Desktop version

Home arrow Political science arrow Dancing with the devil : the political economy of privatization in China

The TVE Spectacle

Township and village enterprises (TVEs), known as xiangcun qiye in Chinese,[1] were publicly owned companies that primarily undertook nonfarm economic activities in rural areas. They were the main driving force of China’s transformation toward a market-oriented economy during the 1980s and early 1990s. The predecessors of TVEs were commune and brigade enterprises (CBEs) formed during the people’s commune era (1958-1983). In 1978 they were the only type of officially allowed off-farm economic organization in rural areas and accounted for 9% of the rural workforce and 17%


of rural GDP (SCTE, 5, 7). In 2008 the percentages of the nonfarm sector (inclusive of both public and private entities) in these categories increased to 33% and 72% respectively (CTEAPPY2009, 99; CSY2010, 38, 117).

Throughout the 1980s and the early 1990s, TVEs led the expansion of the rural nonfarm sector, largely as a result of township and village authorities’ active efforts to promote rural industrialization after agricultural decollectivization (Walder 1995; Oi 1999). There were three possible strategies for local authorities in this undertaking: focusing on output growth, maximizing profit, and expanding sales. Output growth was a major indicator in the performance assessment of local leaders. Profits of TVEs provided the basis for enterprise income tax, and the surplus after tax added to the pool of resources controlled by local officials. Expansion of sales helped generate indirect taxes, which constituted the main source of local revenue, as discussed in chapter 3. Nonfarm employment could also increase as a result of the growth of output and sales, which often required more labor input. Improvement in profitability did not necessarily lead to this, though, depending on whether it was achieved with a concurrent expansion of the nonfarm workforce. It would nevertheless have been ideal for TVEs to grow output, sales, and profits all together.

In reality, however, local officials faced constraints that tended to drive them to anchor their strategies in the growth of sales. A major challenge to focusing on the expansion of output by TVEs was cash flow, which would be seriously affected by large inventories of unsold products.[2] Recounting the experience of running TVEs in the late 1980s and early 1990s, for example, a township official I interviewed in Wuxi county in 1995 had the following observation (informant, 27/1995):

Yes, output mattered greatly to our [local] leaders. But it’s not feasible to grow output independent of sales. We were not part of the [centrally] planned economy, and there was no matching supply of capital for our output. ... We had to work hard to sell our products, as much and as quickly as possible, even below cost in difficult times, so as to make sure that the cash flow would keep going.

Another factor that could accentuate the attention of local officials to sales was a built-in ratchet effect from the careerist pursuits of local leaders. The new political performance assessment system fostered an unintended incentive for local leaders to exaggerate the growth of economic output during their transient tenures (Cai 2000).[3] For an incumbent to outperform his or her immediate predecessor and peers elsewhere, there was pressure to deal with the gap between fabrication and reality. Even if the incumbent chose to refrain from further exaggeration, the bar had already been raised by his or her predecessor for subsequent assessments. Adding more exaggeration on top of it would inevitably increase the level of political risk. Maintaining sufficient maneuvering space between these constraints required real output to grow above the existing level,[4] which could hardly be realized without strong sales given the cash flow problem just mentioned. In fact there is no evidence that TVEs accumulated large amounts of inventories as a result of output growing at faster pace than sales. During 1993-1998 (when pertinent data are available), the average annual inventory turnover rate for TVEs was 7.3 (CTEY, various years), which was roughly at the same level as the concurrent average rate for companies listed on the Standard and Poor 500 Index.

