiaoping Theory for All-Round Advancement of the Cause of Building Socialism with Chinese Characteristics into the Twenty-First Century] (Beijing: Renmin chuban she, 1997), p. 25. Back.
Note 42: Feng Runsheng, “Wo zhege shizhang xianzai mang xie shenmo?” [“As a Mayor, What Am I Busy with Now?”] Juece cankao (Policy References) 28 (1997): 27–28. Back.
Note 43: For example, see Li, “Success in China’s Institutional Market.” Back.
Note 44: Deng maintained that public ownership had to remain dominant in China. See Deng Xiaoping jingji lilun xuexi gangyao [Deng Xiaoping’s Economic Theory: A Study Outline] (Beijing: Renmin chuban she, 1997), p. 36. Back.
Note 45: Jiang Zemin, Hold High the Great Banner. Back.
Note 46: China National Statistics Bureau, China Statistical Yearbook 1996, pp. 403, 409. Back.
Email this citation