Working Paper No. 95. Thriving on a Tilted Playing Field: China s Non-State Enterprises in the Reform Era by. Chong-En Bai*

CENTER FOR RESEARCH ON ECONOMIC DEVELOPMENT AND POLICY REFORM Working Paper No. 95 Thriving on a Tilted Playing Field: China s Non-State Enterprises in the Reform Era by Chong-En Bai* David D. Li** Yijiang Wang*** March 2001 Stanford University John A. and Cynthia Fry Gunn Building 366 Galvez Street Stanford, CA 94305-6015 * School of Economics and Finance, University of Hong Kong ** Department of Economics, School of Business and Management, Hong Kong University of Science and Technology and Center for Economic and Policy Research *** School of Business, University of Minnesota and Center for Economic and Policy Research

Thriving on a Tilted Playing Field: China s Non-State Enterprises in the Reform Era Chong-En Bai University of Hong Kong David D. Li Hong Kong University of Science and Technology and CEPR Yijiang Wang University of Minnesota and CEPR October 1999 Revised: March 2001 Abstract China s non-state sector, which refers to all production units not owned or not directly controlled by the state, has been the engine of the country s rapid economic growth for the past two decades. After documenting this phenomenon, the paper seeks to address three related questions. First, are non-state enterprises facing a level playing field? The answer is no. We show that non-state enterprises have been discriminated against and handicapped in competition with state enterprises. Second, given that the playing field is not level, why and how has China s non-state sector achieved its rapid growth? We show that there are sets of transitional institutions that have helped non-state enterprises to overcome constraints. Third, can transitional institutions continue to propel the growth of China s non-state sector? We do not believe so, since such transitional institutions fall short of what is needed in today s economic conditions, and also create distortions by themselves. By implication, we conclude that China s governments need to implement further reforms to level the playing field in order to sustain the vitality of the non-state sector. Keywords: Non-State Sector, TVE, Transitional Institutions, Playing Field, Property Rights, Privatization. 2

1. Introduction In the past two decades, China s non-state sector, which generally refers to all production units not directly owned or controlled by the state, achieved a growth record unparalleled by other transitional economies. With its growth far outpacing that of the state sector, the share of China s total industrial output produced by the non-state sector rose from about 35 percent in 1985 to more than 72 percent in 1999. During this period, the share of the non-state sector s contribution to total industrial growth increased from about 55 percent to 72 percent. The outstanding performance of the non-state sector is indeed a feature that distinguishes China s reform experience from those of other transitional economies. Given the share of the non-state sector in China s economy, there is little question that this sector has become a major factor in determining the overall performance of the economy. This paper reviews the experience of China s non-state enterprises in the past two decades by studying three closely related questions. First, does the outstanding performance of China s non-state sector imply a level, or even more favorable, playing field for the non-state enterprises (NSOEs) than enjoyed by the state-owned enterprises (SOEs) during the reform? We will argue that the answer to this question is no, on the grounds that financial, legal and other discriminations against the NSOEs are persistent and pervasive. We will argue that these discriminations are detrimental to the NSOEs development because they mean high entry barriers for potential entrants, poorly protected property rights for the existing NSOEs, and distorted internal governance. All these lead inevitably to losses of efficiency. Our second question is a puzzle resulting from the answer to the first question. That is, if the playing field is tilted so disadvantageously for the NSOEs, how did they manage to grow so rapidly in the past two decades? We believe that the answer to this question lies in an understanding of China s transitional institutions. One such transitional institution is fiscal and intra-government decentralization. The relative fiscal and administrative independence of local governments from the center allows the former (collectively and officials individually) to benefit from the rapid growth of the NSOEs in multiple ways, including better employment opportunities and increased (retainable) tax revenues. Another transitional institution has been the vague and ambiguous property right arrangements in many NSOEs. In the early stages of their development, many NSOEs voluntarily chose a government agency as a partner. By 3

yielding some benefit and control rights to the government agency and its bureaucrats, these NSOEs obtained better protection from the local bureaucrats and avoided many of the financial (e.g., tax and credit), administrative (e.g., red tape), and legal difficulties that less well-protected enterprises find hard to overcome. The third question has great policy significance: are the existing transitional institutions adequate to support continued rapid growth of China s non-state sector? We believe the answer is, again, no. There are several reasons for this. First, the playing field remains tilted against the NSOEs, and the impact is becoming more prominent as the NSOEs strive to expand. The transitional institutions are no longer equipped to support their continued growth. Second, due to important institutional changes concerning the NSOEs during the 1990 s, the main benefits of transitional institutions are fading away. Third, some accumulated distortional effects of the transitional institutions are difficult to correct and therefore likely to persist. A case in point is the difficulty in privatizing those NSOEs that were initially established as collective or ambiguously owned firms. Until they are privatized, these NSOEs will have difficulty even to compete in the market place, let alone to expand. Finally, severe difficulties exist for many indigenous private enterprises to grow into large, modern enterprises. The difficulties multiply due to the lack of the rule of law. The remainder of the paper is organized as follows. Section 2 clarifies the concept of the non-state sector in China and summarizes its performance in the past two decades. Section 3 explains why we believe that, during the reform era, the playing field has tilted against the NSOEs in China. Section 4 explores the institutional arrangements responsible for the rapid growth of the non-state sector during most of the reform era. Section 5 explains why those institutional arrangements are now inadequate and unlikely to propel further success of China s non-state sector in coming years. In the last section, we explore policy implications of our study and identify needed further research. 4