On the other hand, promoting profit growth faced two major challenges. Despite local governments’ efforts to help TVEs through regulatory laxity and resource support, the level of profitability was largely shaped by competition, which lay beyond the control of local officials and intensified with the deepening of reforms. A survey of TVEs (n = 300) jointly conducted by the Chinese Academy of Social Sciences and the World Bank in 1991-1992 found that 96% of them considered themselves to be operating in competitive (31%) or highly competitive (65%) markets.[5] The uncertainties in such environments made it difficult for local officials to anchor their economic strategy in profit growth. Moreover, profit maximization could be achieved through cost saving and even organizational downsizing, and thus it could be inconsistent with other important goals for local officials, such as output growth, revenue, and job creation. As a top economic official in a rural township of Suzhou put it (informant, 14/1995):

Profits are important, but not absolutely important if we [the local government] consider the full range of issues we need to address. Actually our collective enterprises had a good number of years of high profitability back in the 1980s, when many SOEs were still bound by the old system and foreign companies were not in yet. Those days are long past. Not only is it more and more difficult to get high returns, but insisting on it may stand in our way forward. We need volume growth to boost the overall scale of our local economy, to broaden the bases for taxes and fees, and to expand employment opportunities. How can a local leader expect to be promoted without these ?

What he meant by “volume growth,” as he clarified at my request, was sales growth. This is not surprising. An examination of the economic statistics of the township reveals that during 1984-1995 the average annual sales growth rate of local TVEs was 25.9%, whereas industrial output, budgetary revenue, and nonfarm employment on average grew by 23.5%, 24.1%, and 13.8% per year respectively. Of the six party secretaries who led the township during this period, four were promoted, one retired, and one was laterally transferred to lead another township in the county (in the aftermath of the economic slowdown following the June Fourth Incident in 1989). Of the five township heads, three were promoted, one died in a car accident, and one left the government to take over the helm of a private business company owned by his extended family. Like those in many other locales in the region, TVEs in the township ran into deep financial difficulties in the following years and had to be privatized around the turn of the century. But none of those who had overseen the earlier boom were held accountable for that; nor were their successors. When I revisited the township in 2003 only three of the thirty-six township-owned enterprises that had existed in 1995 remained. Also, these survivors were all reorganized as shareholding enterprises with diluted stakes held by the township government.

The pivotal importance of TVE sales growth was also recognized by the overwhelming majority of the rural officials I interviewed in thirteen townships in Hebei, Liaoning, Shandong, Jiangsu, Zhejiang, Guangdong, and

Sichuan during 1995-1999.[6] Of the forty-eight interviewees, forty-four strongly agreed that it was clearly the focal strategy for promoting local TVEs during their booming years. One of them (informant, 9/1996), the head of the public finance department in a rural township in Heze of Shandong province, characterized sales growth as the “key link” that helped address multiple agendas of the local government. In particular, if and when sustained, it would help generate revenue that benefited different groups of local officials, as he explained:

Making profit or not, our [collective] enterprises generate fiscal revenue as long as they can sell their products. They provide the basic stream of [fiscal] revenue we count on. This is not only because revenue is an important issue in the performance evaluation package for our township leaders, but because it also provides for the needs of everyone else—including those not even on the regular payroll—in the government, salaries, bonuses, fringe benefits, office supplies, transportation, entertainment of guests, not to mention things like roads, schools, and so on. ... Why not count on private companies for these ? Well, they may be helpful to a certain extent, but they are unwilling to stay the course for long without making money, especially during difficult times. Collective enterprises need to follow orders from the government, and we of course also need to keep them afloat and growing with all the necessary support for as long as we can.

Statistical evidence sheds further light. The 1991-1992 CASS-World Bank survey of three hundred TVEs mentioned above found that during 1984-1990 the sales volume of the sample enterprises on average grew by 17.4% per year, which had a Pearson correlation of .99 with output growth. The average ratio of sales to output in that period was 92.7%, with a standard deviation of 3.5. What these findings indicate is that output grew in close association with sales, but it did not outpace the latter significantly, which could have been the case if output expansion had been the dominant strategy. Sales growth therefore seems to have provided the bedrock for output growth (but not the other way around), which mattered to local governments too. Despite the strong sales growth, however, profits failed to keep pace with it. Figure 4.1 shows that the ratio of profit to sales trended down, whereas the ratio of taxes to sales held steady. The negative correlation of sales and profits suggests that

Taxes and profits versus sales, 1984-1990 Source

FIGURE 4.1 Taxes and profits versus sales, 1984-1990 Source: 1991-1992 CASS-World Bank TVE survey data (n = 300).

the growth of sales was unlikely to be the result of a profit maximization strategy.[7] Instead, the various (tax) revenues derived from the expansion of sales appear to have been the main driving force.

