Ending food subsidies : nutritional, welfare, and budgetary effects

14z.,o THE WORLD BANK ECONOMIC REVIEW, VOL 3, NO 3 39S-408 FIL Ending Food Subsidies: Nutritional, Welfare, and Budgetary Effects Karim Laraki Governments faced with growing budget deficits are cutting many social expenditures, including costly food subsidy programs that have provided benefits to the rich and poor alike Because the poor spend a larger share of their income on food than do the rich, however, such cuts usually have negative distributional, welfare and nutritional effects This article discusses the methodological issues in estimating the effects of price and tax reforms in developing countries I apply a model that Deaton (1988) developed to estimate price elasticities from cross-section data, the only reliable and detailed data available in most developing countries I use measures of both real income and nutrition to evaluate the effects of changes in the Moroccan food subsidy program The analysis suggests that subsidtes on inferior foods not consumed by the wealthy would reduce the welfare costs to the poor and limit the budgetary expenditures required Food price subsidy programs are getting a hard second look in many countries because of their large contributions to government budget deficits The subsi- dies are expensive because they are seldom targeted to the poor a significant part of the food subsidies goes to high-income groups since the rich spend more on food in absolute terms than do the poor But attempts to remove subsidies have been opposed on the grounds that the poor spend higher proportions of their income on food than do the rich An increase in food prices, it is main- tained, would lower the standard of living of the poor How much would removal of food subsidies affect the poor) This article discusses these issues and attempts to determine whether a narrower targeting of subsidies can benefit the poor and relieve budgetary strains Subsidizing a nutritionally rich but economically inferior good would com- pensate the poor for income and nutritional losses from price increases This would also guarantee that the benefits of the subsidy would not leak extensively The author is a professor of agricultural economics at the Institut et Veterinaire Hasan 11, Morocco He is grateful to Jacques van der Gaag and Angus Deaton for contributions to an earlier version of this paper, to Dwayne Benjamin for his help with the estimation of the price elasticities, to Bruce Ross- Larson for his editorial comments, to the referees for their constructive criticism, and to Brenda Rosa for putting together the final version Statistical assistance from the Moroccan Ministries of Planning, Economic Affairs, and Agriculture is gratefully acknowledged @1990 The International Bank for Reconstruction and Development / THE WORLD BANK 39S  396 THE WORLD BANK ECONOMIC REVIEW, VOL. 3, NO. 3 to high-ihcome groups thus reducing the government's cost. In Morocco, food subsidies represented 10 percent of current government expenditures in the early 1980s. Although the percentage has diminished in recent years as prices of subsidized imported foods have fallen, a continuing economic crisis has caused the government to embark on a policy of gradually eliminating the subsidies. Subsidies on meat, butter, and other dairy products have been re- moved since the late 1970s, but attempts to increase prices for the most impor- tant staples (soft wheat, vegetable oil, and sugar powder) have been vehemently opposed by the media and have ignited street riots. Subsidies on these commod- ities are still in place. This study investigates the potential effects of removing the remaining subsi- dies in Morocco. The three criteria used to evaluate the price reform are its effect on real income across income groups; on calorie consumption; and on the government. The method used to evaluate the effect of a food price rise on real income is first described. But for those at the bottom of the income distribution, it is the commensurate declines in consumption that are of concern, as consumption levels could fall below some minimal nutritional requirement. Thus I also explain how induced caloric changes are estimated. Aggregate effects on the government budget will depend on the price and income elasticities of demand for the food products involved. Calculating income elasticities is generally easy, but calculating price elasticities is not-because of the lack of reliable time- series data in developing countries. Nonetheless, many countries have budget surveys of good quality. These surveys do not report on prices but observations on expenditure levels and quantities demanded can be used to estimate price elasticities (Deaton 1988). I apply Deaton's technique to the Moroccan case. The empirical results are then presented including the consequences of a move to subsidies on two inferior foods. The results suggest that such a program could reduce both the budgetary burden and the costs to the poor. 1. METHODOLOGICAL ISSUES IN ANALYSIS OF PRICE REFORM The Welfare Effects A price increase reduces purchasing power and therefore real income (all else equal). The effect of a price reform on consumers' welfare can be calculated by estimating the minimum amount of income by which a consumer must be compensated after a price increase to be as well-off after the change as before. This can be computed by use of Hicks's (1956) compensating variation (CV) index (see Varian 1984). 1 follow the approach adopted by Braverman, Ham- mer, and Ahn (1987). The welfare effect of a price change from p