Wages, Human Capital, and Work Incentives in a Transition Economy: The Case of Mainland China Belton M. Fleisher ∗ and Xiao jun Wang † August 25, 2000 Working Paper 00-03
∗
Department of Economics The Ohio State University Columbus, OH 43210; Tel: (614) 292-6429; E-mail:
[email protected] † Department of Economics The Ohio State University Columbus, OH 43210;Tel: (614) 292-4197; E-mail:
[email protected]
EALE/SOLE 2000 Belton M. Fleisher and Xiaojun WANG, “ Wages, Human Capital, and Work Incentives in a Transition Economy: The Case of Mainland China” Department of Economics, The Ohio State University, 1945 N. High Street; Columbus, OH 43210; USA
Abstract We study the wage structure between production workers (workers with low schooling)and technical/administrative staff (workers with high levels of schooling) in both urban and rural Chinese nonagricultural enterprises and find that employment-wage patterns cannot be explained under the paradigms of the profit-maximizing firm nor a simple interpretation of implications of the labor-managed firm. We find evidence that both urban and rural enterprises achieve productivity-enhancement from paying efficiency wages, but neither group of firms sets efficiency wages at profit-maximizing levels. Our empirical results imply overemployment of production workers relative to a profit-maximizing standard (under monopsony for rural enterprises) and underemployment and gross underpayment of workers with higher levels of schooling and skills. Our estimates suggest that maintaining high employment levels is relatively more important than maximizing profits for firms of all ownership classes. On the basis of our results, we believe that employment growth is limited by investable-funds constraints in the presence of evidently inflexible (downward) capital-labor ratios. Moreover, we estimate the social return to employing additional technical and administrative workers to be of the order of magnitude of 50%, while the private return to obtaining a college education is barely 5%. Theme: Labour markets in transition economies Keywords: Wages, Return to Schooling, Labor-Market Disequilibrium, Productivity, Transition Economies, Chinese Economy JEL-Codes: P23, J24, J31, D33, O15
1
Introduction
A number of puzzles emerge from research on wages and earnings in post-reform mainland China. One might have expected that economic reforms would weaken forces dictating wages in terms of political and/or social priorities, increase wage dispersion by level of formal schooling, and narrow geographical wage differentials. However, such changes have not been observed. Rather, private returns to schooling remain low (Jamison and Van Der Gaag, 1987, Dessi, 1991, Byron and Manaloto, 1990, Fleisher, Dong, and Liu, 1996; Gregory and Xin, 1995; Maurer-Fazio, 1997; Maurer-Fazio, Rawski, and Zhang, 1999; Psacharopoulos, 1985; Wang, Zhu, and Stromsdorfer,1995; Knight and Li, 1996; Li and Zhang, 1998; Zax, 1994; Yang and An, 1997) while geographical and sectoral wage dispersion have increased. (Chen and Fleisher, 1996 and references cited therein, Fleisher and Chen, 1997, Benjamin and Brandt, 1999; Khan, Griffin, and Riskin, 1999; Ravallion and Jalan, 1999; Yang, 1997a, 1997b, and 1999; Yang and Zhou, 1999). Moreover a number of researchers report a gap between estimated marginal products of labor and wages in nonagricultural enterprises, both collectives and state-owned. (Examples include Dong and Putterman, 1996, and references cited therein, particularly p. 601; Xu, 1991 and 1995; Dong and Putterman, 2000; Fleisher, Dong, and Liu, 1996; Parker, 1999; Pitt and Putterman, 1998; and Yang and Zhou, 1999.) We test models of wage determination based on alternative assumptions about the nature of competition and of firms’ objective functions. Svejnar (1990) and Pitt and Putterman (1999) have proposed a multi-objective utility function with arguments that include profit, employment, and wages, which nests models of labor-managed firms (Ward, 1958; Miyazaki, 1984; Bonin and Putterman, 1987), bargaining behavior (Kalai, 1977 and Roth, 1979), and pure profit maximization. In this paper, we add the possible role of efficiency-wage considerations in modeling firms’ wage-setting behavior. Finally, an additional complexity is introduced by consideration of monopsony, which may arise from
1
relatively immobile urban workers being tied to stated-owned enterprise (SOE)-provided housing, medical, food, and other benefits (Parker, 1999; Dong and Putterman, 2000) and from rural workers being located in small labor markets dominated by one or a very few rural enterprises (Dong and Putterman, 1996). In section 2, we first explore the possible role of efficiency-wage considerations in firms’ wage policies. Next, we compare wages paid to production and to technical, administrative, and staff workers (TAS) with their estimated marginal products. As in other studies mentioned above, we find a significant gap (usually negative) between wages and marginal product. The wage gap is substantially greater for TAS workers than for production workers. This differential gap “explains” in a proximate sense numerous observations of low returns to schooling in China. We explore alternative explanations of the estimated wage gaps, including (1)profit maximization under monopsony and (2) the possibility that profit-maximization is only one of firms’ objectives. Section 3 concludes and derives implications for economic reforms and the effects of institutional rigidities on economic growth and the distribution of wages and incomes.
2
Labor, Schooling, Production, and Wages
2.1
Do Chinese Enterprises Pay Efficiency Wages?
We postulate a simple efficiency-wage model in the spirit of Stiglitz (1976), Shapiro and Stiglitz (1984) and Krueger and Summers (1988) in which the excess of wages paid over an estimate of a spot-market competitive wage is included as a regressor in the following augmented Cobb-Douglas production. Y
Y = K α(
(ej Lj )βj )exp(φZ + ²)
j
where: Y
= Gross output or value added
2
(1)
K = Net Capital Stock Lj = Labor of the jth group ej = Effort function for the j th group of employees. Z = Vector of enterprise characteristics ² = an iid disturbance. The effort function is defined as: ej = Bj (
W j ηj Z ) Waj
(2)
Where B and η are parameters, Wj is the observed wage of the jth group of employees, and Waj the estimated spot-market competitive wage. This measure of the efficiency wage assumes that if the a worker loses his or her current job, the alternative wage would be the average predicted wage for his or her group. When data on schooling are available, it is natural to use a Mincerian human-capital wage equation to estimate Waj as the average wage for workers who have achieved a given level of schooling and labor-market experience. We have used such a data set, which we call the Urban Sample, to estimate the wage equation shown in Table 1.2 The Urban Sample provides data on Chinese production workers and technical/administrative workers employed in a cross-section of enterprises in diverse ownership categories.3 All right-hand variables are highly significant, with the exception of job tenure. Workers’ rates of return to schooling are relatively low, as is uniformly found in studies of the Chinese economy, as cited above. Workers who have completed high school (upper middle school) earn about eight percent more than workers with lower levels of schooling. Assuming that the average non-high school graduate has six years of schooling, this would amount to a private rate of return of approximately 1.2% per year of additional schooling for high-school graduates. The private rate of return for college graduates, similarly calculated, amounts to about to five percent per year, which, although higher than that for
3
high-school graduates, is still very low by international comparisons, especially compared to other nations at China’s level of economic development. The coefficients of age and experience have signs and magnitudes typical of other studies, and male workers receive on average about 14% higher earnings than do female workers. (See also Maurer-Fazio, Rawski, and Zhang, 1999.) In order to test the hypothesis that efficiency-wage considerations enter into firms’ wage policies, we estimate the production function defined in equation (1). The dependent variable is gross output, because data on intermediate inputs are not reported in the Urban Sample.
