Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
FOOD POLICY ANALYSIS, NUTRITION, AND ECONOMIC DEVELOPMENT
The interventions discussed above largely operate parallel with, and in response to, a country's development strategy. Their potential is limited, and there is little risk of negative repercussions throughout the economy from a misguided effort. As suggested previously, it is the macro economic environment and micro-economic responses of consumers and producers that are the most important variables in the food system. Therefore, this section begins with a brief discussion of macro-policy. This is followed by an attempt to gain a functional understanding of the micro-economics of consumption and production and to examine the role of price interventions in nutrition and economic development.
Macro Dimensions of Food Policy. A country's macroeconomic policy determines the scope and impact of the micro-economic response of consumers and producers. To illustrate, the rate of inflation and its effects on food prices depend on a country's fiscal and monetary policy. Similarly, taxes not only affect budgetary policy as mediated by fiscal policy, but also have direct effects on income distribution and thereby the demand for food crops. Budgetary allocations are also the mechanism for determining what resources are allocated to intervention programmes.
Of even greater importance is how macro-policy, through foreign exchange rates and trade policy, interest rates and wage policy, determines the price of food, choice of technique used in production (i.e., the mix of inputs and the ratio of capital to labour employed) and the distribution of money incomes. For example, overvaluation of a country's currency allows it to purchase commodities on international markets at lower prices than if foreign exchange were valued at its true economic cost. The result is that domestic supply can be raised through cheap imports, and prices to consumers are lowered. Simultaneously, domestically produced commodities command a lower price. The burden of such a policy falls on the producer as an implicit tax on his production.
Food export taxes and import restrictions, such as those found in Thailand, increase domestic supply and drive down prices. Concessional imports or grants in the form of food aid also may have similar outcomes: lower prices faced by the consumer implicitly paid for by the farmer who must accept a lowered price for his products. Even simple rationing of a commodity throughout a country or region, as has been done with sugar in Pakistan, is a form of an implicit subsidy, because it limits the quantity that may be purchased by high-income consumers, and thus the degree to which they can bid up the price.
Implicit subsidies, in almost all instances, are inadvisable. They turn the terms of trade against, and can do positive harm to, agriculture. This, in turn, not only reduces domestic food production, decreasing self-sufficiency and increasing national food insecurity, but it adversely affects the incomes of those usually in greatest need: rural labourers and small and marginal farmers. Food prices may be lower in the short term because of implicit subsidies, but the long-term prospects of a faltering agriculture, worsening balance of payments, and urban-biased inequitable pattern of growth are all but inevitable.
Similarly, interest rates below equilibrium levels depress the savings rate and availability of credit. Usually it is the small farmer who foregoes access to capital, or is forced to pay black market prices for credit.
Simultaneously, low interest rates foster distortions in factor markets that result in a capital-intensive bias. The result is usually once again an increased skewing in favour of the rich.
It is beyond the scope of this paper to explore macro food policy in greater detail. However, the above illustrations indicate that macro policy is not only the cutting edge of development in general, but functionally affects nutritional status by directly affecting incomes and food prices. Some of these issues will receive further attention in the discussion of micro food policies that follows.
Micro food policy. Food availability at the household level is mediated by a number of factors, especially household income, prices of commodities (food and non food), household characteristics and composition, and consumer tastes (i.e., food habits and taboos). Micro-economic analysis provides researchers with a powerful tool for understanding to what extent these factors affect consumer behaviours, and thereby the scope and potential for various types of intervention.
Specifically, household income, expenditure, and consumption surveys provide the raw data needed to estimate a household consumption function. That is, the household data are usually analyzed using regression techniques in order to provide insight not only into patterns of consumption, but in order to estimate the extent to which food availability will change in response to economic variables such as income and prices, and the array of social and demographic factors that predict consumer behaviour.
In order to attain robust coefficients in a regression equation, regardless of its specification, the data requirements are great, and the analytic procedures sophisticated. This is especially true for price coefficients (i.e., elasticities) that define how food availability will change in response to a shift in prices.
