Even a casual look at agricultural policies around the world reveals many surprising anomalies. In the United States, for example, the government pays farmers not to grow cereals; in the European Community, farmers are paid to grow more. In Japan, rice farmers receive three times the world price for their crop. In 1985, farmers in the EC received 180 a pound for sugar that was then sold on the world markets for 50 a pound; at the same time, the EC imported sugar at 180 a pound. Canadian farmers pay up to eight times the price of a cow for the right to sell that cow’s milk at the government’s support price.
In contrast to industrial market economies, developing countries tend to tax agriculture—even those low-income countries that depend critically on agriculture for their economic growth. Some pay their producers no more than half the world price for grains and then spend scarce foreign exchange to import food. Many have raised nominal producer prices but followed macroeconomic and exchange rate policies that left real producer prices unchanged or lower than before. Many have set up complex systems of producer taxation, and then established equally complex and frequently ineffectual systems of subsidies for inputs to offset that taxation. Many subsidize consumers to help the poor, but end up reducing the incomes of farmers who are much poorer than many of the urban consumers who benefit from the subsidies. Most developing countries pronounce self-sufficiency in food as an important objective, while taxing farmers and subsidizing consumers and thus increasing their dependence upon imported food. And in periods of economic adjustment, when shortages of foreign exchange make export promotion urgent, many have increased taxes on agricultural exports and cut producer support programs, while relying on unrestricted food imports to satisfy urban consumers.
These examples raise many questions. What are the typical agricultural policies of developing and industrial countries? Are they efficient? How well do they serve the objectives of economic growth, the elimination of hunger, and the alleviation of poverty? How do these countries’ agricultural policies affect each other? Even if the external environment facing developing countries is a difficult one, are they making the most of it or are they making matters worse through domestic policy mistakes? If agricultural trade and domestic policies were liberalized throughout the world, could one expect substantial gains for the world economy in general and for developing countries in particular? This article outlines some answers to these questions, drawing on the discussion in the Bank’s 1986 World Development Report.
Policies in developing countries
Agriculture is the basic economic activity of the world’s poorest countries. It employs 70 to 80 percent of the labor force in low-income developing countries and 40 to 50 percent in middle-income developing ones. It is also a main source of gross domestic product, accounting for 35 to 40 percent of GDP in low-income developing countries. Agriculture’s share of national income generally declines as real per capita incomes rise, because people spend a decreasing percentage of their incomes on food as their incomes increase. Almost all of today’s industrial nations had roughly the same percentage of their labor forces engaged in agriculture in the nineteenth century that the low-income developing countries have now. The farmers of the industrial countries have also steadily increased the productivity of their land and labor so that an ever-decreasing share of their country’s resources is needed to grow food for the rest of the population. In most of the industrial economies a farm family produces enough food for itself and as many as 50 others. In the low-income developing countries, on the other hand, a farm family provides enough food for itself and only two other people.
Although the share of agriculture in national income declines over the long term, this trend should occur naturally: neglecting agriculture and forcing its share to go down faster depresses economic growth. The experience of decades suggests that a healthy agricultural sector is essential to overall growth.
Agriculture also plays an important role in short-term economic adjustment for the many countries whose exports are largely agricultural. For these countries, improvements in the balance of payments depend heavily on the foreign exchange that can be earned by agriculture. Because agriculture accounts for large shares of incomes and export earnings in so many developing countries, success there will strongly influence the course of their economies for many years to come.
Many governments have promoted agriculture by making substantial investments in rural infrastructure, in expanding irrigation and flood control, and in strengthening agricultural research and extension services. Other programs have helped to raise productivity through better farm management and improved health, nutrition, and education.
Public spending on these types of services has been extremely helpful in many cases. For example, expansion in irrigated areas and the development of new varieties of wheat and rice have been major factors behind the growth of agricultural production in Asia and South America—two regions in which per capita food production easily exceeded population growth during the last 15 years. Yet the provision of essential public services in rural areas is but one of many elements of economic policy that determine growth in agricultural output and rural incomes. Overall, policies in developing countries have been biased against agriculture.
Developing countries pursue a wide range of agricultural policies, but in many, economywide policies have limited the growth of agricultural production and hampered efforts to reduce rural poverty. Sector-specific pricing and tax policies have also often resulted in substantial discrimination against agriculture. Discriminatory policies have been particularly serious in sub-Saharan Africa, the only region in the developing world that has failed to expand food production sufficiently to match population growth. A reversal of that trend, as well as expansion of agricultural exports, will be necessary if the countries in that region are to cope successfully with the current problems of high indebtedness and attain positive rates of per capita economic growth over the medium term.