4
Econometric results for equation (1) are presented in table 2, where
Waj is taken to be the predicted wage for each worker from the estimated equation reported in Table 1. Column (1) shows estimated coefficients for the simplest specification of the production function, with both labor and the efficiency-wage variable pooled over both production and TAS workers. The estimated production elasticity of labor, 0.33, is estimated with considerable precision, and although low by comparison with estimates for most industrial economies, it is within the bounds of several studies of the Chinese economy (e.g., Chow, 1994; Dollar, 1990; Fleisher, Dong, and Liu, 1996). However, the null of constant returns to scale must be rejected, with p-value of no greater than 1% (not reported). A critical test of the joint hypothesis that enterprises are both profit-maximizing and pay efficiency wages is that the elasticity of work effort with respect to the efficiency wage is unity. (Solow, 1974 and Stiglitz, 1976). By simple substitution of equation (2) into equation (1), we see that the elasticity of effort with respect to the efficiency wage can easily be identified by dividing the coefficient of the variable M T by the coefficient of ln labor. If the effort-efficiency wage elasticity equals unity, then the coefficient of the efficiency-wage variable will equal that of the labor variable. This condition is clearly violated by the estimates reported in column (1). The calculated effort-efficiency wage
4
elasticity in column (1) is more than 3, implying that the impact of paying an efficiency wage on profit is not being fully exploited. The same calculations for the estimates reported in column (2) imply that there is grossly insufficient exploitation of the profitmaximizing potential of efficiency wages, more so for SOE’s than for other ownership forms. While it is to be expected that SOE’s are less profit-motivated than are other ownership forms, the other ownership classes apparently also fall far short of profit maximization with respect to the efficiency wage rate. In column (3) we find that the implied effort-efficiency wage elasticity for production workers is over 20 (although the production elasticity of production workers is VERY imprecisely estimated). The corresponding elasticity for technical/administrative workers is about 1.4, which although greater than the profit-maximizing value, is rather close. Further calculations, using column (4), indicate that the effort-efficiency wage elasticity for technical/administrative workers is closest to profit-maximizing in joint-venture enterprises, which is reassuring in terms of the likely degree of profit motivation in this ownership class (although estimated with considerable imprecision). Another perspective on efficiency wages is available. The Urban Sample includes answers to several questions about worker behavior. One of the questions asks workers how frequently they observe shirking among their coworkers. Table 3 relates reported observations of shirking to ownership type, location (coastal or noncoastal province)5, and the efficiency-wage variable as used to estimate equation (1). The dependent variable takes on a value of 1 for enterprises reporting below-average shirking, 0 otherwise. The estimated coefficients are all highly significant, except that the coefficient of the efficiency-wage variable has a p-value of only 0.23. The pseudo-R2 for the probit is approximately 0.5. The estimated coefficients of the ownership variables imply that shirking is most likely to be observed (or reported) among workers in SOE’s than in the other ownership forms. This conforms with expectations. Joint-venture enterprises
5
are on average 11% more likely to have below-average shirking than are SOE’s. The comparable statistic for collectives is 5%, and for firms located in the coastal provinces, 2%. An increase in the efficiency-wage variable of 1% is associated with a .02% decline in the probability that reported shirking is above average within ownership class groups, but this is a rather imprecise estimate, as indicated by the p-value of the corresponding coefficient in the probit equation. Another data set, which we call the Rural Sample, is a panel survey of 200 large rural enterprises (mostly TVE’s) for the years 1984 to 1990. The survey covers 20 enterprises in each of 10 provinces. The 10 provinces are: Anhui, Hubei, Guangdong, Jiangsu, Zhejiang, Sichuan, Hebei, Liaoning, Shanxi, and Gansu. Among them, Hebei, Guangdong, Liaoning, Jiangsu and Zhejiang are defined as coastal provinces, and the rest non-coastal provinces. The survey not only includes quantitative statistics about the individual firms, but also provides important environmental statistics describing the markets in which the firms operates. For example, not only is there data on an enterprise’s employment, but there is also information on total employment in the village where the firm is located. This allows us to test hypotheses on the impact of market structure on the behavior of the firm.6 Table 4 shows the results of estimating equation (1) using the Rural Sample. The estimated production elasticities for aggregate labor and capital reported in table 3 are close to the ordinary least-squares (OLS) results reported by Pitt and Putterman (1999) and to the GLS estimates reported by Dong and Putterman (1996) using the same data set. Compared to the results for the Urban Sample reported in table 2, there is evidence of unexploited scale economies. The principal difference in the production elasticities between the estimated production functions for the two data sets is that the labor elasticity estimated for the Rural Sample is over twice the magnitude as that reported for the Urban Sample. The relative magnitudes of the labor and capital elasticities are about the same as those reported by Pitt and
6
Putterman (1999). The efficiency-wage variable for the Rural Sample is, of necessity, constructed differently than for the Urban Sample. The Rural Sample does not provide measures of individual-worker levels of schooling. Therefore, we have defined the efficiency wage for this sample to be the deviation (in ratio form) of the worker’s actual wage to the average wage paid to that type of worker (production worker or technical and administrative staff worker) in the same province. The estimated coefficient of the efficiency-wage variable is highly significant, although smaller in magnitude than in the production functions estimated on the Urban Sample. Moreover, when the production function is estimated using two categories of workers, their respective efficiency-wage coefficients are approximately equal in magnitude and in statistical significance, which is in sharp contrast to the estimated coefficients for the Urban Sample. Also in contrast to the results for the Urban Sample, the ratio of the estimated efficiency-wage coefficients to their respective production-labor elasticities is much smaller than unity, and this is true for all workers taken together as well as for production and TAS workers separately. Taken at face value, these coefficients suggests that wage rates are set higher than their profit-maximizing level under an efficiency-wage scenario. The sharp contrast between the “too-low” wage rates estimated for the Urban Sample and the “too-high” wage rates estimated for the Rural Sample constitute a puzzle yet to be solved. The puzzle is made even more apparent when we note that the wage data in the Urban Sample are augmented to reflect housing payments in kind, whereas there is no information in the Rural Sample that permits us to make this adjustment. The panel nature of the Rural Sample permits us to explore the efficiency-wage hypothesis at greater depth. The big question, of course, is whether the estimated relationship between our efficiency-wage measures and production represents a genuine productivity-enhancing effect or whether it reflects reverse causation in the form of
7
bonus-sharing. (As suggested by Coady and Wang, 1998.) Table 5 reports the results of Granger Causality tests of the hypothesis that the efficiency-wage measure used in the estimates reported for the Rural Sample “cause” in the sense of “lead” productivity growth, or vice versa. We define total factor productivity (TFP) in the usual way, as the residual from a regression of log output on log capital and log employment, in the form of equation (1), with the efficiency-wage variable omitted. The estimates reported in columns (1) and (2) of table 5 show that the lagged efficiency-wage variable is a statistically significant “predictor” of TFP, whereas the estimates reported in columns (3) and (4) indicate that lagged TFP is not a statistically significant “predictor” of the efficiency-wage. Therefore, the hypothesis that paying an efficiency wage leads to higher productivity in the Rural Sample sample cannot be rejected.