2. The Non-State Sector as the Engine of China s Growth 2.1. What is the non-state sector? We follow the official definition of the non-state sector in publications such as the Almanac of China s Economy. That is, the non-state sector includes all the production units that are not solely controlled and owned by the state or a government agency. In other words, the concept of the non-state sector is simply defined as all production units except those in the state sector. By this definition, the formal SOEs undergoing major ownership changes, for example by issuing equity shares in the stock market, are still called SOEs, as long as the state maintains controlling ownership. Under this rather broad definition, the non-state sector can be classified into collective enterprises, private enterprises, and enterprises of other ownership forms. Nominally, a collective enterprise is owned either by workers employed in the firm or by a group of community members, such as all the residents in an administrative jurisdiction. In reality, collective enterprises are under close control of the government. Major investment and employment decisions could not be made without government direction or approval. Unlike collective enterprises, private enterprises are owned by private individuals. Private enterprises can be further classified into individually-owned (Ge Ti Qi Ye) and privately-owned (Si Ying Qi Ye) enterprises. The difference between these two types of enterprises is that the former either involves no hired labor or hires immediate family members only, whereas the latter hires non-family members. From the mid-1950s, when private enterprises had been nationalized, until the start of the reform, the government adopted a very hostile attitude towards private businesses, including individually-owned businesses. In a series of political campaigns, private enterprises were confiscated and their owners persecuted. They did not really reemerge in China until years after the market-oriented reforms were introduced in 1979. Enterprises of other ownership types include foreign-invested companies and shareholding enterprises without government as major shareholders (otherwise, they are called stateowned shareholding enterprises, a new category of state enterprises). Typically, joint-venture 5

firms between state-owned enterprises and foreign investors are also counted as non-state enterprises. China s non-state sector can also be classified by its geographical location into the nonstate sector in rural areas and urban areas, respectively. In rural areas, thanks to the rapid decollectivization reform in the early 1980s, all agricultural production, except that of the few stateowned farms, belongs to the non-state sector. As for industrial activities in rural areas, the most prominent are those of rural industrial enterprises under the name of the township and village enterprises (TVEs). The historical origin of the TVE is those industrial accessories to collective farming units established under the People s Commune system, i.e., collectivized agriculture. They were initially termed commune and brigade enterprises. Given their accessory position, most commune and brigade enterprises did not even have independent accounting. Strictly speaking, these early commune and brigade enterprises were community-owned collective enterprises. TVEs acquired their current name in 1983, when the People s Commune system was abolished and Chinese agriculture was essentially de-collectivized. However, in most cases, the township (corresponding to the commune) and village (corresponding to the brigade) governments continued to maintain as close control of the TVEs as commune and brigade administrations did. In many TVEs, it is not clear who are the owners, who are the investors and how much they have invested, and who has what control rights. These features merit discussion and we will return to them later. In urban areas, the non-state sector consists of collectives, private enterprises, and firms involving foreign investors. The firms with foreign investment can be either solely owned by foreign investors or jointly owned by foreign investors and their Chinese partners. The Chinese partner of a joint venture could be a state-owned enterprise or a non-state enterprise. Because of their many unique features, these firms are typically not included in official statistics of the nonstate sector but treated separately. For the same reason, we omit them in our discussion. 2.2. Rapid Growth of China s Non-State Sector China s non-state sector has achieved very rapid development and this has been the most important factor driving the country s economic growth for the past two decades. One may argue about the accuracy of the statistics, but anecdotal evidence and casual observations do suggest 6

that the orders of magnitude portrayed in the official statistics are close to reality. Table 1 lists the annual growth rates of the state sector and non-state sector in total output and employment. As can be seen, from 1986 to 1999, the average annual growth rate of industrial output of collective enterprises and private enterprises was about 17% and 40%, respectively. During the same period, output of the state enterprises grew only 5% a year. Table 1: Growth Rates of Industrial Output: State Sector vs. Non-State Sector (Unit: Percent) Year SOE Non-State Enterprise Collective Other 1986 6.2 18.0 67.6 1987 11.3 23.2 56.6 1988 12.6 28.2 47.3 1989 3.9 10.5 23.8 1990 3.0 9.0 21.1 1991 8.6 18.4 25.3 1992 12.4 33.3 47.0 1993 5.7 35.0 66.2 1994 6.5 24.9 56.2 1995 8.2 15.2 51.5 1996 5.1 20.9 20.0 1997 1.0 10.2 15.4 1998 0.1 9.1 14.7 1999 8.8 6.0 14.4 Average 6.7 18.7 37.7 Note: 1) The growth rates are in real terms. 2) Omits foreign investment firms and Sino-foreign joint ventures. 3) Source: Table 13-4 of China Statistical Yearbook 2000. The fast growth of the non-state sector can also be illustrated by the development of TVEs in the rural areas. From Tables 2 and Table 3, one can see that from 1980 through 1995, the output of the TVE sector grew at an average annual rate of 34.5% in real terms: the output of the TVEs in 1989 was more than 11 times that of 1980 in real terms. In 1995, the output of TVE 7