The larger picture at the national level was quite similar. Figure 4.2 shows the pattern of TVEs’ sales growth rate up to the time of massive privatization. It demonstrated a strong upward trend in the mid to late 1980s. That was followed by a brief and moderate slowdown during the economic retrenchment in the aftermath of the June Fourth incident in 1989. But the upward trend resumed in 1992, with an even stronger tempo. It peaked in 1995 and then began a precipitous decline. On average the annual growth rate of TVE sales was 31% during 1984-1995. Figure 4.3 further shows a relatively stable trend in the correlation of sales to (gross) output and GDP during 1985-1998. On average the ratio of sales to output was 87% and the ratio of GDP to sales was and 29% (standard deviation = 1.8). There is no evidence of output significantly outgrowing sales. As mentioned in the preceding section, gross output was the leading economic indicator in political performance assessment until 1993, whereas GDP became the focal measure thereafter. The growth of sales therefore provided an essential platform for addressing the primary career concern of local officials in the first two decades of reform.

The relationship of sales to profit and taxes tells a different part of the story. Figure 4.4 shows that over time the tax revenues generated through

Sales growth rate (five-year moving average) of TVEs, 1980-1998 Source

FIGURE 4.2 Sales growth rate (five-year moving average) of TVEs, 1980-1998 Source: CTEY, various years.

TVE output, sales, and GDP, 1985-1998 Source

FIGURE 4.3 TVE output, sales, and GDP, 1985-1998 Source: CTEY, various years.

each unit of sales held steady but the profits generated through each unit of sales declined (through 1993) and then stagnated.[8] As discussed before, the bulk of the taxes generated by TVEs consisted of indirect taxes tied to sales (table 3.3). During 1987-1998 (when pertinent data are available) their share

Taxes and profits as percentage of TVE sales, 1978-1998 Source

FIGURE 4.4 Taxes and profits as percentage of TVE sales, 1978-1998 Source: CTEY, various years.

Debt, sales, and profit of TVEs, 1984-1998 Source

FIGURE 4.5 Debt, sales, and profit of TVEs, 1984-1998 Source: CTEY, various years.

in the total tax liability of TVEs ranged between 74% and 83% (CTEY, various years). Given such importance to the interests of incumbent local officials, it is not surprising that TVE sales kept expanding despite declining profitability.

What sustained TVEs’ spectacular growth of sales despite declining and poor profitability throughout the 1980s and early 1990s was the supply of credit leveraged with the help of local authorities. Figure 4.5 shows that external borrowing kept pace with sales during the booming years of TVEs, which received the strongest boost in the late 1980s and early 1990s. It is interesting to note that the importance of direct bank lending declined in TVE liabilities, especially in the 1990s. The 1991-1992 CASS-World Bank survey reveals that on average bank loans accounted for some 81% of the borrowings of the sample enterprises, which tended to be larger in size and thus more resourceful than smaller ones. But even the larger ones might have faced increasing difficulties with the progression of the decade. The percentage of bank lending in total TVE debt was 38% in 1993 (CTEY1994,), which indicates a strong role played by nonbanking sources of financing. One such source lay in rural cooperative funds that began to emerge in 1984. These funds were set up and run by township and village authorities by absorbing deposits from rural residents, oftentimes in competition with the state-owned Agricultural Bank and the rural credit cooperatives it controlled. In 1996, when the central government decided to crack down on rural cooperative funds, there were some twenty-one thousand township-run funds and some twenty-four thousand village-run funds. The amount of funds totaled about 150 billion yuan (Wen Tiejun 2000), which was equivalent to about 12% of the total liabilities of TVEs in that year.[9]