Economy-wide policies. Trade, exchange rate, and macroeconomic policies have a significant influence on agriculture in all developing countries. Given their importance, a companion article by Chhibber and Wilton discusses the issues in some detail. The most important points are:
• Much of the bias against agriculture in developing countries arises from policies to promote industry behind high trade barriers. Inward-looking industrialization strategies accelerate the shift of resources out of agriculture by lowering its profitability vis-á-vis the industrial sector. Agricultural exports suffer, as do agricultural products that compete with imports. This is not just because their domestic prices become lower relative to the prices of protected industrial products but also because the costs of the industrial inputs used by farmers increase. Moreover, protectionist policies result in an appreciation of the real exchange rate. As a result, traded agricultural goods also become less profitable than nontraded goods.
• During the last 15 years the bias against agriculture has often been intensified by the way developing countries have responded to changing economic circumstances. For example, when expansionary monetary and fiscal policies have led to higher inflation at home than abroad, governments have often failed to adjust exchange rates, relying instead on increased protection against imports by means of quotas, exchange controls, and licensing. In such situations the currency becomes overvalued and the bias against agriculture deepens, since the benefits of increased protection usually accrue to industry. Typically, imported foods are exempted from such measures in order to keep urban food prices low. Food imports, in short, are implicitly subsidized. Furthermore, in trying to reduce fiscal deficits in such situations, countries are apt to increase taxes on agricultural exports.
• When capital inflows from abroad or sharp increases in the world prices of key exports cause the real exchange rate to appreciate, countries typically react with expansionary monetary and fiscal policies. These then lead to inflation and augment the appreciation of the real exchange rate that was caused by the favorable change in the external terms of trade. The effects of this reaction continue even after the boom ends, because by then commitments to large investment programs or to large recurrent costs have already been made.
The differential protection of industry and currency overvaluation imply an implicit taxation of agriculture, which can often—as explained in the next article—outweigh the effects of sectoral policies. But the latter have also been extremely important sources of bias against agriculture.
Sectoral policies. Farmers effectively face many sectoral taxes and subsidies apart from conventional trade duties and subsidies. Examples are quotas, domestic sales taxes, and pricing policies of public marketing agencies. The net effects of the various policies can be captured by measuring the differences between farm-gate prices and border prices, at official exchange rates, after adjustments for internal transport and marketing margins. These net effects are additional to the implicit effects of currency overvaluations.
Many developing countries tax export crops—mainly raw materials and beverages—at high rates. Tax rates of 50 percent or more have been common in many countries, especially in Africa. Most of this taxation results from the policies of export marketing boards with statutory monopoly powers. A few years ago, the farm price for coffee in Togo was only one third of the international price; cotton and groundnut farmers in Mali received only half the international prices; and in Cameroon and Ghana cocoa producers received less than half this price. Asian and South American countries frequently tax their exports not only of raw materials and beverages but also of grains.
Some agricultural import substitutes—especially wheat, dairy products, and livestock—are protected in a few developing countries. But in most cases, the domestic producers of import substitutes receive less than import prices, adjusted for internal marketing costs. Taxation rates on import substitutes have often been excessive. For example, Tanzania’s official price for maize has been only about 25 percent of the import price in some years, while in Cameroon and Ghana, rice producers have received only about half the import parity prices.
Typically, the most important reason for taxing agricultural output is to raise revenue or control the costs of other public programs. Although most parastatal export marketing boards were originally required to use their funds to assist farmers, many have become de facto taxation agencies for financing public spending elsewhere. The most important reason for taxing import-competing crops is to control the costs of urban food subsidy programs.
Raising revenue through export taxes may be self-defeating. First, the rates of taxation often seen are likely to generate less, rather than more, revenue than lower rates, because of their effect on incentives to produce. Second, the real national income sacrificed in the process of taxation escalates rapidly as the tax rate increases. Assuming conservatively that a country’s exports of a commodity rise in exact proportion to the commodity’s international price, getting the last dollar of tax revenue will cost only 5.6 cents if the export tax is 5 percent. If, however, the rate is 40 percent, the last dollar of tax revenue will cost two dollars to obtain. And indeed, beyond 50 percent, total revenue will decrease when the tax rate is increased so that it would be pointless to increase the tax further. If export taxes must be used, much lower rates are desirable.