was more than 6 times of that of 1990. From 1980 to 1995, total employment of TVEs more than quadrupled, increasing from 30 million to nearly 130 million. The average annual growth rate of employment is over 10%. Table 2. The Development of TVE 1980 1995 Number of Gross Output of Employees in Social Rural TVE Employees/Rural TVEs TVEs TVEs Labor Force Labor Forces Year (million) (billion RMB yuan) (million) (million) (%) 1 2 3 4 5 6=100X3/4 1980 1.4 65.7 30.0 318.4 9.4 1981 1.3 74.5 29.7 326.7 9.1 1982 1.4 85.3 31.1 338.7 9.2 1983 1.4 101.7 32.4 346.9 9.3 1984 6.1 171.0 52.1 359.7 14.5 1985 12.2 272.8 69.8 370.7 18.8 1986 15.2 354.1 79.4 379.9 20.9 1987 17.5 476.4 88.1 390.0 22.6 1988 18.9 649.6 95.5 400.7 23.8 1989 18.7 742.8 93.7 409.4 22.9 1990 18.5 846.2 92.7 420.1 22.1 1991 19.1 1,047.8 96.1 430.9 22.3 1992 20.8 1,594.2 105.8 438.0 24.2 1993 24.5 2,785.5 123.5 442.6 27.9 1994 25.0 3,761.2 120.2 446.5 26.9 1995 22.0 5,609.2 128.6 450.4 28.6 Note: 1. The Gross Output is in constant prices of 1980. 2. Before 1983, the data just included the Collective enterprises. 1984-1995, the data included all TVEs. 3. 1980 price=1990 price/1.1323, where 1.1323 comes from 958.11/ 846.16, 958.11 is the 1990's gross output of TVEs in 1990 price and 846.16 in 1980 price.(see Source 4 below) Source: 1. 1980-1992 data of column 2 is from China Statistical Yearbook 1993, P. 395, and 1985--1995 from China Statistical Yearbook 1996, P387. 2. 1980-1992 data of column 4 is from China Statistical Yearbook 1993, P395, and 1985-1995 from China Statistical Yearbook 1996, P.388 8

3. 1985--1995 data of column 5 is from China Statistical Yearbook 1996, P353, 1980-1984 data from China Statistical Yearbook 1983, 1984 and 1985. 4. 1980-1990 data (1980 price) of column 3 is from China TVE Statistical Yearbook 1991, P136. Data of year 1990, 1991, 1992, 1993 1994, 1995 in 1990 price are from China TVE Statistical Yearbook 1991 (P.140), 1992 (P. 140), 1993 (P.152), 1994 (P. 178), 1995 (P. 89), 1996 (P. 114) respectively. Table 3: Growth Rate of TVE Output and Employment 1981 1995 (%) Year Gross Output Employment 1981 13.5-1.0 1982 14.5 4.8 1983 19.2 3.9 1984 68.2 61.0 1985 59.6 34.0 1986 29.8 13.7 1987 34.6 10.9 1988 36.3 8.4 1989 14.4-1.9 1990 13.9-1.1 1991 23.8 3.7 1992 52.1 10.1 1993 74.7 16.7 1994 35.0-2.6 1995 49.1 7.0 Average Annual Growth Rate from 1980 to 1995 34.5 10.2 Source: See Table 2. As a consequence of the very rapid growth of the non-state sector and at the same time the much slower growth of the state sector, the landscape of the Chinese economy has completely changed in the twenty years since the inception of economic reform. Table 4 gives the share of non-agricultural employment of the non-state sector. It shows that, in 1985, when economic reform was first introduced in the urban areas, the state-owned enterprises employed about 70% of non-agricultural workers. In 1999, the share was reduced to about 41%. Measured by output, 9

the picture is even more striking. According to Table 5, the share of SOEs in the country s industrial output shrank from 65% in 1985 to less than 28% in 1999. Table 4: Share of Non-agricultural Employment: State Sector vs. Non-State Sector (Unit: percent) Year State- Owned Non-State Enterprise Collective Other Total 1985 70.0 26.1 3.9 30.0 1990 62.3 21.5 16.2 37.7 1995 59.1 16.4 24.6 41.0 1997 54.8 14.1 31.0 45.1 1998 43.8 9.5 46.7 56.2 1999 40.8 8.2 51.1 59.3 Note: 1) Each year s total non-agricultural employment is scaled to 100. 2) Source: Calculated from Table 2-3 of China Statistical Yearbook 1998 (before 1998) and Table 5-1 of China Statistical Yearbook 2000 (after 1998). Table 5: The Share of Industrial Output: State Sector vs. Non-State Sector (Unit: Percent) Non-State Enterprise Year SOE Collective Other Other Enterprises 1985 64.9 32.1 1.8 1.2 1986 62.3 33.5 2.8 1.4 1987 59.7 34.6 3.6 2.1 1988 56.8 36.1 4.3 2.8 1989 56.1 35.7 4.8 3.4 1990 54.6 35.6 5.4 4.4 1991 56.2 33.0 4.8 6.0 1992 51.5 35.1 5.8 7.6 1993 47.0 34.0 8.0 11.0 1994 37.3 37.7 10.1 14.9 1995 34.0 36.6 12.9 16.5 1996 36.3 39.4 15.5 8.8 1997 31.6 38.1 17.9 12.4 1998 28.2 38.4 17.1 16.3 1999 28.2 35.4 18.2 18.2 Note: 1) Each year s total employment is 100. 2) Other Enterprises including foreign invested firm and Sino-foreign joint ventures. 10

3) Source: Table 13-6 of China Statistical Yearbook 1998 (for data before 1990) and Table 13-3 of China Statistical Yearbook (for data from 1990 onward) 11