The methodological problems encountered in using survey data to analyze the economic determinates of household food comsumption often stem from the need to disaggregate price and income elasticities by other explanatory variables such as urban-rural, location, the array of house hold sociocultural characteristicis, food source (e.g., purchased, donated, home-grown), and of greatest concern, income. It is also important to consider to what extent elasticity coefficients are predicting short-term versus long-term adjustments in expected food consumption. Although short-term elasticities can, in principle, be measured using panel surveys, this is not usually feasible because satisfactory estimates of coefficients cannot be made at reasonable cost. It is also difficult to factor out secular and structural changes that occur in the economy. Therefore, cross-sectional household surveys are usually employed. In this case, however, elasticities must be regarded as predicting long-term adjustments in behaviour.
A number of other difficulties in using household surveys ża measure income, price, and cross-price elasticities must be recognized. Specifically, the income, expenditure, and consumption data collected by surveys and used to construct elasticities are all subject to measurement error. There are a series of conceptually simple but operationally complex issues that arise. First is the difficulty of defining the actual household unit about which data are collected. Second, there are also inherent difficulties in measuring income, especially in poor countries. Income is often earned on an irregular or seasonal basis, not received in cash, not recorded, and purposefully misquoted by the survey population. Because of these difficulties, expenditure, rather than income elasticities of demand for foods, are computed. But expenditure data, too, have their limitations. For example, there is the problem of imputing expenditures from self-provisioning agriculture. Furthermore, expenditures are often uneven from week to week. A family, for instance, may purchase a large supply of rice once per month. Therefore, careful attention must be given to the survey's time frame. Consideration of the seasonal dimensions of hunger is also important.
Concerning the measurement of consumption that is the numerator in most income or expenditure elasticities, we also must not overlook the potential for over and under- reporting biases. Purchases made outside of the commercial economy may be forgotten, foods gathered in the bush may be neglected; snacks may be ignored, and so forth.
The examples of vagaries in measurement provided above are best addressed through the careful design of survey instruments and training of field personnel. However, other major methodological problems concerning the effects of prices on consumption are particularly noteworthy. Specifically, there is the potential that only limited price variations will preclude estimating meaningful price elasticities. In addition, another difficulty revolves around the need to account for cross price effects -i,e., how changes in the price of one commodity alter the purchase of other food crops when estimating price elasticities. This requires the construction of a complex cross-price matrix, a difficult task that has seldom been undertaken. Notwithstanding the work by Pinstrup-Andersen et al. (15), Alderman and Timmer, and Thomson (20) that indicates such economic analysis to be feasible, there are a number of limitations because of restrictive conditions that economic theory places on estimation and specification of demand equations (21).
Two important final issues concerning interpreting and using elasticities stem from the fact that they only impure how changes in income and prices affect food consumption, and indirectly nutrition. First, in applying estimated coefficients to the individual household, it must be recognized that elasticities were determined using aggregated market data. Therefore, they provide information on expected responses of a group of consumers, and not necessarily the individual consumer.
Second, the estimated elasticities in many analyses of household consumption data provide information on adjusments in expenditures or food intake, Neglecting to address explicitly to what extent there are changes in actual nutrient intake. A significant attempt to reconcile this problem was undertaken by Pitt (22), who has estimated nutrient-expenditure elasticities, and has also constructed a matrix of nutrient-price coefficients. These elasticities quantify how the intake of protein, fat, carbohydrates, iron, calcium, and other nutrients responds to changes in expenditures and food prices. This lucidly illustrates the relative merits, in terms of nutritional intake, of food crop-specific pricing policies.
This brief discussion concerning how to estimate consumption patterns is designed to elucidate the potentials and constraints of micro-economic analysis in the process of formulating interventions. Refining methodologies, both in terms of the collection of accurate data and the specification of regression equations, represents a fruitful area for study. Nevertheless, the application of economic theory must go beyond estimating consumption functions. This would involve using findings to develop rational price and/or income policies and programmes. While in developed countries, interventions in market prices and income transfers may have little nutritional impact, just the opposite is often the case in developing countries.
