13 SZ(f I?s lCe Labor Supply and Targeting in Poverty Alleviation Programs Ravi Kanbur, Michael Keen, and Matti Tuomala The introduction of variable labor supply raises some fundamental issues in analyzing the targeting of poverty alleviation programs in developing countries. It forces a recon- sideration of the standard obiective function, which is based on income or expenditure and so makes no allowance for the effort made in earning that income. We show that alternative views on the appropriate valuation of effort have very different implications for commodity-based targeting rules. We also establish a benchmark for marginal effective tax rates (inclusive of benefit withdrawal) in income-tested schemes and show that indicator targeting rules may also have to be modified significantly when labor supply responses are recognized. For many governments of developing countries, finer targeting of programs to alleviate poverty appears an attractive option in an era of greatly constrained expenditure budgets. It seems as though policymakers could achieve greater poverty reduction with fewer resources if only they would resort to the magic of targeting. But fine targeting is not without its costs. It is now appreciated that the administrative costs of ensuring that benefits from a program reach the target group can be high (see Besleyand Kanbur 1993). One response is to target by subsidizing commodities largely consumed by the poor or on the basis of other observable indicators-such as age, gender, region, or crop group-that are correlated with deprivation. There is now a literature on how such indica- tors might be used for optimal targeting (see, for instance, Akerlof 1978, Atkin- son 1992, Besley and Kanbur 1988, and Kanbur 1987). An aspect of the costs of fine targeting that has not been as well appreciated in the development literature as it should be is the effect on incentives. Consider, for example, the moves in Sri Lanka to target the rice ration subsidy in the wake of the economic reforms of the late 1970s (see Anand and Kanbur 1991). The system was transformed from one with a universal benefit to one in which the benefit was restricted (in principle) to those with incomes below a critical level Ravi Kanbur is with the Western Africa Department at the World Bank, Michael Keen is with the Department of Economics at the Universityof Essex, and Matti Tuomala is with the Department of Economicsat the Universityof Jyvaskyla.The authors are grateful to Steve Coate, Jonathan Morduch, KimNead, and Dominique van de Walle for helpful comments. © 1994 The International Bank for Reconstruction and Development / THE WORLD BANK 191 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized