Education, enterprise organization, and productivity in the Chinese paper industry

Economic Development and Cultural Change

Volumn 44, Issue 3, 1996, Pages 571-587

  Fleisher, Belton M ^a   Dong, Keyong ^b   Liu, Yunhua ^c  

^a OHIO STATE UNIVERSITY   (United States)

^b RENMIN UNIVERSITY OF CHINA   (China)

^c NANYANG TECHNOLOGICAL UNIVERSITY   (Singapore)

Author keywords

[No Author keywords available]

Indexed keywords

DEVELOPING COUNTRY; ENTERPRISE OWNERSHIP; FIRM OWNERSHIP; INDUSTRIAL PRODUCTIVITY; PRODUCTIVITY IMPACT; PROFIT DISTRIBUTION; PULP AND PAPER INDUSTRY; WORKERS' EDUCATION;

CHINA;

EID: 0030469108     PISSN: 00130079     EISSN: None     Source Type: Journal    
DOI: 10.1086/452232     Document Type: Article

References (24)

1. China Statistical Yearbook, 1991, English ed. (Beijing: State Statistical Bureau, 1991), p. 360.
2. Ibid., p. 26.
3. Two methods were used to deflate raw material values to be commensurate with real output values. Method 1 is the method used in deriving the estimates reported in table 3. It uses the Overall Retail Price Index for China published in China Statistical Yearbook, 1991 (English ed.), p. 199. Method 2 uses an index constructed by dividing the current-year value of output of enterprises in our sample by the constant-price value of output of the same enterprises. The constant-value output data were provided by the enterprises themselves and presumably reflect their own estimate of the base-year prices of their output. Fortunately, the two indexes are very close, although not identical, and our estimates are almost invariant with respect to which index is used.
4. 2, the hypothesis that the unrestricted model is better than the restricted model (i.e., that the slope coefficients of the factors' inputs are different in 1985, 1987, and 1989) is strongly rejected. See G. S. Maddala, Introduction to Econometrics (New York: Collier Macmillan, 1988), p. 429.
5. Ibid.
6. See M. Intriligator, Econometric Models, Techniques, and Applications (Englewood Cliffs, N.J.: Prentice-Hall, 1978), pp. 155-56.
7. Maddala, p. 429.
8. 2 distribution with degrees of freedom r. We cannot reject the null of homoscedasticity even at the 50% significance level.
9. See Maddala, pp. 417-18.
10. In an earlier version of this article, we reported production functions based on the Cobb-Douglas specification in which all production workers, managers, and professional employees were aggregated and the influence of educated personnel was represented by the proportion of educated managers and engineers and of educated production workers, respectively, of the aggregate of all workers. The econometric results for this specification were approximately the same as those reported here for educated managers and engineers. The estimated elasticity for the aggregate of all workers was higher and statistically more significant than that reported here for L(1). The estimated coefficient for the proportion of educated production workers was somewhat more significant than that reported in table 3 for L(2).
11. For a report on the constant returns to scale in the Canadian pulp and paper industry see J. C. Nautiyal and B. K. Singh, "Long-Term Productivity and Factor Demand in the Canadian Pulp and Paper Industry," Canadian Journal of Agricultural Economics 34 (August 1986): 35.
12. When the Cobb-Douglas function is estimated with only aggregated labor, the estimated coefficient of labor is 0.43 with a t-statistic of 4.0.
13. Gregory Chow, The Chinese Economy, 2d ed. (Singapore: World Scientific, 1989), pp. 122-23.
14. David Dollar, "Economic Reform and Allocative Efficiency in China's State-Owned Industry," Economic Development and Cultural Change 39 (1990): 89-106.
15. One of the referees has pointed out that our estimate of materials' elasticity (interpreted as a factor share) appears to be low, presumably due to controlled prices for inputs and inflated prices for output that do not match world relative prices. The referee argues, therefore, that our marginal product calculations for nonmaterial inputs may be biased upward (if the output prices we have used to aggregate output are inflated). We want to point out, however, that a proportional measurement error of a single input or of output will not result in a biased elasticity estimate, given the definition of elasticity as (∂y/ ∂x)(x/y). Obviously, a proportional change in either x or y will leave the resulting elasticity unaffected. Note, however, that a proportional error in measuring either x or y will result in an equal-proportion effect on the marginal product calculated from that elasticity. Therefore, the referee's caveat regarding a possible upward bias in our calculations of the marginal product of labor inputs should be borne in mind when drawing conclusions from the results of this study.
16. Jörg Becker, Paper Technology in the Third World (Braunschweig/Wiesbaden: Friedr. Vieweg & Sohn, 1988), p. 19.
17. Michael Mohr, A Quarterly Econometric Model of the Long-Term Structure of Production, Factor Demand, and Factor Productivity in 10 U.S. Manufacturing Industries, BLS Staff Paper 9, U.S. Technical Information Service no. PB-292-545 (Washington, D.C.: U.S. Bureau of Labor Statistics, 1978), p. 126.
18. Michel A. Amsalem, Technology Choice in Developing Countries: The Textile and Pulp and Paper Industries (Cambridge, Mass.: MIT Press, 1983), p. 121.
19. It is perhaps noteworthy that these figures lie on either side of the recently reevaluated per capita GDP in China, which is approximately $2,000 per year. Our estimated marginal products do not seem unreasonable in this light, considering that the substantial majority of Chinese labor is still found in agriculture, where lower incomes predominate. See Stephen Greenhouse, "New Tally of World's Economies Catapults China into Third Place," New York Times (May 20, 1993).
20. It is very important to emphasize that the comparison of the present value of the additional output attributable to one additional college-trained manager or engineer to the present cost of the forgone output involved in educating such personnel is not distorted by the possible upward bias in the estimated marginal product of nonmaterial inputs. Any systematic errors in deriving our marginal products will affect both the cost and benefit sides in equal proportion. Therefore, the ratio of costs and benefits is unaffected by proportional errors in estimating marginal products.
21. 0.1 = 1.0399.
22. Roberta Dessi, "Income, Occupation and Education in China," DAE Working Paper no. 9107 (Cambridge University, Department of Applied Economics, March 1991), p. 191.
23. Raymond P. Byron and Evelyn Q. Manaloto, "Returns to Education in China," Economic Development and Cultural Change 38 (1990): 783-96.
24. Boqing Wang, Jianfang Zhu, and Ernst Stromsdorfer, "The Rate of Return to Education and Experience for Workers of State-Owned Enterprises in the People's Republic of China" (Department of Economics, Washington State University, Pullman, June 1995).