The effect of food subsidies on labor supply in Sri Lanka

Economic Development and Cultural Change

Volumn 45, Issue 1, 1996, Pages 124-145

  Sahn, David E ^a   Alderman, Harold ^b  

^a Department of Molecular Biology and Genetics

^b WORLD BANK   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

DEVELOPING COUNTRY; FOOD ASSISTANCE PROGRAMME; INCENTIVES; LABOUR SUPPLY; POLICY IMPLICATION; RICE RATION; TRANSFER PROGRAMME; WORK DISINCENTIVE;

SRI LANKA;

EID: 0030429352     PISSN: 00130079     EISSN: None     Source Type: Journal    
DOI: 10.1086/452260     Document Type: Article

References (30)

1. Ravi Kanbur, Michael Keen, and Matti Tuomala, "Labor Supply and Targeting in Poverty Programs," in Public Spending and the Poor: Theory and Evidence, ed. D. van de Walle and K. Nead (Baltimore: Johns Hopkins University Press, 1995).
2. Robert Moffitt, "Incentive Effects of the U.S. Welfare System: A Review," Journal of Economic Literature 30, no. 1 (1992): 1-61.
3. Sheldon Danziger, Robert Haveman, and Robert Plotnick, "How Income Transfers Affect Work, Savings, and the Income Distribution: A Critical Review," Journal of Economic Literature 19, no. 3 (1981): 975-1028; Thomas Fraker and Robert Moffitt, "The Effect of Food Stamps on Labor Supply: A Bivariate Selection Model," Journal of Public Economics 35, no. 1 (1988): 25-56; Anne Winkler, "The Incentive Effects of Medicaid on Women's Labor Supply," Journal of Human Resources 26, no. 2 (1991): 308-37.
4. Sheldon Danziger, Robert Haveman, and Robert Plotnick, "How Income Transfers Affect Work, Savings, and the Income Distribution: A Critical Review," Journal of Economic Literature 19, no. 3 (1981): 975-1028; Thomas Fraker and Robert Moffitt, "The Effect of Food Stamps on Labor Supply: A Bivariate Selection Model," Journal of Public Economics 35, no. 1 (1988): 25-56; Anne Winkler, "The Incentive Effects of Medicaid on Women's Labor Supply," Journal of Human Resources 26, no. 2 (1991): 308-37.
5. Sheldon Danziger, Robert Haveman, and Robert Plotnick, "How Income Transfers Affect Work, Savings, and the Income Distribution: A Critical Review," Journal of Economic Literature 19, no. 3 (1981): 975-1028; Thomas Fraker and Robert Moffitt, "The Effect of Food Stamps on Labor Supply: A Bivariate Selection Model," Journal of Public Economics 35, no. 1 (1988): 25-56; Anne Winkler, "The Incentive Effects of Medicaid on Women's Labor Supply," Journal of Human Resources 26, no. 2 (1991): 308-37.
6. David E. Sahn and Neville Edirisinghe, "The Politics of Food Policy in Sri Lanka: From Basic Human Needs to Increased Market Orientation," in The Political Economy of Food and Nutrition Policies, ed. Per Pinstrup-Andersen (Baltimore: Johns Hopkins University Press, 1993); Neville Edirisinghe, "The Food Stamp Program in Sri Lanka: Costs, Benefits, and Policy Implications," International Food Policy Research Institute (IFPRI) Research Report no. 58 (IFPRI, Washington, D.C., 1987); James Gavan and Indrani Chandrasekera, "The Impact of Public Foodgrain Distribution on Food Consumption and Welfare in Sri Lanka," IFPRI Research Report no. 13 (IFPRI, Washington, D.C., 1979).
7. David E. Sahn and Neville Edirisinghe, "The Politics of Food Policy in Sri Lanka: From Basic Human Needs to Increased Market Orientation," in The Political Economy of Food and Nutrition Policies, ed. Per Pinstrup-Andersen (Baltimore: Johns Hopkins University Press, 1993); Neville Edirisinghe, "The Food Stamp Program in Sri Lanka: Costs, Benefits, and Policy Implications," International Food Policy Research Institute (IFPRI) Research Report no. 58 (IFPRI, Washington, D.C., 1987); James Gavan and Indrani Chandrasekera, "The Impact of Public Foodgrain Distribution on Food Consumption and Welfare in Sri Lanka," IFPRI Research Report no. 13 (IFPRI, Washington, D.C., 1979).
8. David E. Sahn and Neville Edirisinghe, "The Politics of Food Policy in Sri Lanka: From Basic Human Needs to Increased Market Orientation," in The Political Economy of Food and Nutrition Policies, ed. Per Pinstrup-Andersen (Baltimore: Johns Hopkins University Press, 1993); Neville Edirisinghe, "The Food Stamp Program in Sri Lanka: Costs, Benefits, and Policy Implications," International Food Policy Research Institute (IFPRI) Research Report no. 58 (IFPRI, Washington, D.C., 1987); James Gavan and Indrani Chandrasekera, "The Impact of Public Foodgrain Distribution on Food Consumption and Welfare in Sri Lanka," IFPRI Research Report no. 13 (IFPRI, Washington, D.C., 1979).