In this regard, interventions in terms of pricing policy are of major interest, because tansferring income directly to the poor is rarely a politically viable action. An exception to this generalization is an income transfer that is inkind, taking the form of food. Food stamp programmes are the most prominent example of this approach. In theory, food stamp programmes that involve a purchase requirement can affect consumer behaviour at the margin, increasing food acquisition in excess of what it would be with a straight cash transfer. However, in practice, purchase requirements are administratively cumbersome (especially when they are variable, involving differential pricing of stamps to individual households), and may, in fact, be regressive in the absence of a variable requirement. In other words, if there is a fixed payment required to purchase food stamps, this policy will discriminate against households in greatest need and not capable, because of scarce income, of coming forth with the amount of money needed to purchase the stamps. Others may not be willing to purchase stamps because of perceived tradeoffs, such as the time required for participation in the programme.
In the final analysis, the marginal propensity to consume food, i.e., the proportion of additional income used to purchase food, from a food transfer, even if it be inframarginal, may be higher than a cash income transfer of equal value. This may be attributable to the form. and timing of the transfer, and factors such as its accruing to the female head of household. However, the empirical evidence indicates that families also place a high value on, and therefore have a high marginal propensity to buy fuel, clothing, shelter, and so on. This, combined with the fact that households can be expected to purchase more expensive foods (i.e., a higher price per calorie and usually higher in protein quality), clearly limits the potential impact of income transfers. Nevertheless, there is a need for further research in this area in order to gain insight into the nutritional implications of various cash and in-kind transfer policy options.
While creating demand for calories through transferring income remains an intervention requiring further consideration, more often than not, countries manipulate food prices to meet welfare and nutritional objectives. The manipulation of food prices to achieve nutritional as well as related economic and political objectives has not only been a preferred option in many countries, but has great implications in terms of promoting long-term economic development.
Bole of price policy. Concurrent with the recognition of the role of prices as a determinant of nutritional status, there has been increasing attention given to the fact that price policy cuts with more than one edge. It directly affects consumption and production decisions, income distribution, and the process of economic development in general. Drawing on a framework developed by Timmer (23), these functions are briefly discussed here.
Consumption. Prices work directly in two dimensions to affect consumption. The first is through the effects of prices on real income available to the family. Because food is usually the major consumer good, especially among the poor, the real income effect of a price change can be significant, freeing additional income for food and non-food expenditures.
The second dimension is through the substitution effect of a price change. Not only does the consumer's utility level shift because of changes in real purchasing power, but prices also dictate how relatively attractive food is compared to other wage goods. As such, a price-related intervention has greater potential impact on food intake because of the reordering of consumer preferences in favour of the food crop with a lower price, which is a corollary of the substitution effect.
Production Prices at the farm gate are central components of a healthy and dynamic agriculture-a prerequisite to any long-term effort to alleviate malnutrition. The evidence that farmers are price-responsive is convincing. Although short-term elasticities of supply are relatively low for staple crops-usually in the range of 0.1 to 0.2 for starchy staples (24) the mix of output is even more subject to price changes. Furthermore, evidence that higher prices result in increased agricultural research, and commensurately induce innovation in agricultural technology and institutional capacity, indicates that long-term elasticities are considerably higher (25).
The importance of improving the productive environment for small farmers and the rural landless, who characteristically are among the most nutritionally vulnerable segments of the population, requires attention beyond maintenance of price incentives. First and foremost, attention must be given to patterns of land ownership and tenure, as well as to access to resources and the fruits of production. Efforts are also necessary to enhance production and marketing abilities of marginal farmers through measures such as developing appropriate low-cost, low-risk technology packages, improving post-harvest storage techniques, ensuring access to markets and accurate market information, and improving availability of credit.
Income distribution and economic growth. The differential effect of prices on the incomes of producers and consumers, coupled with the repercussions related to employment generation, make pricing policy a central tenet in determining income distribution. As suggested by Mellor (26), in the short-run the poor " . . . spend a high proportion of their income on food and depend directly or indirectly on agriculture for their employment and income. In the longer run, food price policy may affect the supply function for wage goods and thereby influence the extent to which total wage employment and hence income of the laboring classes can be expanded."
It is difficult to determine the exact nature of the distributional effects of price policy without a general equilibrium analysis. Complicating the task is the need to take into account both the expected short-term and long-term supply response to price changes disaggregated by food crop, as well as to specify a consumption function that disaggregates elasticities by income group, urban versus rural, and so forth. Nevertheless, certain aspects of the complex relationship can be elucidated.
For example, we know that low--income consumers usually suffer disproportionately from a price increase, although the wealthy are more affected in absolute terms. Similarly, the effect of prices on producer incomes depends on the crop mix being produced; the amount of output; the elasticity of marketable surplus; and the quantity of agricultural inputs purchased. Using similar variables, Mellor's analysis suggests that it is the high-income producers who benefit mainly from high market prices (26).
Perhaps the most difficult aspect of the price policy nexus revolves around the issue of employment generation. First, price changes alter consumption patterns, and thereby demand for commodities. This, in turn, indirectly affects employmet opportunities. Prices also affect levels of production, and thus employment and the supply of wage goods. Prices may also affect choice of technology, which could have an adverse or positive impact on employment. This depends on the direct and indirect effect of technological change, the mix of labour and capital used, and the growth of output. Further more, any changes in food crop prices will alter the urban-rural terms of trade, which will also have unforeseen results throughout the economy.
To illustrate the opposing points of view concerning how food prices affect the process of employment and economic development, we need only to contrast two important models of economic development. In the Lewis (27) model, low food prices owing to the surplus of manpower are viewed as a method of extracting the surplus form the agricultural sector. Lewis postulates that this, in turn, can be invested in the modern, commercial sector, spurring industrial growth and expansion, and eventually providing jobs for the rural unemployed.
Juxtaposed with the Lewis model is Mellor's work (28), stressing that the process of economic development necessitates the maintenance of high food prices to the producer. Specifically, it is argued that agriculture must play the leading role, or at least be an equal partner, in the process of economic growth. Simply, the theory suggests that producer incentives, increased by high prices, will lead to a dynamic agriculture. Backward and forward linkages will thereafter result in new employment opportunities, especially for the poor. As income accrues to the poor, with their high elasticities of demand for food, this will fuel production, especially of basic grains. In fact, it is expected that, as income distribution shifts to the poor, demand will increase for domestically produced agricultural inputs, labour-intensive primary products and consumer goods, rather than for capital-intensive and imported goods. Ideally, a dynamism results throughout the economy, precipitating growth and development.
Naturally, presentation of the stylized theories alluded to above is not an attempt to analyse and draw conclusions about the appropriateness of various development strategies. Nor is it an attempt to examine the validity of such alternatives. In reality, the links between prices, employment generation, and economic growth are complex and contextual. However, it is very clear that getting food prices right is a difficult and inherently contradictory process. The fact that pricing policy cuts with more than one edge inevitably means that, while one income class is helped, another is hurt; or alternatively, the reality of price formation may mean hurting the rural poor while assisting the urban poor, or vice versa. In fact, in many cases, the same policy may both aid and harm the same group simultaneously through different mechanisms.
A few attempts at reconciling inherent conflicts in setting food policies have been undertaken (29). Inevitably, the food policy dilemma arises of keeping prices high for producers while at the same time not imperiling the poor consumer. Food price subsidies represent one attempted solution to this conflict.
CONSUMER SUBSIDIES
Consumer food price subsidies have been implemented throughout the world because they achieve multiple purposes, including socioeconomic and political objectives, in addition to increasing food consumption. The fact that consumer subsidies are designed to realize broader goals than reducing malnourishment becomes obvious when one considers their cost in relation to their impact on caloric consumption among the most vulnerable groups, and the experience that consumer-oriented food subsidies have largely been limited to urban areas, not reaching the rural poor who are at greatest nutritional risk.