Perhaps an even more important source of nonbank lending consisted of interfirm loans incurred by TVEs through transactions on credit (CTEY 1994, 385-386), with township governments serving as guarantors (Ma Rong, John Wong, Wang Hansheng, and Yang Mu 1994). The exact magnitude of such lending is unknown. A large part of it was ultimately owed to state-owned banks, as the credit extended by the lending firms was often backed with loans from banks. When serious delays in loan repayment occurred, it contributed to the growing problem of “triangle debt” (sanjiao zhai), which depicts a triangular debt loop (where A owes to B, B owes to C, and C owes to A) but more generally refers to a series of intertwined debt chains that plagued public enterprises and state-owned banks alike in the early to mid-1990s (Lardy 1998). Indeed, the ability of TVEs to fuel sales growth depended greatly on cash flows facilitated by delays and rollovers in the repayment of principal.

In the 1991-1992 CASS-World Bank survey of TVEs mentioned above, 84% of the enterprises indicated that, when faced with difficulties in servicing their loans, they were able to obtain relief from local governments by deferring repayment, covering old loans with new borrowing, and/or using pretax profit for repayment. In 1995 the total principal repayment by TVEs was 9.7 billion yuan, which was equivalent to only 0.77% of the outstanding debts

(of 1.25 trillion yuan) (CTEY1996, 336, 341).[10] On the other hand, interest payment totaled 62.6 billion yuan, which was equivalent to 42% of the profits earned by TVEs in that year (CTEY 1996, 336, 341).[11] Apparently out of line with common practices of banking and financing, this peculiar structure of interest-principal amortization nevertheless reveals a major twist in the TVE growth story. With the accumulation of outstanding debts and the decline and stagnation of net profits relative to sales, the ability of TVEs to service their growing debts deteriorated. This is illustrated by the second curve (i.e., ratio of profit to debt) in figure 4.5. When the extensions for principal repayment could no longer be obtained or sustained, the sales growth strategy would inevitably run into difficulties.

Table 4.1 shows the debt level of TVEs relative to their total equity capital and to the revenues of township governments during 1993-1998 (when pertinent data are available). Although the debt-equity ratio of TVEs was generally lower than that of SOEs (to be discussed below) and registered some

Table 4.1 Financial liabilities of public enterprises versus equity and government revenue


Debt-equity ratio

Debt-revenue ratio


All SOEs

Local SOEs


Local SOEs





































Note: Debt-revenue ratio is defined as total TVE liability divided by total revenue of township governments (including budget revenue, extrabudget revenue, and “self-raised funds") for TVEs, and as total local SOE liability divided by budget revenue of provincial, prefectural, and county governments for local SOEs.

Sources: CSOAY (various years); CTEY (various years); FSPCC (various years).

decline from 1993 to 1998, it still exceeded their equity value by more than 50% in 1998. That implies that there would have remained a 34% gap had they been forced to liquidate all the assets of their own to pay off their debts. In fact the “improvement” shown by the downward movement of the debt- equity ratio in 1993-1998 reflects the changes resulting from early privatization of TVEs in an increasing number of locales during that period, which saw the disappearance of some 37% of all TVEs and, along with it, the financial liabilities on their books.[12] As I will show later, TVEs with more serious financial problems were more likely to be among the early dropouts (hence diluting the debt statistics).

The growing liabilities of TVEs also added to the fiscal strains faced by township governments, as can be seen from the debt-revenue ratio in the table. The problem was even more formidable than that faced by the “owners” of local SOEs, as the denominator for TVEs includes not only budget revenue but extrabudget revenue and “self-raised” funds (discussed in chapter 3). Moreover, the overleveraged growth of TVEs eventually turned them into casualties of their own earlier success in job creation. In 1978 they employed a total of 28 million workers; in 1995 TVE employment peaked at 61 million (SCTE, 103-105). Their inability to further contribute to nonfarm employment afterward was soon complicated by the need to accommodate large numbers of workers who were rapidly rendered redundant by the drastic shrinkage of sales growth. In face of the surge of combined financial and un(der)employment pressures, privatization became the only viable way out for local authorities, as will be discussed in chapter 7.

