Wage premiums for education and location of South African workers, by gender and race

Economic Development and Cultural Change

Volumn 48, Issue 2, 2000, Pages 307-334

  Mwabu, Germano ^a   Schultz, T Paul ^b  

^a UNIVERSITY OF NAIROBI   (Kenya)

^b YALE UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EDUCATIONAL ATTAINMENT; RACIAL DISPARITY; WAGE DETERMINATION; WAGE GAP;

SOUTH AFRICA;

EID: 0034079753     PISSN: 00130079     EISSN: None     Source Type: Journal    
DOI: 10.1086/452460     Document Type: Article

References (43)

1. L. Thompson, A History of South Africa (New Haven, Conn.: Yale University Press, 1990).
2. T. Paul Schultz and Germano Mwabu, "Labor Unions and the Distribution of Wages and Employment in South Africa," Industrial and Labor Relations Review 51, no. 4 (1998): 680-703.
3. T. Paul Schultz, "Human Capital and Economic Development," in Agricultural Competitiveness, ed. G. H. Peters et al., Twenty-Second International Conference of Agricultural Economists (Aldershot: Dartmouth, 1995).
4. Germano Mwabu and T. Paul Schultz, "Wage Premia for Education and Location, by Gender and Race in South Africa," Working (Discussion) Paper no. 785 (Yale University, Economic Growth Center, New Haven, Conn., 1998).
5. J. Mincer, Schooling, Experience and Earnings (New York: Columbia University Press, 1974); S. Rosen, "Human Capital: A Survey of Empirical Research," in Research in Labor Economics, ed. R. G. Ehrenberg, vol. 1 (Greenwich, Conn.: JAI Press, 1977); Z. Griliches, "Estimating the Returns to Schooling: Some Econometric Problems," Econometrica 45, no. 1 (January 1977): 1-22; Robert J. Willis, "Wage Determinants: A Survey and Reinterpretation of Human Capital Earnings Functions," in Handbook of Labor Economics, ed. O. Ashenfelter and R. Layard (Amsterdam: North Holland, 1986), 1:525-602 ; Schultz, "Human Capital."
6. J. Mincer, Schooling, Experience and Earnings (New York: Columbia University Press, 1974); S. Rosen, "Human Capital: A Survey of Empirical Research," in Research in Labor Economics, ed. R. G. Ehrenberg, vol. 1 (Greenwich, Conn.: JAI Press, 1977); Z. Griliches, "Estimating the Returns to Schooling: Some Econometric Problems," Econometrica 45, no. 1 (January 1977): 1-22; Robert J. Willis, "Wage Determinants: A Survey and Reinterpretation of Human Capital Earnings Functions," in Handbook of Labor Economics, ed. O. Ashenfelter and R. Layard (Amsterdam: North Holland, 1986), 1:525-602 ; Schultz, "Human Capital."
7. J. Mincer, Schooling, Experience and Earnings (New York: Columbia University Press, 1974); S. Rosen, "Human Capital: A Survey of Empirical Research," in Research in Labor Economics, ed. R. G. Ehrenberg, vol. 1 (Greenwich, Conn.: JAI Press, 1977); Z. Griliches, "Estimating the Returns to Schooling: Some Econometric Problems," Econometrica 45, no. 1 (January 1977): 1-22; Robert J. Willis, "Wage Determinants: A Survey and Reinterpretation of Human Capital Earnings Functions," in Handbook of Labor Economics, ed. O. Ashenfelter and R. Layard (Amsterdam: North Holland, 1986), 1:525-602 ; Schultz, "Human Capital."
8. J. Mincer, Schooling, Experience and Earnings (New York: Columbia University Press, 1974); S. Rosen, "Human Capital: A Survey of Empirical Research," in Research in Labor Economics, ed. R. G. Ehrenberg, vol. 1 (Greenwich, Conn.: JAI Press, 1977); Z. Griliches, "Estimating the Returns to Schooling: Some Econometric Problems," Econometrica 45, no. 1 (January 1977): 1-22; Robert J. Willis, "Wage Determinants: A Survey and Reinterpretation of Human Capital Earnings Functions," in Handbook of Labor Economics, ed. O. Ashenfelter and R. Layard (Amsterdam: North Holland, 1986), 1:525-602 ; Schultz, "Human Capital."
9. J. Mincer, Schooling, Experience and Earnings (New York: Columbia University Press, 1974); S. Rosen, "Human Capital: A Survey of Empirical Research," in Research in Labor Economics, ed. R. G. Ehrenberg, vol. 1 (Greenwich, Conn.: JAI Press, 1977); Z. Griliches, "Estimating the Returns to Schooling: Some Econometric Problems," Econometrica 45, no. 1 (January 1977): 1-22; Robert J. Willis, "Wage Determinants: A Survey and Reinterpretation of Human Capital Earnings Functions," in Handbook of Labor Economics, ed. O. Ashenfelter and R. Layard (Amsterdam: North Holland, 1986), 1:525-602 ; Schultz, "Human Capital."
10. The hourly wage is defined to include the value of bonuses, subsidies, and income in kind.
11. Mincer; Rosen.
12. F. Welch, "Human Capital Theory: Education, Discrimination and Life Cycles," American Economic Review 65, no. 2 (May 1975): 63-73; Griliches.
13. J. Heckman, "Sample Selection Bias as a Specification Error," Econometrica 47, no. 1 (January 1979): 153-61.
14. W. P. M. Vijverberg, "Educational Investments and Returns for Women and Men in Côte d'Ivoire," Journal of Human Resources 28, no. 4 (1993): 933-74.