With its very rapid growth, the non-state sector has become the most important engine of growth of the Chinese economy. There are different ways to project this picture. Table 6 decomposes each year s increase in industrial output into that from the state sector and non-state sector, respectively. As can be seen from the table, before 1990, the state sector s contribution to the increase in industrial output fluctuated around 47% to 52%. Since 1996, the share of the state sector s contribution has been significantly lower. 1 Table 6: The Share Contributed to Industrial Growth: State Sector vs. Non-State Sector (Unit: percent) Year State-Owned Non-State Collective Other 1986 45.3 42.9 11.8 1987 48.8 39.3 11.8 1988 47.6 40.9 11.4 1989 52.5 33.5 14.0 1990 37.8 34.8 27.4 1991 43.7 36.1 20.1 1992 32.5 45.5 21.9 1993 31.4 39.1 29.5 1994 14.4 46.3 39.3 1995 33.5 14.6 51.9 1996 8.8 57.7 33.5 1997 4.7 29.1 66.2 1998-44.1 44.8 99.3 1999 27.6-15.9 88.3 Note: 1) Each year s total increase in industrial output is scaled to 100. 2) The negative share of contribution is partly due to changes of ownership forms of many enterprises in 1998 from State-owned enterprises to collective enterprises and from collective enterprises to private enterprises, respectively. 3) Source: Calculated from Table 13-3 of China Statistical Yearbook 2000. 1 One reason for the diminishing share of SOEs contribution to industrial output is privatization that transformed some SOEs into collective or private enterprises. We do not think this is all the reason. In fact, privatization was not adopted as a major policy until mid-1997 but the share SOE s contribution to output growth has been consistently lower than that before 1990. Poor performance of existing SOEs is a major reason for this. 12

3. How Level Is the Playing Field? Does the very rapid growth of the non-state sector in the past two decades imply that China has created a level (or even more favorable) playing field for the sector? We argue that to the contrary, the playing field has been tilted to the disadvantage of the non-state enterprises, especially the indigenous and small private enterprises. The tilted playing field has prevented the non-state sector from growing at its full potential. More recently, it has caused a dramatic slowdown of the growth of the non-state sector. 3.1. Legal and Bureaucratic Restrictions to Entry One major disadvantage of the non-state sector is a government ban on the NSOEs in many sectors of the economy, i.e., the NSOEs are not allowed to enter certain industries. A partial list of these sectors includes telecommunication, commercial and investment banking, insurance, import and export trade, and wholesale trade. In many cases, the NSOEs inability to enter these sectors of high-profit margins and high-growth potential is due to complicated bureaucratic approval procedures or entry requirements that are too hard for non-state enterprises to meet. For example, until recently, it was very difficult for the NSOEs to complete the bureaucratic procedures needed to obtain import permits (Jiang, 1999). Without direct channels of imports, many potentially profitable operations are not viable for the NSOEs. In another example, many hi-tech NSOEs are now looking to become contractors of the national defense industry, which enjoys high profit margins. But it is very difficult for a NSOE to clear the security scrutiny of the government and therefore to be approved to entry the industry. Table 7 shows the result of a survey of 338 private enterprises in 1999 by the International Financial Corporation. Over 50% of all types of enterprises indicated that the most important entry barriers are license requirements and government policy restrictions. Moreover, for larger private enterprises (over 500 employees), by far the most important barrier for them to enter an industry of their choice is obtaining a license. This shows that the larger the private enterprise, the more prominent is the constraint of government restriction on entry. 13

Table 7: Types of Entry Barriers by Firm Size (% of Survey Response) (Unit: percent of responses) Size of Employment Licenses (%) Policy Restriction (%) Local Protection Industry Monopoly Less than 51 22.6 29.0 6.6 29.0 12.9 51-100 50.0 12.5 12.5 25.0 0.0 100-500 37.5 37.5 18.8 0.0 6.3 More than 500 80.0 0.0 20.0 0.0 0.0 Market Size Note: 1) Based interviews with CEO s 338 private enterprises in Beijing, Shunde (Guandong province), Chengdu (Sichuan Province), and Wenzhou (Zhejiang Province) in the summer of 1999; 2) This is compiled from Table 4.1 of page 59 of International Finance Corporation (2000). Recently, China s commitments for membership of the WTO have created an interesting new twist to government restrictions on the entry of the NSOEs into certain industries. Under the accession agreements, China is required to phase out restrictions on foreign companies entry into Chinese markets. The liberalization, however, does not apply to domestic private enterprises. In addition to the unequal treatment that disadvantaged the NSOEs compared with the SOEs, the indigenous NSOEs will be disadvantaged with respect to foreign corporations as well. This awkward situation prompted many Chinese economists to argue and petition on behalf of the indigenous NSOEs for equal treatment of domestic private firms with foreign companies (under the interesting slogan of national treatment for national industries ). Some argue that markets should be opened first to domestic private firms to allow them time to prepare for anticipated foreign competition, and not yield too much market share to foreign entrants after the WTO accession. Reasonable as these proposals seem, it is far from certain that Chinese policy makers will adopt them, or, if they do, to what extent. 3.2. Credit Restriction Another often mentioned form of discrimination against non-state enterprises is state banks credit allocation. As many Chinese economists have observed (e.g., Jiang, 1999 and Fan, 1999), a disproportionately low amount of bank credit has been allocated to non-state enterprises. 14

Table 8 shows that, during the reform era, the state sector has been obtaining more funds for fixed investment than the non-state sector. This pattern of state banks favorable credit allocation to SOEs continued even when their share of output diminished dramatically. Table 9 is even more telling of this point. It shows that from 1994 through 1999, the non-state sector each year obtained no more than 15% of total short-term loans and almost no long-term loans from the state banks, while contributing to over 65% of the country s growth of industrial output during the same period! Table 8: Share of Total Fixed Investment: State Sector vs. Non-State Sector (Unit: percent) Year State- Owned Non-State Collective Other 1980 81.9 5.0 13.1 1981 69.5 12.0 18.6 1982 68.7 14.2 17.1 1983 66.6 10.9 22.5 1984 64.7 13.0 22.3 1985 66.1 12.9 21.0 1986 66.6 12.6 20.8 1987 64.6 14.4 21.0 1988 63.5 15.0 21.5 1989 63.7 13.0 23.4 1990 66.1 11.7 22.2 1991 66.4 12.5 21.1 1992 68.1 16.8 15.1 1993 60.6 17.7 21.6 1994 56.4 16.2 27.4 1995 54.4 16.4 29.1 1996 52.4 15.9 31.7 1997 52.5 15.4 32.1 1998 53.4 14.8 31.8 1999 54.1 14.5 31.4 Source: Table 6-3 of China Statistical Yearbook 2000. 15