9. At the time that the income declaration forms were filled out, the Rs 300 cutoff point had not been announced.
10. Edirisinghe indicates that one of the reasons for the rural bias and the regional differences in the incidence of receiving the transfer was the difficulty in valuing agricultural related and seasonal income in monetary terms.
11. Variation in the amount received was also a result of actual degree of utilization of the ration shops, given that a portion of the ration needed to be purchased, albeit at a subsidized value.
12. Edirisinghe; Sahn and Edirisinghe.
13. Robert Moffitt, "An Economic Model of Welfare Stigma," American Economic Review 73, no. 5 (1983): 1023-35; Christine Ranney and John Cushman, "Cash Equivalent, Welfare Stigma, and Food Stamps," Southern Economic Journal 13 (April 1985): 1011-27.
14. Robert Moffitt, "An Economic Model of Welfare Stigma," American Economic Review 73, no. 5 (1983): 1023-35; Christine Ranney and John Cushman, "Cash Equivalent, Welfare Stigma, and Food Stamps," Southern Economic Journal 13 (April 1985): 1011-27.
15. Edirisinghe.
16. The Mill's ratio is the sample selection correcting variable derived from the participation equation. The selection approach is based on J. J. Heckman, "Shadow Prices, Market Wages and Labor Supply," Econometrica 41, no. 4 (1974): 679-94. More recent work, such as James Heckman, "Variety of Selection Bias," American Economic Review 80, no. 3 (1990): 313-18; and W. Newey, J. Powell, and J. Walker, "Semiparametric Estimation of Selection Models: Some Empirical Results," American Economic Review 80, no. 3 (1990): 324-28, suggests that often identification cannot be achieved. These studies also claim that often the assumptions about error structure, which make selection important, do not prevail. Nevertheless, it is still worthwhile to decompose total labor force participation into probability and conditional response as often the magnitude of the parameters differ.
17. The Mill's ratio is the sample selection correcting variable derived from the participation equation. The selection approach is based on J. J. Heckman, "Shadow Prices, Market Wages and Labor Supply," Econometrica 41, no. 4 (1974): 679-94. More recent work, such as James Heckman, "Variety of Selection Bias," American Economic Review 80, no. 3 (1990): 313-18; and W. Newey, J. Powell, and J. Walker, "Semiparametric Estimation of Selection Models: Some Empirical Results," American Economic Review 80, no. 3 (1990): 324-28, suggests that often identification cannot be achieved. These studies also claim that often the assumptions about error structure, which make selection important, do not prevail. Nevertheless, it is still worthwhile to decompose total labor force participation into probability and conditional response as often the magnitude of the parameters differ.
18. The Mill's ratio is the sample selection correcting variable derived from the participation equation. The selection approach is based on J. J. Heckman, "Shadow Prices, Market Wages and Labor Supply," Econometrica 41, no. 4 (1974): 679-94. More recent work, such as James Heckman, "Variety of Selection Bias," American Economic Review 80, no. 3 (1990): 313-18; and W. Newey, J. Powell, and J. Walker, "Semiparametric Estimation of Selection Models: Some Empirical Results," American Economic Review 80, no. 3 (1990): 324-28, suggests that often identification cannot be achieved. These studies also claim that often the assumptions about error structure, which make selection important, do not prevail. Nevertheless, it is still worthwhile to decompose total labor force participation into probability and conditional response as often the magnitude of the parameters differ.
19. The human capital variables that determine wages include education and job experience (proxied by age minus years of schooling). In addition, landholding variables as well as a series of seasonal dummy variables are included. See David E. Sahn and Harold Alderman, "The Effects of Human Capital on Wages, and the Determinants of Labor Supply in a Developing Country," Journal of Development Economics 29, no. 2 (1988): 157-83, for a more detailed discussion of the wage equation. The standard errors have not been corrected for the use of predicted rather than observed wages. Lung-Fei Lee, "Unionism and Wage Rates: A Simultaneous Equation Model with Qualitative and Limited Dependent Variables," International Economic Review 19 (1978): 415-33, indicates that there are some additional problems in computing the standard errors when the predicted variables come from a truncated model, as with the wages here. This concern is a limitation of the estimates, but peripheral to the main theme.