2.2
Do Chinese Enterprises Exploit Monopsony in Their Employment and Wage-Setting Behavior?
Another view of the wage structure is obtained by comparing wages with the marginal product of labor. In table 6 (Urban Sample), the marginal product of aggregate labor is about 7.7 times average earnings including in-kind housing allocations. This wage gap is not only surprising for its magnitude, but also because it implies substantial underemployment of labor. Although this contradicts the conventional wisdom of overemployment in Chinese industry, it is consistent with empirical results in other studies. (Pitt and Putterman, 1999; Xu, 1995, Chapter 5) What is perhaps equally surprising, though, is that when employment is disaggregated into production workers and TAS workers, we observe that the wage gap for production workers is opposite in sign, with marginal product averaging about 80% of earnings. This means that the excess of marginal product over earnings is extremely large for TAS workers, with a magnitude in ratio terms of about 43. This large excess of MPL over earnings for TAS is consistent with results reported in Fleisher, Dong, and Liu (1996) for the Chinese paper industry.
8
The ratio of the marginal product of production labor to annual wage (inclusive of housing in kind) is 0.8 for SOE’s, 0.6 for collectives, and 0.9 for joint ventures. This is consistent with the hypothesis that employment decisions for production workers are politically driven with the motive of avoiding political disruption, especially for collectives, where dissatisfaction and political unrest would affect local employers directly and immediately. Even joint ventures are probably affected by this motive, through the need to obtain permission to enter Chinese markets and in order to achieve cooperation with Chinese business partners, especially local governments. One hypothesis that has been proposed to explain a positive MPL-wage gap in Chinese industry is monopsony power of Chinese nonagricultural employers. Dong and Putterman (1996) argue that monopsony power of rural employers in the presence of restrictions on inter-community migration can explain the wage gap in rural industry. Parker (1999) and Dong and Putterman (2000) both argue that Chinese State Industry (SOE’s) consciously exploited monopsony power in the pre-reform era. Parker, using micro data for the machine-tool industry in Nanjing provides evidence that increasing competition from collective- and privately owned firms has diminished SOEs’ monopsony power under reform. The monopsony explanation has a priori appeal, especially for urban SOE’s, given the well-known restrictions on rural-urban migration in China. The idea that SOE monopsony power is diminishing under reform is consistent with our finding a non-positive MPL-wage gap for production workers in the 1991 Urban Sample. It appears plausible to us that the supply of TAS workers is less elastic than that of production workers, because there is surely a smaller pool of skilled workers from which to attract new employees. It may well be that political pressures induce urban SOE’s to “overemploy” production workers, who are much more numerous than TAS workers, to minimize the risk of political unrest, and that this pressure is spread, through regulation and the need to acquire permits and licenses to operate, to urban collectives and joint
9
ventures. However, the immense magnitude of the wage gap for TAS workers in the urban sample leads us to question whether the observed wage-employment bundles we observe are due strictly to profit maximization under monopsony. The estimated gaps suggest an extremely small supply elasticity of TAS workers. 7 The hypothesis that the wage gaps we and others have estimated represent something like profit-maximization under monopsony can be tested in more than one way. An obvious direct test is to obtain estimates of labor-supply elasticities and compute from them profit-maximizing wage gaps. Estimates of simple labor-supply functions for the Rural Sample are reported in table 7 and a comparison of profit-maximizing and estimated wage gaps are shown in table 8. (The magnitudes of the wage gaps reported in table 8 are very close in magnitude to those reported by Xu (1995, p. 36), where the marginal product estimates are based on provincial aggregate data for TVE’s during about the same period as the data in our rural sample.)8 We assume that any exogenous shifts in communities’ labor-supply functions are captured by year dummy variables and their interaction with the wage regressor. Our estimation results with the complete set of interaction terms imply that there was no statistically significant trend in the labor-supply elasticities over time, but that there is a statistically significant difference between the labor-supply elasticity of TAS workers and that of production workers. Therefore only a single estimated elasticity for each class of workers is reported in table 7. The highly significant elasticities imply profit-maximizing wage-gap ratios of approximately 6.3 and 4.3 for production workers and TAS workers, respectively. A more detailed comparison between estimated profit-maximizing and observed wage gaps is shown in table 8. In almost all cases for production workers, the estimated profit-maximizing wage gap is larger than the observed wage gap, implying that enterprises place a positive value on employment in addition to profit. Thus, the conventional wisdom that overemployment is the rule in Chinese enterprises is supported, if one views the profit-maximizing
10
criterion under the monopsony hypothesis. On the other hand, the estimated profitmaximizing wage gap is uniformly lower than the observed wage gap for TAS, implying underemployment of this class of workers. This is consistent with the discussion of the econometric results for TAS reported in tables 2 and 3. Bearing in mind that the annual differences in the estimated profit-maximizing gaps are highly insignificant, it is interesting to note that the observed gaps tend to drift upward, implying perhaps a reduced tendency over time to “overemploy” production workers and an increased tendency to “underemploy” TAS workers. A less direct test of the monopsony hypothesis takes a more “instrumental variable” approach, regressing the wage gap on presumably exogenous correlates of monopsony power. Table 9 reports a test of the monopsony hypothesis for the Rural Sample. The mean wage gap for production and TAS workers, respectively, is regressed on the following variables: employer-concentration ratios and available land per worker in our sample of communities; provincial measures of foreign direct investment per worker, and unemployment.9 We hypothesize that under monopsony, the estimated coefficients of employer concentration and per-capita land will be positive and that of unemployment will be negative, because employer concentration is a direct measure of monopsony power, more land per person should increase agricultural labor productivity and reduce the elasticity of labor supply, while higher unemployment will increase the elasticity of labor supply. The foreign-investment variable is included to represent funds available for investment in physical capital. We include this variable to jointly test the alternative hypothesis that the wage gap is due to capital constraints limiting employment growth in the presence of available labor. We take a negative coefficient for the foreign-investment variable to be consistent with the hypothesis that capital constraints limit employment and wage growth. The estimation results reported in table 9 are notable mainly their very low ex-
11
planatory power. The unadjusted R2 are 0.06 and 0.03 for production workers and TAS, respectively. For production workers, the coefficient of the employer concentration ratio is negative and marginally significant, while the coefficient of per-capita land is statistically insignificant. Both of these results are inconsistent with the monopsony hypothesis. The coefficient of foreign direct investment is negative, although its p value is rather large, 0.19, and the coefficient of unemployment is positive and statistically significant. Therefore, we are inclined to explain the rural wage gap for production workers in terms of a highly elastic supply of unemployed production workers in the presence of capital constraints limiting employment growth. It remains to be explained, though, why capital-labor ratios don’t adjust downward to increase employment-output ratios and the number of workers employed with a given stock of capital. This puzzle is also addressed by Liu (1997). Further work on this question and on better ways to measure the supply of investable funds available to rural enterprises should be very productive in helping to understand more about the production-worker wage gap. The estimates reported in table 9 for TAS are less revealing than those for production workers. While the estimated coefficients of both employment concentration and available land per capita are positive, consistent with the monopsony hypothesis, the coefficient of the concentration ratio is hardly significant, and that of per capita land very insignificant. The estimated coefficient of foreign investment is negative, but very insignificant, and that of unemployment positive, but again insignificant.
2.3
Direct Estimation of Enterprise Goals
So far, we have found evidence that Chinese enterprises tend to overemploy production workers compared to a profit-maximizing standard, although the profit-maximizing standard must be viewed under the assumption of monopsony for rural enterprises. In
12
particular, firms have the following multiple-objective function: U = LωL (W/Wa )ωW Π1−ωL−ωW
(3)
where Wa is a measure of alternative wage. Firms maximize this objective function, subject to the production technology defined in the text (including the efficiency wage function). Profit Π is defined as: Π = Y − W L − rK
(4)
where r is the exogenous interest rate on capital. Firms chose L, W and K, so the first order conditions can be simplified as: ωL 1 − ωL − ωW dΠ + =0 L Π dL ω 1 − ωL − ωW dΠ W: W + =0 W Π dW dΠ K: =0 dK L:
(5) (6) (7)
Using the profit function and production function, we can easily obtain the following: dΠ βY = −W dL L dΠ βγY = −L dW W dΠ αY = −r dK K
(8) (9) (10)
Substitute them into the first order conditions, define sL = W L/Y and rearrange, we obtain: L:
1 − ωL − ωW α − 1 + sL = ωL β − sL
(11)
ωL ) 1 − ωW
(12)
or alternatively, sL = β + (1 − α − β)( 13
And W:
1 − ωL − ωW α − 1 + sL = ωW βγ − sL
(13)
or alternatively, sL = βγ + (1 − α − βγ)(
ωW ) 1 − ωL
(14)
The last four equations can be used to construct the weights for the objective function once the production function parameters are obtained. The constructed weights based on the estimation results for equations (12) and (14) are reported in table 10. Although the implied weights are not all positive, it is still instructive to compare them. The relative size of the weights implies that employment creation is highly valued relative to establishing higher wages or to earning profits. Moreover, there appears to be an upward trend in the difference between the magnitude of the employment weight and the weights for wages and profits, implying an increasing trend in the relative value of employment creation over the period 1994-1990. This is consistent with the views expressed by Svejnar (1990), although Pitt and Putterman (1998) conclude that enterprises in this sample assign rent-sharing a higher weight than employment creation, they do not evaluate employment levels against a profit-maximizing standard under monopsony as do Dong and Putterman (1996).
3
Conclusion and Outline for Further Work
We have attempted to provide a “broad-brush” sketch of major features of wage structure in Chinese business enterprises. The most outstanding feature of the landscape that emerges is one of immense disequilibria when viewed either from the perspective of the standard profit-maximization, competitive-market paradigm. Overemployment of production workers would appear to rule out a simple interpretation of observed behavior as explanable in terms of the pure labor-managed firm. Despite the complexities
14
and peplexities that emerge of analysis of production and wage characteristics, a few systematic tendencies in our empirical results can be enumerated. 1. Productivity is higher in firms that pay wage rates above the norm as measured by the average paid to all workers with similar characteristics. Evidence from panel data suggest that this correlation results at least in part from “efficiency-wage” effects rather than from profit or rent sharing. Evidence from cross-section data indicates that shirking is less likely in enterprises that pay efficiency wages. 2. Wage rates appear not to be set at profit-maximizing levels when judged in terms of efficiency wages. In the Urban Sample, efficiency wages appear to be lower than profit-maximizing, while in the Rural Sample, wages appear to be “too high.” There is weak evidence, however, that in the Urban Sample the productivityenhancing potential of efficiency wages is more fully exploited by joint ventures and collectives than by SOE’s and more among enterprises located in the coastal provinces than in the interior. 3. Wages for production workers exceed their marginal products in the Urban Sample but fall short of their marginal products in the Rural Sample. However, when the rural wage gap is compared with the criterion of profit-maximization under monopsony, rural firms are also seen to “overemploy” production workers relative to the profit-maximizing levels. 4. The joint hypothesis that Chinese firms are both profit-maximizing and monopsonists is rejected in that the marginal product-wage gap for production workers is negative in the urban data set and smaller than the profit-maximizing monopsony criterion in the rural data set. 5. The production-worker wage gap appears to have increased between 1984 and 1990, although, we cannot find statistically significant evidence of increased monopsony
15
power over this period. This could be taken as evidence that the wage gap in the rural data set is converging toward profit-maximization. However, an “instrumental variables” test of the monopsony hypothesis is rejected for the Rural Sample. 6. Consistent evidence of overemployment of production workers relative to a profitmaximizing level and evidence, albeit flawed, from the estimated weights of firms’ objective function, implies that greater weight is given to increasing employment of production workers than to maximizing profits. 7. We find (weak) evidence that investable-funds constraints limit employment growth among firms in the Rural Sample. 8. Wages for TAS workers are everywhere lower than their marginal products. This is very pronounced in the Urban Sample and less so in the Rural Sample. However, in the Rural Sample, even under the monopsony hypothesis, firms “underemploy” or “underpay” TAS workers relative to their profit-maximizing levels. We have not estimated a profit-maximizing wage gap for TAS workers in the Urban Sample, but the labor-supply elasticity implied by the estimated wage gap is absurdly small. 9. The low private rate of return to schooling found in our research and in numerous other studies of the Chinese labor force is “explained” in a proximate sense by this wage gap and reflects a similar large gap between the private and social returns to schooling in China. This rate of return gap has been noted elsewhere10and remains, to us, one of the major unresolved puzzles of the Chinese economy. We estimate the social return to educating and employing additional technical and administrative workers to be of the order of magnitude of 50%. This estimate is comparable to one obtained for college graduates using a national sample of provincial aggregate
16
data reported by Fleisher and Chen (1997). At the same time, the private return to obtaining a college education for the workers in our sample (and as reported in a number of other studies using different data) is around 5%. Explaining this “exploitation” of skilled workers in the presence of the implied gigantic profit potential of employing more of them is one of our greatest challenges. On a more positive note, our empirical results so far indicate not only the “bad news” of immense labor-market disequilibrium (relative to a profit-maximizati on criterion), but also the “good news” of tremendous potential continued economic growth from reallocation of resources toward more schooling and training of skilled workers. When these workers begin to be paid anywhere near what they seem to be worth, incentives to acquire further education will be greatly enhanced, and those with lower levels of schooling should perceive much greater incentives to advance themselves by remaining in school longer whenever economically feasible. Thus, all levels of society should benefit, although the short-term impact may be to widen inequality of living standards. Exploiting these growth opportunities should be one of the greatest challenges to Chinese policy makers.