Table 9: The Non-State Sector s Share in Total Short-Term Loans from State Banks (Unit: Percent) Year Collective Private Sino-Foreign Join Ventures 1994 7.4 0.6 2.9 1995 7.5 0.6 3.0 1996 7.0 0.7 3.3 1997 9.1 0.7 3.4 1998 9.2 0.8 4.1 1999 9.6 0.9 4.7 Note: 1) Each year s total short-term loan is scaled to100. 2) Source: Calculated from Tables 18-3, 19-3 of China Statistical Yearbook 1998, China Statistical Yearbook 2000, respectively. To illustrate the problem of credit constraint, let us look at the source of enterprises initial investment funds. Not surprisingly, China s NSOEs mostly replied upon self-finance for their initial investments. Naturally, this restricts the size of their operations at least at the beginning. Table 10 is a result of a survey of 338 private enterprises in the summer of 1999 by the International Financial Corporation. The survey asked for the source of the enterprises initial investment. By and large, over 80% of the initial investments were self-financed. Less than 6% came from bank loans. Interestingly, those newer enterprises in the survey seemed to have relied more heavily on self-financing than the older ones, indicating that in recent years, the problem of credit constraint has not been loosened for China s NSOEs. Table 10: Sources of Private Firms Initial Investment (Unit: percentage of total amount) Year of Operation Self Financed Bank Loans Institution Other By mid-1999 Less than 3 Years old 92.4 2.7 2.2 2.7 3 5 Years 92.1 3.5 0.0 3.8 5 10 Years 89.0 6.3 1.5 3.0 Older than 10 Years 83.1 5.7 9.9 1.3 Note: 1) Based interviews with CEO s 338 private enterprises in Beijing, Shunde (Guandong province), Chengdu (Sichuan Province), and Wenzhou (Zhejiang Province) in the summer of 1999; 2) This is compiled from Table 5.1 of page 67 of International Finance Corporation (2000). 16

3.3. Poor Legal Protections Poor legal protection of their property rights is another major problem faced by non-state enterprises, and especially by the indigenous enterprises. As is well known, China has yet to develop an independent judiciary system, so legal protection of rights is generally ad hoc. This is not nearly as much of a problem for the SOEs as it is for the NSOEs because the former group is itself part of the government apparatus and has close relations with administrative branches. Therefore, it is easy to understand why it is difficult for a non-state enterprise to get a fair ruling when engaging in a dispute with a state-owned enterprise. Because of this concern, non-state enterprises typically avoid legal cases against state enterprises, even when that entails an unfavorable settlement outside the legal system. For example, in a survey of 1171 private enterprises in 1993 and 1995, two Chinese researchers from the Chinese Academy of Social Sciences (CASS), Zhang and Ming (1999) found that, when facing business disputes, the overwhelming majority of the managers (71%) tried to deal with the disputes on their own. Only 6.5% of those surveyed said that they would resort to legal means. (See Table 11) We do not have similar statistics for the SOEs. But anecdotes tell us that the percentage of resorting to legal means should be much higher. Table 11: How Do Private Firms Deal with Business Disputes? (Unit: percent) Approaches Response (%) Ignore it 0.6 Deal with it by the firm itself 71.4 Seek help from Government 6.1 Resort to Legal Means 6.5 No Response 15.2 Note: 1) Based on a survey of 1171 private enterprises in 1993 and 1995; 2) This is compiled from Table 15 of page 150 of Zhang and Ming (1999). Meanwhile, China s regulatory agencies typically have the biased perspective that nonstate enterprises thrive by violating government regulations and evading taxes. The agencies are therefore much more stringent on non-state enterprise. Stringent inspections generate opportunities for extortion that in turn induces more frequent inspections. Thus, the regulatory burden on non-state enterprises is disproportionately higher than that on state enterprises. 17

Poor protection of the NSOEs property leaves them vulnerable to several forms of predatory behaviors. One problem is excessive taxes and fees. Another problem is the demand by bureaucrats for bribes before they provide services and protection. In the same survey mentioned above, Zhang and Ming (1999) found that over 37% of private enterprise managers interviewed listed trade between bureaucratic power and money (corruption) as a major factor slowing down their business. Interestingly, in 1995, arbitrary taxation was a major concern, but by 1997, it became less important. But corruption was consistently important between the two batches of the surveys. (See Table 12) Table 12: Most Important Social Problems Troubling the Operation of Private Firms? (Unit: percent) Social Problems Response in 1995 Responses in 1997 (%) (%) Unfair Income Distribution 5.1 9.9 Trade between bureaucratic power and 37.3 37.6 money (corruption) Public Crimes 20.6 41.1 Arbitrary Taxes and Fees 31.4 6.3 Others 10.7 5.2 Note: 1) Based on a survey of 1171 private enterprises in 1993 and 1995; 2) This is compiled from Table 13 of page 148 of Zhang and Ming (1999). 3.4. Tilted Playing Field: A Further Comment During the reform period, it was often said that the SOEs are disadvantaged in many ways in their competition with NSOEs. They are subject in some areas to more strict regulations. Government interference in SOEs operational decisions is also more direct and probably more frequent. As a consequence, the SOEs cannot shed surplus labor as easily, for example. They need to pay their workers (even those with no work to do) well enough to maintain their basic living standard and thereby help to maintain social stability. But they are not allowed to pay wages high enough to compete with private and foreign employers for special talents, such as managerial and technical professionals. Prompt and timely decisions are also more difficult to make because of strict bureaucratic reviewing and approval procedures. Based on these 18

observations, some people argue that, when all the distortions are considered and averaged out, the playing field is level for, or possibly even tilted to the advantage of, the NSOEs. Granted that the SOEs have their own problems, but it is a gross mis-perception to think that their problems essentially offset those of the NSOEs and conclude that the playing field is level after all. It would be even worse to assume that no further effort is needed to create an environment more conducive to the survival and growth of the NSOEs. The institutional constraints under which the SOEs operate and the associated problems caused thereby are not a consequence of the NSOEs. These constraints are long-standing; they existed long before the NSOEs became a major force in the Chinese economy. Policy reform is needed to eliminate constraints detrimental to both the SOEs and the NSOEs. In both cases, constraints cause inefficiencies. Both cases need to be resolved in the process of further reforms, with the government-business relationship and the country s legal systems at the core. 4. Non-State Sector Thriving on the Tilted Field: The Role of Transitional Institutions With all the discriminations against the NSOEs, it is very puzzling how China's non-state sector could have achieved such rapid growth in the past two decades. We argue here that several transitional institutional arrangements played an essential role in promoting the development of the non-state sector. 4.1. Intra-government Decentralization Decentralization within government itself was a main scheme of China s political reform in the past two decades. That reform provided local incentives for local governments to offer their helping hands to NSOEs, and proved to be indispensable in promoting the growth of the non-state sector. From the mid-1980's to mid-1990's, fiscal decentralization was implemented in China. In the fiscal reform, governments at different levels (e.g., the center and the provinces, provinces and municipals, etc.) first negotiated a set of revenue sharing rules amongst themselves. Also, in the mid-1980's, local government agencies were granted the right to create their own 19

business entities. The fiscal and administrative decentralization provided local governments both incentives and latitude to develop local businesses. But, for at least two reasons, local governments incentives for local economic development did not directly lead to a friendly environment for the development of private enterprises. First, many policies that discriminate against private enterprises extend directly from the center and it is beyond local governments power to remove them. Examples are sector entry restrictions and discriminatory access to credit. Second, rent-seeking motivations, either in the form of more tax revenues or more private benefits, mean that local governments real interest lies less in creating an environment in which private enterprises can prosper than in extracting benefits from them. One result has been that local government officials and entrepreneurs created many NSOEs with ambiguous property rights, in terms of both control rights and benefit distributions. (Weitzman and Xu, 1994, Chang and Wang, 1994, Li, 1996, Che and Qian, 1998). Under ambiguous property rights arrangements, a local government will offer an NSOE its protection and facilitation. Local governments have helped steer the NSOEs away from many existing and burdensome regulations, taxes, fees and harassment. They also have used their own power to grant land use rights to the NSOEs, lobby on their behalf for bank loans, and secure low-priced raw materials and transportation services for the NSOEs within their jurisdictions. Meanwhile, the arrangements allow the entrepreneurs to operate their enterprises. In exchange, the NSOE will yield to the local government some of its control and economic benefit. Given the discriminatory environment faced by the NSOEs, the protection and facilitation that local governments had to offer were valuable contributions, making the exchange worthwhile. In the 1980's and early 1990's, it was very common for a private founder of an NSOE to choose to register the firm as a collective rather than a private firm, although the latter option was officially also available. The popularity of the unusual ambiguous property rights arrangements is reflected in aggregate data. In fact, for most of the reform era, the rapid growth of the non-state sector was mainly attributable to that of collective enterprises. In contrast, before 1993, private enterprises had constituted a relatively small proportion of the non-state sector. As data in Table 5 show, before 1993, the private enterprises share of industrial output was no more than 6%. Their share only gradually increased after 1993. 20

How did local governments benefit from the rapid growth of the non-state sector? They did so in two ways political and economic thanks to the institutional arrangements that we explained above. By directly controlling the NSOEs in its jurisdiction, a local government found it easy to collect taxes and fees from the enterprises. More tax revenue meant more tax contribution to higher level governments. It also meant more investments in local public goods. Also, the development of the NSOEs led to more employment. This is especially important in areas with few SOEs (so that the emergence of non-state enterprises does not crowd out existing state enterprises and the net employment gain is large). Rising employment mitigated problems of social instability. Tax contributions, public goods, low unemployment and social stability are all measurable and valued indicators of the political achievements of local government officials and as bases for promotion. Rapid growth of non-state enterprises can indeed lay a solid foundation for the successful political career of a local government official. The second benefit to local officials of rapid growth of the non-state sector is economic. More tax revenue can be easily transformed into more on-the-job consumption such as better offices and cars, as well as more resources under the control of the government official in general. In many cases, local government officials were able to find ways to legally or illegally convert a considerable portion of the newly created community wealth into their personal wealth. 4.2. Information decentralization As we noted earlier, the lack of legal protection leads to discriminatory taxes and fees on NSOEs. To a certain extent, the problem is mitigated by their ambiguous property rights arrangement, since they can rely upon their sponsoring government agencies to fend off arbitrary taxes and fees from other agencies. However, the sponsoring government agencies themselves often made demands for excessive taxes, fees and profit remittance. In fact, many NSOEs surveyed complained that the demand (not actual payment) for taxes and fees exceeded their total operations revenue. 2 Table 12 shows that one of the top two concerns of surveyed private 2 According to Zhang and Ming (1999), the amount of fees paid by NSOE was as high as 30% of formal taxes. Some Chinese authors claim that the ratio is 170%. Also, the number of fees collected by local governments was more than 18,200. A newspaper article revealed that one local government collected fees based on the measure of air density. (Zhang and Ming (1999), page 140) 21