Food price subisidies vary along a number of dimensions. These include:
The brief list above illustrates some of the major design and policy issues that relate to food price interventions. These are discussed in detail in the literature (30,311. Of equal importance, the micro-level response of consumers to a subsidy programme has macro-economic linkages and implications. For example, there is considerable potential for lower prices to stimulate demand and thus production. The extent to which demand is increased is partially a function of price incentives, although the expectation of significant production increases is contingent upon improvement in agricultural practices and the availability of needed inputs such as seeds, fertilizers credit, and extension services. Without these ingredients, there will be only a limited short-term supply response that will be largely in the choice of food crop to produce. The expected result will thereafter be higher food prices, as domestic increases in production will not keep pace with demand. The implication of such a scenario is that inflation caused by lagging production will imperil the poor. Gains achieved through subsidization can thereby be negated. As such, the risk of inflation adds further impetus to the argument that consumption-oriented strategies must be tied to initiatives to increase output of food crops.
Finally, an important development impact that militates against food subsidies is their high cost. For example, in Sri Lanka, 16.6 per cent and in Bangladesh, 21.5 per cent of total government expenditure was for the food subsidy ;321. Fiscal effects in the form of increased taxation and expenditures inevitably portend serious macroeconomic consequences. In fact, depending on how and from whom tax is extracted, such policies may result in changes in aggregate demand that are more consequential than those resulting from price changes.
Despite the problems inherent in food subsidies, there are potential mechanisms for circumventing some of the major design and operational problems, both in terms of the micro-response of consumers and producers and macro-level linkages. While there are inevitable risks in the process of controlling food prices, few governments do not already drive a wedge between international, producer, and consumer prices. The creative task is to perform careful policy analysis so as to increase the substantial benefits of price-setting, while minimizing potential negative repercussions.
CONCLUSION
Most health and nutrition planners have focused their efforts on formulation, implementation, and evaluation of micro-level nutrition interventions. While this orientation is appropriate, given the nature of the hunger problem and the direct measures to reduce malnutrition, more attention should be given to developing macro-models that predict the nutritional implication of changes in macro-economic policy. While a formal, general equilibrium analysis is probably beyond the scope of most countries, and will quickly reveal how little we know about the science of economic modeling, it is feasible and desirable to begin the process of looking beyond traditional interventions We suggest an attempt to quantify, or at least qualify, how nutritional status and the scope and potential for pricing interventions are conditioned by macro-policies. Inevitably, this leads to the formulation of economic decisions in the appropriate political context.
While a paper of this kind can urge policy-makers to take the bold initiatives required to bring about needed reform, it cannot provide them with the methods for doing so.
Nevertheless, as a beginning to this process, it would be fruitful to integrate explicit nutrition considerations into a broader programme and policy framework. This is not to suggest that nutritional considerations be accorded priority over all other sector goals.
Rather, that when macro-policy in general, and specific agricultural and rural development projects in particular, are being; formulated, nutrition concerns should be incorporated into the process. Doing so will at least call to the attention of national planners the needs of the poor and indigent and perhaps favourably affect the outcome of economic development.
REFERENCES
1. T.W. Schultz, "Investment in Human Capital." Amer. Econ. Rev. 51: 1 (1961).
2. A. Berg, The Nutrition Factor. The Brookings Institution Washington D. C. (1973).
3. P. Hakim and G. Solimano, "Nutrition and National Development: Establishing the Connection." MIT International Nutrition Planning Program Paper No. 5, Massachusetts Institute of Technology, Cambridge, Massachusetts, 1975.
4. S. Reutlinger and M. Selowsky, "Malnutrition and Poverty: Magnitude and Policy Options." World Bank Staff Ocassional Papers No. 23 (1976).
5. S. Kuznets, Economic Growth of Nations Harvard University Press, Cambridge, Massachusetts, 1972.
6. I. Adelman, M. Adelman, and C. Taft, Economic Grovvth and Social Equality in Developing Countries. Stanford University Prep, Stanford, California 11973).
7. W. R. Cline, "Distribution and Development: A Survey of the Literature." l Develop. Econ. 1: 359 (1975).
8. J. Field, "The Importance of Context: Nutrition Planning and Development Reconsidered." In: Nutrition in the Community, D. McLaren (Ed.), John Wiley and Sons, Chichester, Sussex, U. K. (in press).