17
Notes 1
The cited references are Svejnar (1990) Hay, et al. (1994), and Jefferson and Rawski (1994). Jefferson, Rawski, and Zheng (1992) report estimated nominal marginal products of labor in 1988 to be 2,974 and 1,648 yuan for enterprises and collectives (urban and TVE’s), respectively. The China Statistical Yearbook 1991 reports average annual wages of staff and workers in state-owned enterprises in 1988 to be 1,853 yuan and in urban collectives to be 1,426 yuan. 2
The value of the in-kind housing provided by employers is included in earnings.
3
The Urban Sample was collected in the second half of 1992 in a survey funded in part by the Ford Foundation in a grant to the Institute of Economics of the Chinese Academy of Social Sciences and the Labor Science Research Institute of the Ministry of Labor of the People’s Republic of China. We are grateful to Ernst Stromsdorfer, Elizabeth Li, and Jun Cao for making us aware of these data and for their help in using them. The survey is a stratified random sample of enterprises within a randomly selected sample of locales. All major regions of China are represented, but provincial capitals are not necessarily included. Of the major self-governing cities, only Guangdong is included in the sample. Two survey instruments were administered, one to an official of each enterprise, and another to a random sample of employees of the enterprise. 4We experimented with
alternate corrections for this defect, including using provinciallevel estimates of intermediate inputs. The resulting estimates are insensitive to these alternative specifications. 5
Coastal provinces are Hebei, Jiangsu, Shandong, Fujian, Guangdong, and Hainan. Non-coastal provinces are Shanxi, Jilin, Anhui, Hunan, Henan and Sichuan. 6
We are grateful to Dennis Yang, Yaohui Zhao, Xiao-yuan Dong, Isabelle Perrigne, and Gary Jefferson for their help in obtaining and using these data. 7After
all, the profit-maximizing wage gap would be 1 + 1/η, where η is the elasticity of supply. A wage gap of 43.6, the average for TAS workers in the Urban Sample, implies a supply elasticity of only 0.02. 8
The panel nature of the Rural Sample makes it easier to estimate labor-supply elasticities. Only cross-section estimates would be possible in the Urban Sample. 9
Foreign investment per worker is obtained from the Statistical Yearbook of China. Unemployment estimates are reported in Liu (1997) and are based on an estimate of available labor force minus the sum of workers required to operate family farms and nonagricultural employment.
18
10
See Fleisher, Dong, and Liu, 1996; Fleisher and Chen, 1997.
19
4
References • Benjamin, Dwayne and Loren Brandt, 1999, Markets and Inequality in Rural china: Parallels with the Past, American Economic Review 89, 292-295. • Bonin, John and Louis Putterman, 1987, Economics of Cooperation and the LaborManaged Economy, Fundamentals of Pure and Applied Economics, v. 14, London: Harwood Academic Publishers. • Byron, Raymond P. and Evelyn Q. Manaloto, 1990, Returns to Education in China, Economic Development and Cultural Change 38, 783-796. • Chen, Jian and Belton M. Fleisher 1996, Regional Income Inequality and Economic Growth in China,Journal of Comparative Economics 22, 141-164. • Chow, Gregory C., 1994, Understanding China’s Economy, World Scientific Publishing Company. • Coady, David P. and Limin Wang, 1998, Incentives, Allocation and Labour-Market Reforms During the Transition: The Case of Urban China, 1986-1990, Queen Mary and Westfield College, London, England. • Dollar, David, 1990, Economic Reform and Allocative Efficiency in China’s StateOwned Industry, Economic Development and Cultural Change 39, 89-106. • Dessi, Roberta, 1991, Income, Occupation, and Education in China, Department of Applied Economics University of Cambridge Working Paper No. 9106 • Dong, Xiao-Yuan and Louis Putterman, 2000, Pre-Reform Industry and State Monopsony in China, Journal of Comparative Economics, (forthcoming).
20
• Dong, Xiao-Yuan and Louis Putterman, 1996, China’s Rural Industry and Monopsony: An Exploration, Pacific Economic Review, Pacific Economic Review 1, 59-78. • Fleisher, Belton, Keyong Dong, and Yunhua Liu, 1996, Education, Enterprise Organization, and Productivity in the Chinese Paper Industry, Economic Development and Cultural Change 44, 571-87. • Fleisher, Belton M. and Jian Chen, 1997, The Coast-Noncoast Income Gap, Productivity, and Regional Economic Policy in China ), Journal of Comparative Economics 25, 220-36. • Gregory, R. G. and Xin Meng, 1995, Wage Determination and Occupational Attainment in the Rural Industrial Sector of China, Journal of Comparative Economics 21, 353-374. • Hay, D. A., D. J. Morris, G. Liu, and S. Yao, 1994, Economic Reform and Stateowned Enterprises in China 1979-1987, Oxford University Press. • Hellerstein, Judith, David Neumark, and Kenneth R. Troske, 2000, Wages, Productivity, and Worker Characteristics: Evidence from Plant-Level Production Functions and Wage Equations, Journal of Labor Economics 17, 409-446. • Huang, Tzu-Line, Arne Hallm, Peter F. Orazem, and Elizabeth M. Paterno, 1998, Empirical Tests of Efficiency Wage Models, Economica 65, 125-43. • Jamison, Dean T. and Jacques Van der Gaag, 1987, Education and Earnings in the people’s republic of China, Economics of Education Review 6, 161-166. • Jefferson, Gary, Thomas G. Rawski, and Yuxin Zheng, 1992, Growth, Efficiency, and Convergence in China’s State and Collective Industry, Economic Development and Cultural Change, 40, 239-266.
