The Political Economy of Chinese-Style Privatization: Motives and Constraints

Table 4. The Progress of Ownership Restructuring in the Chinese SOEs (1997 – mid 2004): Evidence from the Survey Data

	Total Firms Restructured			In Which
				Central Gov. Controlled Firms		Local Gov. Controlled Firms
	Firm No.	% of the restructured firms	% of the whole sample	Firm No.	% of the sub-sample	Firm No.	% of the sub-sample
Pre-1997	13	1.1	0.5	2	0.2	11	0.7
1997	30	2.6	1.1	7	0.6	23	1.4
1998	40	3.5	1.5	10	1.0	30	1.8
1999	80	7.0	3.0	15	1.4	65	3.9
2000	121	10.6	4.4	24	2.2	97	5.9
2001	221	19.4	8.1	90	8.6	131	7.9
2002	217	19.0	8.0	31	3.0	186	11.3
2003	305	26.7	11.3	54	5.2	251	15.2
2004	115	10.1	4.2	24	2.3	91	5.5
Sub Total	1,142	100	42.1	257	24.6	885	53.6
Whole Sample	2,696		100	1,044	100	1,652	100

Source: Liu and Liu (2005), the Unpublished Report of the Year 2004 Enterprise Survey on Ownership Transformation in China.

Decentralized State Asset Administration

Fiscal & Banking Reform

Increased Market Competition

Control & Income Rights of Local SOEs

Harder Budget Constraints of Local Economies

Economic Growth Determined by the Performance of Local Firms

Private Benefits Derived from Local Firms

Career Advancement

Cadre Evaluation System

The Payoffs of Local Bureaucrats

Figure 1. Institutional Environments Shaping the Payoff Structure of Local Bureaucrats since the Mid-1990s. The figure illustrates the payoff structure of Chinese local government leaders, which is largely shaped by the evolving institutional environments during the reform era. The payoff of local leaders is composed of two parts: their career concerns for further promotion and the private benefits they can derive from their control of local firms. Since in the post-Mao cadre management system, the economic performance under each leader’s jurisdiction has been given a large weight on future promotion decisions made by her superiors, local leaders have strong incentives to promote their local GDP growth. On the other hand, boosting the profitability of local firms has become indispensable to achieving impressive growth records due to the following factors: first, the decentralized state asset management system delegates the control and income rights to local governments of a large number of small and medium SOEs. Second, with the intensified cross-regional competition, it is more and more difficult for local governments to help their firms maintain monopoly profits by erecting effective trade barriers. Finally, the fiscal reform in the 1990s and the ongoing banking credit centralization served to significantly harden the budget constraints of local governments, in the sense that 1) they need to finance local economies chiefly from their own revenue resources such as local firms, 2) cheap and easy credit from the banking sector is harder to obtain for propping up local firms and financing fixed-asset investment.
Table 5. What Major Constraints do Firms Face during

Ownership Restructuring?

Major Constraints		No. of All Choices		% of All Choices
Difficulty in raising funds to cover the restructuring costs		434		35.7
Debt problems hard to solve		259		21.3
The higher authority does not allow the firm to be sold		15		1.2
Sale price is too high		15		1.2
No buyers interested in the firm		35		2.9
Others		458		37.7
Total		1,216		100

Source: Liu and Liu (2005), the Unpublished Report of the Year 2004 Enterprise Survey on Ownership Transformation in China.

Notes: The main question of the questionnaire is shown as the table title, and the managers are allowed to make multiple choices. The category of ‘Others’ pooled all the other miscellaneous factors the managers mentioned during the survey, so the largest percentage shown in the last column does not necessarily indicate that they face another major constraint besides worker compensation and bank debts.

Table 6. Probit Model Estimations Examining the Motives and

Constraints of Privatization in China

	Ownership Switch from State (0) to Private (1)
	Whole Sample	Sub-Sample 1	Sub-Sample 2	Sub-Sample 3
Intercept	0.521 (0.602)	5.542 (1.117)	6.057 (0.747)	5.558 (0.708)
Log (Sales) t-1	0.177*** (6.321)	0.240*** (4.898)	0.175** (2.244)	0.221*** (2.456)
(Profit Margin) t-1	-0.210 (-1.25)	-0.536*** (-2.404)	-0.708** (-2.219)	0.552 (0.802)
(Worker Compensation Rate) t-1		-0.012 (-0.444)	-0.008 (-0.190)	-0.108** (-1.964)
(Debt Ratio) t-1	-0.445*** (-2.514)	-0.405* (-1.5)	-0.711** (-1.896)	0.107 (0.291)
Year Dummies	Included	Included	Included	Included
Industry Dummies	Included	Included	Included	Included
Log-likelihood	-409.12	-152	82.468	-39.666
No. of Observations	847	272	153	119

Notes: a. Firm-level data are from the Year 2004 Enterprise Survey on Ownership Transformation in China, which was organized by the Institute of Enterprise Research, the State Council Development Research Center and the World Bank. All the sample firms were originally SOEs but have already started ownership restructuring, with a significant number of them being fully privatized during 1997-2003.

b. Sub-sample 1 includes the firms for which the management estimation of labor compensation costs is available. Within sub-sample 1, sub-sample 2 includes firms currently or previously controlled by municipal governments, while sub-sample 3 includes those currently or previously controlled by provincial and central governments.

c. The dependent variable: Ownership Switch is a dummy variable that equals 0 if the firm remains state-controlled and equals 1 if it was fully privatized during the period. Here ‘fully privatized’ means the change of the largest shareholder from government agencies to private ones.

d. Explanatory variables: Log (Sales)_t-1 is the sales revenue in logarithm form with one year lagged. (Profit Margin)_t-1 is the one-year lagged ratio of net profit to sales revenue. (Worker Compensation Rate) _t-1 is the lagged labor compensation cost per employee estimated by firm managers should privatization happen in year t. (Debt Ratio) _t-1 is the one-year lagged ratio of total debts to total assets.

e. T-Statistics are reported in parentheses, with ***, **, and * marking significance at the 0.01, 0.05, and 0.10 level respectively.