20. The human capital variables that determine wages include education and job experience (proxied by age minus years of schooling). In addition, landholding variables as well as a series of seasonal dummy variables are included. See David E. Sahn and Harold Alderman, "The Effects of Human Capital on Wages, and the Determinants of Labor Supply in a Developing Country," Journal of Development Economics 29, no. 2 (1988): 157-83, for a more detailed discussion of the wage equation. The standard errors have not been corrected for the use of predicted rather than observed wages. Lung-Fei Lee, "Unionism and Wage Rates: A Simultaneous Equation Model with Qualitative and Limited Dependent Variables," International Economic Review 19 (1978): 415-33, indicates that there are some additional problems in computing the standard errors when the predicted variables come from a truncated model, as with the wages here. This concern is a limitation of the estimates, but peripheral to the main theme.
21. As is often the case, a variety of home-production activities such as cooking and fetching fuel are not included in the data.
22. There is a potential sample truncation issue that enters the analysis insofar as the small share of households without two persons between the ages of 15 and 65 is left out of the analysis.
23. Elsewhere, we compare results of treating the subsidy as an exogenous variable versus using the endogenized value, as we do in this article. The direction of bias was consistent with expectations; unobserved factors that increase the probability of subsidy participation correlate with labor supply leading to a smaller subsidy impact on labor supply. See David E. Sahn and Harold Alderman, "Incentive Effects on Labor Supply of Sri Lanka's Food Subsidies," in van de Walle and Nead, eds. (n. 1 above).
24. The inclusion of occupation in the instrumenting equation was felt to be justified since it is largely a lagged endogenous variable that predates the implementation of the targeted subsidy.
25. A quadratic term for the subsidy variable was negative, although not significant, and hence not included in the estimates shown.
26. An alternative means of estimating the above labor participation and supply equations is by means of a Tobit equation that treats entry and supply in a single estimation. Such an estimate has some conceptual advantages when the same variables appear in the participation and supply equations, as is the case here, since they avoid the need for identification by means of functional form. In such cases one can test the restriction that the parameters that determine participation are the same as those in the labor-supply equation up to a scaling term (see William Green, Econometric Analysis [New York: Macmillan, 1990], p. 736). We have done so and find that for all estimates, female and male, urban and rural, the restrictions are rejected at probability less than 0.01. Nevertheless, we report the Tobit estimators in appendix table A3. Since the Tobit does not have the flexibility of the two-equation approach used in the main text, some of the parameters in appendix table A3 mask the magnitude of the supply effects in table 3.
27. Moffitt, "Incentive Effects" (n. 2 above).
28. Harold Alderman and David E. Sahn, "Substitution between Goods and Leisure in a Developing Country," American Journal of Agricultural Economics 75, no. 4 (1993): 875-83, estimated a complete demand system, including leisure as a commodity. As with the current study, the marginal propensity to consume leisure was found to be high.
29. Moreover, the results here are inconsistent with a reverse causality interpretation of our predicted subsidy variable in another respect. If officials determined subsidy eligibility on the basis of observations of labor supply, one might expect a stronger negative correlation between subsidies and participation than in the difficult-to-observe marginal supply estimates. We find the opposite.
30. Sahn and Alderman, "The Effects of Human Capital" (n. 12 above).