15. The survey provides few ideal variables with which to determine the selection of the wage earner status. The exception is assets, some of which are potentially a reflection of occupational choice. We therefore repeated the selection correction based on other sets of wage identifiers (Mwabu and Schultz, "Wage Premia," table A4) and found decreased significance of the vector of identifiers and decreased statistical significance of an already marginal coefficient on lambda. Most of the land use rights and financial asset variables were associated with the individual being less likely to work for wages, as expected, but there were anomalous positive coefficients on irrigated land, salable crop land, and self-employment equipment for the Africans. Our identifiers for the colored and Indian samples, who are concentrated in urban areas, have even less predictive power than for the whites.
16. Mwabu and Schultz, "Wage Premia," table A2; cf. P. G. Moll, "The Decline of Discrimination against Colored People in South Africa, 1970-1980," Journal of Development Economics 37 (1992): 289-307, and "The Collapse of Primary School Returns in South Africa, 1960-1990" (Mt. Prospect, III., unpublished manuscript, 1995). We assumed it required 2 years of full-time study to obtain a teaching or nursing certificate, 3.5 years on average for a technikon diploma, 4 years for a university degree, and 1 year for a diploma from a junior or vocational college. This final credential or qualification category is probably the most diverse and is mostly obtained by Africans, presumably before they were admitted in greater numbers to universities. If individuals reported going to a university but not receiving a degree, they are attributed 2 years of higher education. To illustrate how the estimates are to be interpreted, assume an African woman who receives a nursing certificate; the coefficient for African women on this dummy variable is -.67, the coefficient for years of higher education is .71, and the expected value of her log wage is .75 = (-.67 + 2*.71), larger than for an African woman who only completed secondary school. By contrast, if the African woman had completed teacher training, her wage would only be 1.06 = (-.36 + 2*.71), larger for the same 2 years of training (coefficients are from table 4).
17. Mwabu and Schultz, "Wage Premia," table A2; cf. P. G. Moll, "The Decline of Discrimination against Colored People in South Africa, 1970-1980," Journal of Development Economics 37 (1992): 289-307, and "The Collapse of Primary School Returns in South Africa, 1960-1990" (Mt. Prospect, III., unpublished manuscript, 1995). We assumed it required 2 years of full-time study to obtain a teaching or nursing certificate, 3.5 years on average for a technikon diploma, 4 years for a university degree, and 1 year for a diploma from a junior or vocational college. This final credential or qualification category is probably the most diverse and is mostly obtained by Africans, presumably before they were admitted in greater numbers to universities. If individuals reported going to a university but not receiving a degree, they are attributed 2 years of higher education. To illustrate how the estimates are to be interpreted, assume an African woman who receives a nursing certificate; the coefficient for African women on this dummy variable is -.67, the coefficient for years of higher education is .71, and the expected value of her log wage is .75 = (-.67 + 2*.71), larger than for an African woman who only completed secondary school. By contrast, if the African woman had completed teacher training, her wage would only be 1.06 = (-.36 + 2*.71), larger for the same 2 years of training (coefficients are from table 4).
18. Mwabu and Schultz, "Wage Premia," table A2; cf. P. G. Moll, "The Decline of Discrimination against Colored People in South Africa, 1970-1980," Journal of Development Economics 37 (1992): 289-307, and "The Collapse of Primary School Returns in South Africa, 1960-1990" (Mt. Prospect, III., unpublished manuscript, 1995). We assumed it required 2 years of full-time study to obtain a teaching or nursing certificate, 3.5 years on average for a technikon diploma, 4 years for a university degree, and 1 year for a diploma from a junior or vocational college. This final credential or qualification category is probably the most diverse and is mostly obtained by Africans, presumably before they were admitted in greater numbers to universities. If individuals reported going to a university but not receiving a degree, they are attributed 2 years of higher education. To illustrate how the estimates are to be interpreted, assume an African woman who receives a nursing certificate; the coefficient for African women on this dummy variable is -.67, the coefficient for years of higher education is .71, and the expected value of her log wage is .75 = (-.67 + 2*.71), larger than for an African woman who only completed secondary school. By contrast, if the African woman had completed teacher training, her wage would only be 1.06 = (-.36 + 2*.71), larger for the same 2 years of training (coefficients are from table 4).