managers was arbitrary taxation in 1995. 3 This raises the question: How could any business survive, let alone prosper, under such heavy burdens of government taxation? An important transitional institution that has helped China s NSOEs in fending off the arbitrary demand of taxes and fees is information decentralization. Information decentralization greatly reduced the amount of information available to government agencies on NSOEs transactions and profitability and therefore limited the ability of the agencies to collect the arbitrary taxes and fees. The major institutional arrangement of information decentralization is what we call anonymous banking (Bai, Li, Qian, and Wang, 2000). That is, in the reform era, the Chinese government gradually but steadily loosened monitoring of enterprise transactions by the banking sector, which had been an essential part of the socialist central planning system. Not only was cash permitted for most business transactions, anonymous bank accounts were also allowed for NSOEs. Before the reform, enterprises had to conduct virtually all of their transactions in real name bank accounts. Cash and anonymous bank accounts were meant for use by households. Since the reform, the NSOEs are able to hide their transactions and profits from governments, which in turn serves the purpose of more conveniently and safely evading taxes and fees. Thus, although the government and its various branches and agencies at different levels have the coercive power to collect arbitrary taxes, in practice, taxes on NSOEs, especially smaller private enterprises, were difficult to collect. Effective tax rates are therefore much lower than what they are officially. This is clear from Table 13, which shows that, although towards the end of the 1990s the non-state sector annually produced about twice as much industrial output as the state sector, its contribution to government fiscal revenues remained less than half of that by SOEs. In contrast, although accounting for less than 26% of total industrial output and being generally in poor financial shape, the SOEs still contributed 70% of Chinese government tax revenues annually during this period. The low effective tax rates provide some de facto protection of NSOEs property and incentives for private businesses to grow. 3 In the 1997 survey, complaints on arbitrary taxation were much less than the 1995 survey. We suspect that this was due to the tax reform of 1994, which began to take effect after the three-year phase-in period. We have discussion on this in Section 5. More importantly, by 1997, the problem of arbitrary taxation has evolved into outright demands from government officials for bribes and bribe taking, which was the leading concern reflected in the 1997 survey. 22

Table 13. State vs. Non-State Sector: Government Budgetary Revenue (1) Budgetary revenue (billion yuan) (2) Budgetary revenue from state ownership (billion yuan) (3) Share of budgetary revenue from state ownership (%) (4) Share of state industrial output (%) 1978 113.2 98.5 87.01 77.6 1979 114.6 100.2 87.43 78.5 1980 116.0 100.7 86.81 76.0 1981 117.6 101.7 86.48 74.8 1982 121.2 103.3 85.23 74.4 1983 136.7 114.7 83.91 73.4 1984 164.3 136.0 82.78 69.1 1985 200.5 155.6 77.61 64.9 1986 212.2 166.2 78.32 62.3 1987 219.9 162.1 73.72 59.7 1988 235.7 168.8 71.62 56.8 1989 266.5 187.7 70.43 56.1 1990 293.7 209.5 71.33 54.6 1991 314.9 224.6 71.32 56.2 1992 348.3 248.3 71.29 51.5 1993 434.9 311.6 71.65 47.0 1994 521.8 372.7 71.43 37.3 23

1995 624.2 444.1 71.15 32.6 1996 740.8 28.5 Note: Source: Column (1): Government budgetary revenue is net of planned subsidies for the losses of stateowned enterprises. China Statistical Yearbook, 1997, p. 235, 237, 238; p. 413. China Industrial Statistical Yearbook, 1994, p. 27. 5. Transitional Institutions as Obstacles to the Non-State Sector s further Growth Since the mid- to late 1990 s, China s non-state sector has begun to show signs of slowing down. The pattern can be discerned from aggregate statistics. From Table1, one can see that, between 1986 and 1996, the average growth rate of industrial output of the collective enterprises was 23%; it was 8.3% between 1997 and 1999 and decreased by year. For private industrial enterprises, the average growth rate from 1986 to 1996 was 43%; it was 14.8% from 1997 to 1999 and also decreased by year. A contributing factor is of course the macroeconomic shock due to the Asian financial crisis. But more worrisome are major institutional reasons underlying the aggregate statistics. One of the major and positive measures in China s overall reform effort in the mid-1990s is the 1994 tax reform. However, the tax reform also had the unintended effect of reducing local governments incentives to promote the development of the non-state sector. Freer products and labor markets had similar effects. Meanwhile, continued local government interference has turned many collective enterprises into mini-soes laden with many of the long-lasting problems typical of the SOEs, such as poor incentives, inflexibility in adjusting employment levels, soft budget constraints and persistent financial losses. Difficulties in securing bank credits are limiting the NSOEs growth more than ever before, as the bigger NSOEs find greater difficulty in relying on the informal financial sector to satisfy their financial needs. The slowdown since the mid-1990s poses the question whether the transitional institutions discussed above can support continued growth of the non-state sector. In the subsections below, we argue that this is unlikely to be the case since the transitional institutions themselves have become increasingly ineffective in dealing with the difficulties faced by the nonstate enterprises. Moreover, while facilitating rapid growth of non-state enterprises in the past, 24