21
• Jefferson, G. H. and Thomas G. Rawski, 1994, Enterprise Reform in Chinese Industry, Journal of Economic Perspectives 8, 47-70. • Kalai, Ehud, 1977, Nonsymmetric Nash Solutions and Replications of Two-Person Bargaining, International Journal of Game Theory 6, 129-33. • Khan, Azizur Rahman, Keith Griffin, and Carl Riskin, Income Distribution in Urban China During the period of Economic Reform and Globalization, American Economic Review, 89, 296-300. • Knight, John and Li Shi, 1996, Educational Attainment and the Rural-Urban Divide in China, Oxford Bulletin of Economics and Statistics 58, 83-117. • Krueger, Alan B. and Lawrence H. Summers, 1988, Efficiency Wages and the Interindustry Wage Structure, Econometrica 56, 259-93. • Li, Tianyou and Junsen Zhang, 1998, Returns to Education under Collective and Household Farming in China, Journal of Development Economics 56, 307-335. • Liu, Yunhua, 1997, Labor Absorption in China’s Township and Village Enterprises, paper presented to the International Conference of Economics of Greater China, Perth, Australia. School of Accountancy and Business, Nanyang Technological University, Singapore. • Maurer-Fazio, Margaret, 1997, Earnings and Education in China’s Transition to a Market Economy: Evidence from 1989 and 1992, Department of Economics, Bates College. • Maurer-Fazio, Margaret, Thomas G. Rawski, and Wei Zhang, Inequality in the Rewards for Holding up Half the Sky: Gender Wage Gaps in China’s Urban Labour Market, 1988-94, The China Journal No. 41, 55-88.
22
• Miyazaki, Hajime, 1984, On Success and Dissolution of the Labor-managed Firm in the Capitalist Economy, Journal of Political Economy 92, 909-930. • Parker, Elliot, 1999, Are Wage increases in Chinese State Industry Efficient? Productivity in Nanjing’s Machine-Building Industry, Contemporary Economic Policy 17, 54-67. • Pitt, Mark M. and Louis Putterman, 1999, Employment and Wages in Township, Village, and Other Rural Enterprises, in G. Jefferson and I.J. Singh (ed.), Enterprise Reform in China: Ownership, Transition, and Performance, Oxford University Press, 197-215. • Psacharopoulos, George, 1985, Returns to Education: A Further International Update and Implications, Journal of Human Resources 20, 583-604. • Ravallion, Martin and Jyotsna Jalan, 1999, China’s Lagging Poor Areas, American Economic Review, 89, 301-305. • Roth, Alvin E., 1979, Axiomatic Models of Bargaining, Lecture Notes in Economics and Mathematical Systems 170, New York: Springer-Verlag. • Shapiro, Carl and Joseph E. Stiglitz, 1984, Equilibrium Unemployment as a Worker Discipline Device, American Economic Review 74, 433-44 • Solow, Robert M., 1979, Another Possible Source of Wage Stickiness, Journal of Macroeconomics 1, 79-82. • State Statistical Bureau, 1991, China Statistical Yearbook, China Statistical Information and Consultancy Center. • Svejnar, Jan, 1990, Productive Efficiency and Employment, in: William A. Byrd and Lin Qingsong (ed.), China’s Rural Industry: Structure, Development, and
23
Reform, published for the World Bank by Oxford University Press, 243-254. • Wang, Boqing, Jianfang Zhu, and Ernst Stromsdorfer, 1995, The Rate of Return to Education and Experience for Workers of State-owned Enterprises in the People’s Republic of China, paper presented at the Meetings of the Southern Economic Association New Orleans, November 18-20, 1995, Department of Economics Washington State University. • Ward, Benjamin, 1958, The firm in Illyria: Market Syndicalism, American Economic Review 48, 566-589. • Xu, Chenggang, 1991, Innovation, Productivity, and Labor Mobility in Socialist Economies, unpublished Ph.D. dissertation, Harvard University. • Xu, Chenggang, 1995, A Different Transition Path: Ownership, Performance, and Influence of Chinese Rural Industrial Enterprises. New York: Garland Publishing, Inc. • Yang, Dennis Tao, 1997a, Education and Off-Farm Work, Economic Development and Cultural Change 45, 613-632. • Yang, Dennis Tao, 1997b, Education in Production: Measuring Labor Quality and Management, American Journal of Agricultural Economics 79, 764-772 • Yang, Dennis Tao, 1999, Urban-Biased Policies and Rising Income Inequality in China, American Economic Review, 89, 306-310. • Yang, Dennis Tao and Mark Yuying An, 1997, Human Capital, Entrepreneurship, and Farm Household Earnings, Department of Economics Duke University. • Yang, Dennis Tao and Hao Zhou, 1999, Rural-Urban Disparity and Sectoral Labour Allocation in China, Journal of Development Studies, 35, 105-133.
24
• Zax, Jeffrey S., 1994, Human Capital in a Worker’s Paradise: Returns to Education in Urban China, Discussion Paper in Economics 94-7, Department of Economics, University of Colorado, Boulder.
25
5
Tables Table 1: Wage Equation : Urban Sample (Dependent Variable: ln W ) Variable Description CON ST . AGE AGE 2 SEX 1 if male EDUH 1 if a high school graduate EDUC 1 if at least collegegraduate EXP ER experience in year T E N U RE year with current employer No. of Obs.=731 6, Adj. R2 =0.20
Estimate 6.9486 0.0211 -0.0002 0.1433 0.0836 0.2273 0.0139 -0.0004
Std 0.0746 0.0042 0.0001 0.0107 0.0122 0.0147 0.0019 0.0009
P-Value 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.6201
Note: The wage variable, W , is adjusted for housing benefits. If the employee lives in an employerprovided house, an estimated amount of annual market rent for the same size house in the same province is added to the wage income. This estimated annual rent is based on the average annual rent per square meter paid by those who rent an apartment in the same province, multiplied by the size of the house.
26
Table 2: Production Function : Urban Sample (Dependent Variable: ln Y ∗) Variable CONST .