19. See M. Ainsworth and J. Munoz, "The Côte d'Ivoire Living Standards Survey," Working Paper no. 26 (World Bank, Living Standards Measurement Study, Washington, D.C., 1986); and M. Grosh and P. Glewwe, "A Guide to Living Standards Measurement Study Surveys and Their Data Sets," Working Paper no. 120 (World Bank, Living Standards Measurement Study, Washington, D.C., 1995) for a description of these surveys.
20. See M. Ainsworth and J. Munoz, "The Côte d'Ivoire Living Standards Survey," Working Paper no. 26 (World Bank, Living Standards Measurement Study, Washington, D.C., 1986); and M. Grosh and P. Glewwe, "A Guide to Living Standards Measurement Study Surveys and Their Data Sets," Working Paper no. 120 (World Bank, Living Standards Measurement Study, Washington, D.C., 1995) for a description of these surveys.
21. SALDRU, South Africans Rich and the Poor: Baseline Household Statistics (University of Cape Town, School of Economics: Cape Town, South Africa, 1994). A random sample of population clusters was first drawn from a countrywide listing of census districts, from which, in turn, the required number of households was picked: 360 clusters were chosen with an expectation of picking 25 households per cluster so as to yield a total of 9,000 households and some 45,000 individuals. The actual number of households interviewed was 8,848, while the number of individuals enumerated was 43,974, or approximately a 1-in-1,000 self-weighting clustered sample of the South African population, which is estimated to have then been 40.1 million.
22. Ibid.
23. Mwabu and Schultz, "Wage Premia" (n. 4 above), tables A3 and A4. The Heckman sample selection corrected estimates are based on the working assumption that variables representing an individual's ownership of land and assets enter the probit function, determining who is employed in a wage job, and the wage function is then fit, conditional on a transformation of the probit estimates (inverse of the Mills ratio). Implicit in this model is the parametric assumption that the error in the wage-employment equation is normally distributed. Although the land and asset variables explain a significant fraction of the variation in wage-employment outcomes, the estimated coefficient on lambda (the inverse of the Mills ratio) in the wage function is not statistically significantly different from zero in seven of the eight race and gender subsamples described in table 1. Only for African males is there statistical evidence that the selection into wage employed is related to the unexplained wage residual (the coefficient on lambda is .266, t = 2.99). In other words, African men with unexplained higher wage opportunities are more likely to be employed in a wage job. The selection is also evident only in the rural and not the urban African male subpopulations, justifying the disaggregation of rural and urban wage estimates as reported in table 2. In the case of African men, the Heckman estimates .of educational returns in the wage function are slightly higher than the OLS estimates for secondary and higher schooling and marginally lower for primary schooling. Given our limited empirical basis for identifying the sample selection bias, and the limited effect of correcting this bias on the wage returns to schooling, our discussion here focuses on the OLS estimates in tables 2, 3, and 4. Additional research should be directed toward developing a better basis for correcting for potential sample selection bias among African males, but we do not expect that it would substantially change our conclusions.
24. See Mwabu and Schultz, "Wage Premia," table A2, for sources and coding conventions used to construct all variables.
25. The various categories of higher education reported by individuals in the survey may have involved different numbers of years of study for different individuals at different institutions and at different times. We consulted a variety of sources to obtain our estimates reported in Mwabu and Schultz, "Wage Premia," table A2, and summarized in the text. Any systematic error in our estimates of years required for each type of higher education category could then bias the degree category coefficients reported in table 4. For example, suppose that university degrees required on average 4 years of study, and we assumed standard 10 plus only 3 years. Then the university coefficient in table 4 would be biased downward, or about three-fourths its correct magnitude.
26. R. L. Oaxaca, "Male-Female Wage Differentials in Urban Labor Markets," International Economic Review 14, no. 3 (1973): 693-709; for derivation see Mwabu and Schultz, "Wage Premia," appendix B.
27. R. L. Oaxaca, "Male-Female Wage Differentials in Urban Labor Markets," International Economic Review 14, no. 3 (1973): 693-709; for derivation see Mwabu and Schultz, "Wage Premia," appendix B.
28. The Heckman selection corrected estimates are reported in Mwabu and Schultz, "Wage Premia." Because they are statistically significant only for African males and often unstable for other subsamples, they are not regarded as reliable as the OLS estimates. The decompositions differ using the Heckman model estimates for African and colored men.