The deterioration of financial health and longer-term sustainability was not the only problem associated with the sales growth strategy that made TVEs increasingly vulnerable to major external shocks. The internal organization and governance of TVEs were also negatively impacted as a result of local officials’ using them as instruments for careerist pursuits and revenue manipulation. A common practice adopted by township authorities during the booming years of TVEs was to hide taxes in enterprise accounts (Ma Rong et al. 1994), which had been particularly widespread before the 1994 fiscal restructuring.[13] What they did was to refrain from “overfulfilling” the fiscal contracts with higher-level authorities by not collecting the full amounts of taxes, especially indirect taxes, owed by TVEs so as to stash away revenues that otherwise would have to be shared with higher-level authorities, and even to inflate profitability (which could also be a plus in political performance assessment). As shown in table 3.3, there was a sizable gap between “taxes due” and “taxes paid” by TVEs. Although the taxes exempted had the effect of helping some TVEs to stay in competition, they were funds subject to recall by the (local) government rather than belonging to the enterprises concerned. Indeed the uses of such funds were oftentimes not for the longer-term development of the enterprises concerned, but for various discretionary purposes defined by local officials, ranging from private and collective consumption to the financing of activities (e.g., “image projects”—see Guo 2009 for a discussion) irrelevant to the pertinent account holders that provided the hiding service. Because of the shuffling of closely related accounting items in revenue manipulation, local officials sometimes even “conveniently” expanded these uses beyond the slush funds parked in TVE accounts, thus undermining the financial independence and health of the enterprises affected.[14]

On the other hand, the involvement of TVE managers in the manipulation of fiscal flows through enterprise accounts fostered a collusive relationship with local officials. Their collusion could degenerate into a common base for naked pursuit of private gains at the expense of TVEs.[15] Even without further slippage down this path, the elevation of informal dependence of local officials on TVE managers in revenue maneuvering could compromise their formal fiduciary role (as representatives of rural communities) in monitoring the behavior of the latter. This problem was exacerbated by a concurrent rise of monitoring cost amid fast expansion of TVEs driven by the sales growth strategy. From 1985 to 1995 the total assets of TVEs increased from 118 billion yuan to 1.99 trillion yuan, whereas the workforce grew from 42 million to 61 million (CTEY1988, 273; CTEY1996, 100). The resultant increase of organizational scale and complexity added to the information asymmetry faced by supervising authorities (Li and Rozelle 2004), leaving open more opportunities for opportunism to grow.[16] As I will discuss in chapter 7, managerial corruption and asset stripping were among the factors contributing to the decline of public enterprises, which was further hastened by insiders’ strategic manipulation of enterprise asset valuation before and during massive privatization.