9. B. Winikoff, "Political Commitment and Nutrition Policy." In: Nutrition and National Policy, B. Winikoff (Ed.), MIT Press, Cambridge, Massachusetts (1978).
10. D. R. Gwatkin, "Food Policy, Nutrition Planning, and Survival-The Cases, Kerala and Sri Lanka." food Policy 4: 4 (1979).
11. A. O. Krueger, Foreign Exchange Regimes and Economic Development: Liberalization Attempts and Consequences. Baldinger Cambridge Massachusetts (1978).
12. N. Eberstadt, "Hunger and Ideology." Commentary 72 (1): 40 (1981),
13. Ministry of Health, Sri Lanka, White Paper on the Reorganization of the Ministry of Health. Colombo, Sri Lanka (1980).
14. G. H. Beaton and H. Ghassemi, "Supplementary Feeding Programes for Young Children in Developing Countries." ACC Sub-Committee on Nutrition of the United Nations, UNICEF New York (1976).
15. P. Pinstrup-Andersen N. Ruiz de Londono, and E. Hoover "The Impact of Increasing Food Supply on Human Nutrition: Implications for Commodity Priorities in Agricultural Research and Policy." Am. J. Agric. Enco. 58 131 (1976).
16. J. E. Austin and M. F. Zeitlin, Nutrition Intervention in Developing Countries Oelgeschlager, Gunn and Hain, Publishers Inc., Cambridge, Massachusetts (1981).
17. D. E. Sahn and R. Pestronk, "A Review of Issues in Nutrition Program Evaluation." AID Program Evaluation Discussion Paper No. 10, U. S. Agency for International Development, Washington, D. C. (1981).
18. S. Chafkin, "Nutrition Policies, Programs and Politics." Am. J. Agric. Econ. 50: 806 11978).
19. C. Schuftan, "Nutrition Planning-What Relevance to Hunger?'' Food Policy 3: 1 (1978).
20. Dissertation, Stanford University, Stanford, California (1979).
21. C.P. Timmer, "Estimation, Consumption Parameters for Food Policy Analysis. Am. J. Agric. Econ. 61: 5 (1979).
22. M. Pitt, "Food Preferences and Nutrition in a Poor Developing Country: An Analysis of a Household Expenditure Survey of Rural Bangladesh." Department of Economics, University of Minnesota, Minneapolis (1981),
23. C.P. Timmer, "Food Prices and Food Policy Analysis in LDCs." Food Policy 5: 3 (1980)
24. R. Herdt, "A Disaggregate Approach to Aggregate Supply." Am. J. Agric. Econ. 52:512 (1970).
25. Y. Hayami and V. Ruttan, Agricultural Development: An International Perspective. Johns Hopkins University Press. Baltimore, Maryland 11972).
26. J. Mellor, "Food Price Policy and Income Distribution in Low-lncome Countries." Econ. Develop. Cult. Change 27: 1 (1978).
27. W.A. Lewis, "Economic Development with Unlimited Supplies of Labour." The Manchester School, 1955.
28. J. Mellor, The New Economics of Growth, Cornell University Press, Ithaca, New York (1976).
29. D. McCarthy and L. Taylor, "Macro Food Policy Planning: A General Equilibrium Model for Pakistan." Fox. Icon. Statist. LXII: 107 (1980).
30. B.L. Rogers, C. Overholt, F. Sánchez-Carrillo, A. Chávez C.P. Timmer, and T. Beldins Consumer Price Subsidies, in Special Study VI in AID Series Nutrition Interventions in Developing Countries. Oelgeschlager, Gunn and Hain, Publishers, Inc., Cambridge, Massachusetts (1980).
31. L. Taylor, S. Hortin, and D. Raft, "Food Subsidy Programs: A Survey." Report to the Ford Foundation, blew York (1980).
32. J. David "The Fiscal Role of Food Subsidy Programs," International Monetary Fund Staff Papers 24: 1 (1977).