29. The difference between the actual mean log wage of wage earners (table 5, row 2) and the mean log wage predicted for all persons ages 16-65 based on the OLS estimates of the wage equations (table 2) that neglect sample selection bias (table 5, row 3) is -.19 for African men, -.34 for African women, -.10 for colored men, -.16 for colored women, -.17 for Indian men, -.18 for Indian women, -.17 for white men, and -.077 for white women.
30. Our estimates of the selection corrected wage function, with their noted limitations, imply that the sample selection bias reduces the average log wage of all persons by -.34 compared with that predicted on observables for African men. However, for the other race and gender groups the implied effect of unobservables is positive and smaller (Mwabu and Schultz, "Wage Premia").
31. G. S. Becker, Human Capital, 2d ed. (New York: Columbia University Press, 1975).
32. It is assumed here that any abilities that are complementary to the returns from education are uniformly distributed across the different groups, but this assumption is relaxed elsewhere (Germano Mwabu and T. Paul Schultz, "Education Returns across Quantiles of the Wage Function," American Economic Review 86, no. 2 [1996]): 335-39. The other commonly considered hypothesis for group differences in educational investment is credit constraints, which could plausibly increase the cost of borrowing for Africans compared with white groups, and thus reduce the incentive for Africans to invest in schooling as much as whites do. National examinations are set for different racial groups to determine which primary school graduates may advance to secondary school and which secondary school graduates may enroll at an institution of higher education. These exams ultimately function as a race-specific rationing device, although credit constraints could reinforce this restriction mechanism. Estimated graduation rates by race, gender, and age are provided in the next to last column in table A5 in Mwabu and Schultz, "Wage Premia."
33. The OLS regression estimates of the rate of return to the three levels of schooling, along with their standard errors, and the proportion of individuals with that level of schooling, of each race and sex group, are reported in Mwabu and Schultz, "Wage Premia," appendix table A5. The regressions in table 6 are weighted by the inverse of the regression coefficient standard errors. The cell for the oldest age group of higher educated Indians is empty, and hence only 23 observations are available for higher education coefficients and 71 in the overall regressions.
34. P. G. Moll, "Quality of Education and the Rise in Returns to Schooling in South Africa, 1975-1985," Economics of Education Review 11, no. 1 (1992b): 1-10, and "The Collapse of Primary School Returns" (n. 12 above).
35. P. G. Moll, "Quality of Education and the Rise in Returns to Schooling in South Africa, 1975-1985," Economics of Education Review 11, no. 1 (1992b): 1-10, and "The Collapse of Primary School Returns" (n. 12 above).
36. Moll, "The Decline of Discrimination" (n. 12 above), and "Quality of Education."
37. Vijverberg (n. 10 above).
38. According to the survey, however, only 1% of the African male wage earners ages 16-65 with less than a secondary education are bom in another country.
39. Moll, "Quality of Education," and "The Collapse of Primary School Returns," previously estimated returns to schooling for Africans in the 1970s and 1980s and found them to be lower than those we report for 1993. Because Moll's data are from urban manufacturing surveys, we restricted our sample to these groups but continued to obtain estimates of wage returns larger than those he reports. Perhaps different sampling frames were used in the two surveys, or a radical change in wage structures occurred for at least urban Africans in the early 1990s.
40. Elsewhere we estimate quantile wage regressions to evaluate whether estimates of the wage returns to education decrease from the top to the bottom of the distribution of wage residuals for various race and sex groups in South Africa (Mwabu and Schultz, "Education Returns"). Only among the whites is there evidence that the returns to higher education decrease markedly among the lower deciles of the distribution of wage residuals. This pattern suggests that reducing the proportion of whites who attend higher education by raising the admission standards might increase the average wage return to those matriculating. For Africans the data do not confirm that a similar increase in returns would be achieved by retrenchment in secondary and higher education, which suggests that expansion of education for Africans will not lead to an observable deterioration in educational quality or even to a noticeable decline in the additional marketable skills that these levels of education provide for African workers.
41. Moll, "The Collapse of Primary School Returns."
42. T. Paul Schultz, "Education Investment and Returns," in Handbook of Development Economics, ed. H. Chenery and T. N. Srinivasan, vol. I (Amsterdam: North-Holland, 1988).
43. Moll, "Quality of Education."