Description
K
ln net capital stock
L
ln total employment
PW
ln production workers
T AS
ln technical/administrative staff
MT
efficiency wage∗∗
MP W
efficiency wage∗∗ for production workers efficiency wage∗∗ for technical/administrative staff interaction between MT and collective dummy interaction between MT and joint venture dummy interaction between MT and coastal dummy interaction between MPW and collective dummy interaction between MPW and joint venture dummy interaction between MPW and coastal dummy interaction between MT AS and collective dummy interaction between MT AS and joint venture dummy interaction between MT AS and coastal dummy
MT AS MT 2 MT 3 MT C MP W 2 MP W 3 MP W C MT AS 2 MT AS 3 MT AS C No. of Obs. Adj. R2 F test statistics MT + MT 2 = 0
(1) 7.3411 (14.238) 0.4761 (11.494) 0.3324 (7.336) -
(2) 7.4902 (14.128) 0.4701 (11.228) 0.3283 (7.145) -
-
-
1.1497 (7.541) -
1.5506 (4.480) -
-
-
-
-
-0.2214 (-0.593) -0.3413 (-0.942) -0.2632 (-0.712) -
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
319 0.69
319 0.69
-
-
(3) 8.4293 (14.169) 0.3878 (8.117) -
(4) 8.5810 (13.736) 0.3801 (7.698) -
0.0260 (0.614) 0.467 (7.180) -
0.0233 (0.541) 0.469 (7.008) -
0.4454 (2.221) 0.6587 (2.973) -
0.2546 (0.795) 1.212 (2.533) -
-
-
-
-
-
262 0.73
0.1855 (0.371) 0.3556 (0.708) 0.1244 (0.296) -0.4312 (-0.675) -0.5027 (-0.921) -0.3974 (-0.809) 262 0.72
-
-
-
-
-
-
-
1.0473 (0.3071) 1.8875 (0.1707) 0.7937 (0.3739) 1.8969 (0.1697) 2.4485 (0.1189) 2.8881 (0.0909)
MT + MT 3 = 0
-
MT + MT C = 0
-
MP S + MPS 2 = 0
-
14.6022 (0.0002) 9.2998 (0.0025) 17.2664 (0.0001) -
MP S + MPS 3 = 0
-
-
-
MP S + MPS C = 0
-
-
-
MT AS + MT AS 2 = 0
-
-
-
MT AS + MT AS 3 = 0
-
-
-
MT AS + MT AS C = 0
-
-
-
Note: t-statistics are in parentheses for the top panel, and p-values for the bottom panel. ∗
Y is gross output. Efficiency wage is the residual if the regression reported in table 1.
∗∗
27
Table 3: Probit Model on Shirking : Urban Sample (Dependent Variable: 0 (shirking) or 1 (non-shirking)) Variable CON ST . Collectiv e J ointV enture Coastal EW
Estimate 0.507 0.153 0.331 0.074 0.051
Std 0.028 0.037 0.049 0.038 0.043
χ2 320.288 16.844 45.923 3.861 1.435
P-Value 0.000 0.000 0.000 0.049 0.231
No. of Obs. = 8238 Log Likelihood = -4575.76 6 Pseudo R2 =0.498 Note: EW is the wage differential at individual level. It is different from M T in that the latter is simply the average of all interviewed employees in the same firm.
28
Table 4: Production Function : Rural Sample (Dependent Variable: ln Y ∗) Variable CONST.
Description
K
ln net capital stock
L
ln total employment
PW
ln production workers
T AS
ln technical/administrative staff
MT
efficiency wage∗∗
MPW
Y R85
efficiency wage∗∗ for production workers efficiency wage∗∗ for technical/administrative staff interaction between MT and coastal dummy interaction between MPW and coastal dummy interaction between MT AS and coastal dummy year 1985 dummy
Y R86
year 1986 dummy
Y R88
year 1988 dummy
Y R89
year 1989 dummy
Y R90
year 1990 dummy
MT AS MT C MPW C MT AS C
No. of Obs. Adj. R2
(1) -1.3289 (-6.346) 0.4221 (13.728) 0.7376 (16.645) -
(2) -1.2780 (-6.141) 0.4257 (13.950) 0.7294 (16.576) -
(3) -0.7898 (-3.962) 0.3998 (12.677) -
(4) -0.7420 (-3.764) 0.4057 (13.007) -
0.4934 (9.888) 0.3024 (5.912) -
0.4851 (9.833) 0.2967 (5.870) -
-
-
0.3636 (6.213) -
0.6763 (7.068) -
0.2048 (3.168) 0.1724 (2.533) -
0.5020 (4.775) 0.3437 (3.388) -
-
-
-
-0.4803 (-4.111) -
-
-0.4136 (-3.150) -0.2632 (-1.977) 0.2557 (2.224) 0.0495 (0.431) 0.4822 (4.252) 0.4971 (4.311) 0.6175 (5.364) 979 0.61
-
-
-
0.2383 (2.059) 0.0288 (0.248) 0.4875 (4.262) 0.5178 (4.461) 0.6322 (5.479) 988 0.60
0.2205 (1.919) 0.0106 (0.092) 0.4707 (4.145) 0.4981 (4.322) 0.6053 (5.280) 988 0.61
0.2540 (2.184) 0.0545 (0.470) 0.4893 (4.266) 0.5081 (4.361) 0.6328 (5.434) 979 0.60
Note: t-statistics are in parentheses. 1987 observations are dropp ed due to inadequate data. So nominal number of observations should be 1200, and the discrepancies reflect missing values. ∗ Y is value added. ∗∗
Efficiency wage is the ratio of actual wage and average wage paid to the group of workers in that province.
29
Table 5: Granger Causality Tests : Rural Sample (Dependent Variable: ln T F P ∗f or(1), (2); ln EW ∗∗for(3), (4)) Variable C ON ST .
Description
T F P −1
ln TFP lagged 1 period
T F P −2
ln TFP lagged 2 period
E W− 1
ln EW lagged 1 period
E W− 2
ln EW lagged 2 period
No. of Obs. Adj. R2 F test statistics E W− 1 = 0, EW −2 = 0 T F P −1 = 0, T F P −2 = 0
(1) 0.1028 (3.374) 0.4777 (16.503) 0.2196 (3.548) 770 0.29
(2) 0.1261 (3.454) 0.3955 (9.302) 0.1191 (2.890) 0.1245 (1.380) 0.2085 (2.356) 577 0.27
(3) -0.0517 (-3.032) 0.0183 (1.159) 0.5964 (17.457) 795 0.29
(4) -0.030 (-1.497) 0.0122 (0.535) -0.0128 (-0.591) 0.4983 (10.172) 0.1788 (3.730) 595 0.30
-
8.3242 (0.0003) -
-
0.2114 (0.8095)
Note: t-statistics are in parentheses for the top panel, and p-value for the bottom panel. ∗ : ln T F P is the residual from the regression of log output on log capital and log employment. ∗∗: Efficiency wage is the ratio of actual wage and average wage paid to the same type of worker in the province.