  • [1] TVEs should not be confused with enterprises under township authorities, known as xiangzhen qiyein Chinese. The latter consist of both public and private nonfarm economic organizations in rural areas.
  • [2] See Ma Rong et al. 1994 for a sample of case studies that illustrate this problem.
  • [3] In 1993, for example, TVEs reported a total output of 2.065 trillion yuan, a total sales volume of1.742 trillion yuan, and a total inventory value of 262 billion yuan (CTEY 1994, 367). The differencebetween output and sales was 362 billion yuan. Even if all the inventory consisted of current-year unsoldoutput (with unused supplies and past unsold output being assumed as zero), one would still find a gapof 61 billion yuan unaccounted for, which most likely came from exaggerated output.
  • [4] Debunking the statistical fabrication by the predecessor was another way to cope with this issue. Butit was not without serious drawbacks. As one former township head in Shunde of Guangdong province(informant, 5/1996) pointed out: “You don’t want to create new enemies unnecessarily; you may not befully aware of how well connected your predecessor is in the county and even beyond; you don’t knowwhether what you do will offend those (assessors) who have given a pass to his performance; and youdon’t want to trigger a close scrutiny of your own record—past and future; in short it (reporting exaggeration by predecessor) is a very risky move, even if he (the predecessor) is already in trouble.”
  • [5] Full documentation of the survey is posted at the book site.
  • [6] The basic biographic profile of these officials is posted at the book site.
  • [7] This, of course, does not imply that profits were unimportant to local governments. But, rather,declining profitability alone would not have undermined their dominant strategy.
  • [8] Profit figures are adjusted to reflect the effect of tax exemptions booked as profits. An explanatorynote on this is posted at the book site. It should be noted that this adjustment does not even accountfor the fact that as of 1993 a significant portion of what was booked as profit had to be used for interestpayments on the loans that TVEs incurred to finance and sustain their expansion (see the discussionbelow).
  • [9] This, of course, does not mean the full amount was channeled to TVEs. For an account of the riseand demise of rural cooperative funds during 1984-1999, see Wen 2000.
  • [10] Detailed breakdown figures are unavailable for earlier years. The percentages for 1996, 1997, and1998 were 0.9% (of 1.45 trillion yuan), 0.88% (of 1.55 trillion yuan), and 0.88% (of 1.59 trillion yuan)respectively (CTEY 1997, 356, 367; CTEY 1998, 392, 297; CTEY 1999, 421, 427).
  • [11] During 1986-1992 public enterprises were allowed to use part of their gross profit (before income tax) for repayment of loan principal. That practice, known as shuiqian huandai (repayment of loans before tax), was discontinued as of 1993.
  • [12] In 1993 the total number of TVEs was 1.69 million; in 1998 it dropped to 1.07 million (SCTE,55-56).
  • [13] As noted in chapter 3, the 1994 fiscal restructuring did not change the various arrangements ofrevenue sharing between township and county governments. But it did make the exemption of certaintaxes (e.g., value-added tax) more difficult. I will discuss the constraining effect of this change on theincentives for local governments to continue the sales growth strategy in chapter 7.
  • [14] A common observation that I made during my fieldwork trips to the thirteen townships (mentioned above) during 1995-1999 is that the bills for entertaining township officials’ visiting guestswere always paid by their subordinate TVEs. This, of course, was only part of the “bursary” service provided by TVEs. A TVE manager I interviewed in Shunde (informant, 12/1996) acknowledged having receiving tax breaks for the township government but claimed that “we purchasedthe Toyota car that chauffeurs the (township) party secretary; we share [with other enterprises] theexpenses for some of their duty trips, as well as the gifts they carry along, to the county [government], the provincial capital, and other cities; we contribute to the fees and other expenses incurredby township leaders’ children attending boarding schools in the county; last year we were asked topay for the cement and sand used in the construction [at the suggestion of the county party secretary] of the monument commemorating revolutionary martyrs in the war of resistance againstJapan; we shared the cost for replacing the water pipes for the canteen of the township government;we were one of the sponsors for the lantern fair after the Lunar New Year. ... There are just toomany items to list out in full. What I can say is that in return we have actually contributed morethan what we have received.”
  • [15] In a previous study (Lin 2001) I have discussed how such collusion undermined the organizationalhealth of public enterprises. For a sample of corruption cases involving collusion between townshipofficials and TVE managers, see Wang Guangying 1996 and Jin Qiang 2005.
  • [16] According to a township official (informant, 21/1995) in Wuxi county—the heartland of the TVEboom—during 1993-1994 more than a dozen local TVEs had to be liquidated because of heavy lossesand debts. Their remaining assets were taken over by private business people. “They expanded too fastsuch we could not keep a close eye on them. The managers took advantage of this and entered deals thatbenefited their friends and family members at the expense of their enterprises. Some of them have nowbeen disciplined, and we are still investigating these cases"
< Prev   CONTENTS   Source   Next >

Related topics