the transitional institutions have generated distortions of their own, such as the ambiguous property rights in many non-state enterprises. In order for the non-state sector to develop further, such distortions should be corrected and there are substantial obstacles to overcome. 5.1. Increased Adverse Impact of the Tilted Playing Field As has been argued above, the playing field has never been level for China s non-state enterprises. However, in the early years of the reform, most of the start-up non-state enterprises were able to overcome many government-created difficulties quite successfully by relying on the transitional institutions. Hence, the rapid entry of non-state enterprises that we observe. The situation is very different now, as the non-state enterprises have gone beyond the start-up stage and are aspiring to grow larger. The negative impact of the tilted playing field becomes ever more prominent, and the pain ever more acute. The transitional institutions are not nearly as helpful for the growing non-state enterprises as they were in the past. To understand the deepening impact of the tilted playing field, let us first look at the issue of entry barriers. Twenty years ago, the prohibition of entry to the high profit margin sectors such as financial services, telecommunications, wholesale trade, and import and export trade had more nominal than real impact on the development of the non-state sector, as most small-scale, start-up NSOEs did not target those sectors for entry. Indeed, restaurants, small-scale hotels, light manufacturing plants, small-scale retail outlets, etc, were their top choices for entry, and the barriers to these sectors were much lower. Today, however, competition in these sectors has become very fierce, and many successful non-state firms want to but cannot enter high profitmargin sectors such as financial services and telecommunication. Thus, legal prohibition of NSOEs entry into these sectors is now much more restrictive on non-state firms than in the early years of reform. Similarly, today s non-state sector suffers more from restricted access to bank credit and financial markets. Twenty years ago, when start-up non-state enterprises needed to raise capital, they bypassed the state-controlled financial institutions by relying on personal savings and borrowing from close relatives and friends. Since the initial capital requirement for the start-up was rather small and often below the minimal lending amount of a bank or other financial institutions, the lack of access to formal financial services was not a fatal restriction for many of 25

the new NSOEs. Today, for a successful non-state enterprise to grow bigger, the amount of capital needed is far beyond what can be raised from the informal financial sector. Thus, the lack of access to bank credit and formal financial markets has become ever more damaging. The same can be said about the former lack of impact of discriminatory regulation and legal framework on the NSOEs. In the preceding section, we mentioned that many NSOEs were able to protect themselves from government discrimination and intrusion by hiding their transactions and wealth, and this was possible because information in the economy was decentralized. It is more difficult for the NSOEs to do the same thing nowadays because, first, many of them are much bigger than before and the government scrutinizes them more carefully. Second, institutions such as anonymous banking have been abolished, as part of China s effort to modernize its economic institutions and prepare itself for the WTO accession. While the impact of a tilted playing field is increasingly prominent, the local governments, which used to be very supportive of non-state sector development and played active roles in helping non-state enterprise in overcoming the restrictions, are no longer effective in this regard. The reason is that the restrictions that today s non-state enterprises face are mostly sector-wide and nation-wide and are very difficult for individual local government agencies to change. For example, the sector entry prohibitions are set by national government agencies and are virtually impossible to evade by local agencies. Likewise, it is a set of policies and institutional arrangements of the central bank and nation-wide state commercial banks that have restricted non-state enterprise from obtaining necessary bank credit and accessing formal financial markets. Such restrictions are much more difficult to evade than those on land use, tax rates and business registration, even with the help of a local government. 5.2. The Diminishing Role of the Transitional Institutions Some transitional institutions responsible for the rapid growth of the non-state sector are themselves tapering off. This is due to a number of important changes. Specifically, the 1994 tax reform and the emergency of cross-regional labor migration led to major changes in economic incentives of governments. The 1994 tax reform aimed to consolidate the taxation right to the central government and give the central government a much larger share of tax revenue. The centerpiece of the reform was the establishment of a value-added tax (VAT) of 26

17% collected by the central government tax agencies residing in local areas. As much as 70% of the VAT goes to the central government. Moreover, the VAT, unlike any other previously collected taxes in the Chinese economy, is much easier to collect by the tax authorities due to the possibility of cross-checking among enterprises. While the merit of the 1994 tax reform needs to be judged by its longer run effects and warrants separate study, a prominent, near-term impact is the diminished interest of local governments in promoting non-state enterprises. Much of the tax contribution of these enterprises goes to the central government, while efforts to promote their growth stay with local government officials. As a consequence, local government officials are less motivated to promote the growth of the NSOEs and rationally divert their efforts to other areas. Recent years have seen increased cross-regional labor migration in China. Again, while this is good for China s long-run efforts to establish a market economy, it has a negative effect on local governments incentives in promoting non-state enterprises. Historically, non-state enterprises, especially TVE s, played the role of mitigating unemployment pressure in a local region. Reducing unemployment among local residents by increasing employment in the nonstate sector is a major policy objective of many local governments. In recent years, relaxed domicile control, improved transportation, and better information flow are three of many factors that have contributed to increased labor migration from one region to another. As a result, a typical labor-intensive non-state enterprise hires mostly cheap labor from outside the region in which it resides. Meanwhile, local residents have the option to leave for other regions for employment. Thus, non-state enterprises are no longer as effective instruments in providing employment for local residents, further reducing the interests of local government officials in promoting the growth of non-state enterprises. A number of recent political events also contributed to diminished interests of local government officials in non-state enterprises. The most obvious is the intensified anti-corruption campaign in recent years. The motivation for quelling corruption is political, i.e., to secure political support and shore up the legitimacy of the communist party. Although the anticorruption campaign should bring some long-term economic benefit, it also has a short-term economic cost. For local government officials, personally benefiting from the development of the NSOEs is now more risky. This is more so since the anti-corruption campaign is often 27