30
Table 6: Comparison of Marginal Product of Labor and Average Wage: Urban Sample (unit: yuan) Worker Group T otal
P roduction W orker
T echnical/Administ rativeStaf f
Enterprise Character -
MPL 21,082
StateOwned
19,503
Collectiv e
14,853
J ointV enture
33,282
Coastal
25,183
N on − Coastal
17,333
-
2,008
StateOwned
2,008
Collectiv e
1,328
J ointV enture
2,644
Coastal
2,417
N on − Coastal
1,642
-
131,970
StateOwned
117,645
Collectiv e
103,041
J ointV enture
213,824
Coastal
156,368
N on − Coastal
109,517
Note: Sample standard deviation in parentheses.
31
Average Wage 2,729 (1,557) 2,633 (1,228) 2,463 (1,383) 3,362 (2,088) 3,297 (1,910) 2,268 (922) 2,508 (1,408) 2,495 (1,219) 2,260 (1,252) 2,955 (1,732) 2,986 (1,681) 2,123 (904) 2,946 (1,683) 2,762 (1,224) 2,677 (1,541) 3,699 (2,263) 3,595 (2,059) 2,386 (929 )
Table 7: Elasticity of Labor Supply: Rural Sample
P roduction W orkers T ech/mana Staff
Elasticity 0.19 (4.10) 0.30 (6.67)
Profit-Maximizing Gap 6.26 4.33
Note: t-statistic in parenthesis. Labor supply function is estimated by regressing log employment on log wage, year dummies, worker type dummies, and interaction terms between wage and type. On the basis of an F-test we cannot reject the hypothesis that labor-supplied elasticities are constant over time for both production and TAS workers. However, we can reject the hypothesis that the elasticity for production and TAS workers are equal with p-value equal to about 0.1.
32
Table 8: Observed and Profit-Maximizing Wage Gap: Rural Sample M P L/W AGE AllW orkers
P roduction W orkers
T ech/mana Staff
YEAR 1984 1985 1986 1988 1989 1990 1984 1985 1986 1988 1989 1990 1984 1985 1986 1988 1989 1990
Observed∗ 2.56 2.51 2.72 3.50 3.78 4.84 2.10 1.86 2.27 3.09 3.31 3.98 8.44 10.08 7.24 9.18 9.44 12.77
P rof it − M aximiz ing ∗∗ 5.75 7.25 3.04 3.86 3.08 11.00 5.76 8.14 3.44 4.13 4.57 11.00 6.88 4.13 4.03 4.13 4.13 6.55
*: Observed gap is the ratio of calculated marginal product of labor using the estimates from the production function and wage rates. **: Profit-Maximizing gap is one plus the inverse of the elasticity of labor supply. This is the ”hypothetical” gap if the enterprise is indeed a profit-maximizing monopsony.
33
Table 9: Monopsony Regression: Rural Sample Variable Const.
Description
CR
ln Concentration Ratio∗ ln acreage per person
MPC FDL LF UNEM
Foreign direct investment per labor force Unemployment rate
Number of observations Adjusted R2
Production Workers 0.3167 (2.130) -0.0904 (-1.771) 0.0366 (0.906) -0.003 (-1.305) 1.216 (2.180) 187 0.04
Tech./Admin. Staff 1.8017 (11.610) 0.0609 (1.144) 0.0281 (0.667) -0.002 (-0.794) 0.584 (1.003) 187 0.01
Note: The dependent variable is the log of the gap between marginal product of labor and wage rates. t-statistics in the parentheses. ∗: This is the employment share of this enterprise among all industrial enterprises in the township or
village.
34
Table 10: Constructed Weights of the Objective Function: Rural Sample Year 1984 1985 1986 1988 1989 1990
L
W Π Mean(std. dev.)/Me dian 1.36(8.20)/1.99 0.81(4.37)/0.47 -1.17(3.84)/-1.46 2.28(10.26)/1.99 0.32(5.46)/0.47 -1.60(4.80)/-1.46 1.16(19.89)/1.94 0.91(10.5 8)/0.50 -1.08(9.29)/-1.44 2.01(7.85)/1.97 0.46(4.18)/0.48 - 1.47(3.67)/-1.46 2.19(8.44)/1.96 0.37(4.49)/0.49 -1.56(3.95)/-1.45 1.91(14.12)/1.96 0.51(7.52)/0.49 -1.43(6.60)/-1.45
Note: Splitting the sample to coastal and non-coastal does not produce significant difference, so only the whole sample statistics are reported above.
35
Table 11: Sample Statistics For Employees: Urban Sample (N=9397) Variable Age Sex Experience T enure Education − − Hig hSchool − − C ollege Job Title − − P roduction W orker − − T echnical/Administ rativeStaf f Location − − C oastal − − N on − Coastal Income and Benefits − − W age − − Housing − − M edicalE xpenses
Description year dummy, 1 if male year year
Mean 34 0.55 15 10
Std 10 0.50 10 9
dummy, 1 if only high school graduate dummy, 1 if at least college graduate
0.44 0.22
0.50 0.41
dummy, 1 if production worker dummy, 1 if tech./adm. staff
0.28 0.48
0.45 0.50
dummy, 1 if coastal dummy, 1 if non-coastal
0.44 0.56
0.50 0.50
yuan/yr dummy, 1 if public house dummy, 1 if public medic care
2470 0.91 0.84
1229 0.29 0.36
36
Table 12: Sample Statistics For Enterprises: Urban Sample (N=422) Variable GrossOutput T otalE mploy ment N etC apital Ownership Classification − − State − − Collective − − JointV enture − − P rivate Location − − Coastal − − N on − Coastal
Unit mil.yuan person mil. yuan
Mean 70.455 1376 29.300
Std 225.408 3733 110.520
percent percent percent percent
0.40 0.34 0.25 0.01
0.49 0.47 0.43 0.11
percent percent
0.47 0.53
0.50 0.50
Note: Ownership Classification and Location are dummy variables. For example, the variable State has a mean of 0.40, which means that 40% of the surveyed firms are State-owned enterprises. Coastal provinces in the sample are: Hebei, Jiangsu, Shandong, Fujian, Guangdong, and Hainan. Noncoastal provinces in the sample are: Shanxi, Jilin, Anhui, Hunan, Henan and Sichuan.
37
Table 13: Sample Statistics For Enterprises: Rural Sample (N=200) UNIT 1984 1985 1986 1987 1988 1989 1990
Gross Output 10,000 yuan 336.139 (593.218) 486.576 (849.214) 532.570 (1025.870) 624.188 (1231.110) 631.756 (1018.180) 709.366 (1262.700) 757.204 (1317.700)
Net Capital 10,000 yuan 65.183 (88.725) 96.718 (132.202) 162.001 (289.581) 237.213∗ (509.598) 171.432 (277.068) 209.628 (358.718) 214.514 (353.842)
Note: standard deviation in the parentheses. ∗: original price of fixed capital.
38
Employment person 298.475 (309.381) 369.968 (354.787) 419.590 (472.879) 409.046 (521.383) 387.355 (439.702) 382.523 (438.833) 374.410 (481.779)