Similarly, where import-competing food crops are taxed, excessive rates depress domestic supplies and cause a high dependence on imports. The priority should rather be to lower the costs of food subsidy programs to levels the country can easily afford and to target them better to reach the poorest income groups.
Among the other reasons for high taxation of export crops in developing countries are: promotion of agroindustries by lowering the costs of their exportable agricultural inputs; promotion of food production by diverting resources from export crops; and the exploitation of perceived monopoly powers in world markets. Upon close examination of cross-country comparisons, these reasons appear to be untenable. In the case of agroindustries, for example, taxes or quotas on their raw material inputs have imposed high economic losses in countries as diverse as Brazil, Ghana, and Tanzania.
As shown by detailed studies in Argentina and Chile, excessive rates of sectoral taxation, when combined with the taxation implicit in persistent overvaluations of the exchange rate, can have extremely harmful effects on economic and agricultural growth. If discrimination against agriculture is sustained for some time, farmers and private capital move out of farming altogether. Farms then deteriorate, and migration to cities becomes excessive. Those who remain on the land find it less profitable to use input-intensive methods. As in Africa, this inhibits technical progress and hastens soil degradation.
It is often argued that the generous subsidies that many developing countries provide on such farm inputs as chemical fertilizers, machinery, seeds, pesticides, and credit offset the bias against agriculture caused by the heavy taxation of farm outputs. This, however, is not true. It is extremely difficult to implement such subsidies in an equitable and efficient manner. The availability of credit and modern inputs is often jeopardized by the budgetary costs of the subsidies and of the inefficiency of public distribution systems. The main beneficiaries of these types of subsidies are typically larger and relatively affluent farmers. In addition, input subsidies cause distortions in the choice of crops and farming techniques, often compounding rather than offsetting the adverse effects of output taxes.
Neither does the widespread intervention by the government in the marketing of agricultural products help to offset the effects of taxation. Such intervention has in practice tended to be inefficient. The experience in many developing countries suggests that better results can be achieved through greater reliance on private markets. The losses of marketing parastatals are often extremely large, in some cases reaching 1 or 2 percent of national income. Public monopolies impose high costs, even when farmers are able to bypass official channels and sell in parallel markets.
Public marketing agencies are often required to stabilize consumer and producer prices, in addition to carrying out other functions. Partly because of multiple and often conflicting objectives, efficient and profitable price stabilization schemes are hard to find in developing countries. The methods typically used for price stabilization can increase the economic costs of price fluctuations in world markets; they frequently result in excessively large buffer stocks, erratic changes in “floor” and “ceiling” prices, and high budgetary subsidies. In addition, public stabilization schemes displace private sector operations. Crowding out becomes especially serious when public agencies try to impose the same price across different regions of the country and across different seasons.
Regardless of their original motivation, both economy-wide and agricultural policies in developing countries have evolved in ways that discourage growth of agricultural output and rural incomes. There is great scope for improving policies and performance. In recognition of this, several countries, notably China and Turkey, have undertaken broad programs of reform. Many others—including countries in Africa—are changing their policies more gradually, to make programs for urban food subsidies more efficient, to curtail or eliminate farm input subsidies, to reduce state intervention in marketing, and to improve farm output prices. Policy reforms of this type, along with better economy-wide policies and more expenditures on rural infrastructure, offer developing countries their best chance of promoting growth in agriculture and in the wider economy, thereby facilitating sustained progress toward food security—the eradication of poverty, malnutrition, and the periodic occurrence of famines.
Policies in industrial countries
The main objectives of agricultural policies in industrial countries are to stabilize and increase farmers’ incomes and slow the migration of people out of the agricultural sector. Underlying these objectives are the social and political aims of stable food prices and self-sufficiency in food production, which go hand in hand with such other goals as preventing environmental damage to the countryside and preserving the traditional farming unit, usually the family farm.
A variety of measures are used to support farm incomes at levels higher than would result from free trade. Apart from import tariffs and quotas, variable import levies are often used—particularly in the EC—to maintain high and stable domestic producer prices. If protection creates excess supplies, the excess is ultimately disposed of in world markets through subsidized sales or as food aid. Export restitutions are the exporter’s equivalent of variable import levies, permitting domestic prices to be independent of world prices. State-controlled marketing boards, direct payments to producers, and subsidies on inputs and credit are also widely used to aid farmers. When farmers produce too much, several countries use acreage controls to keep down surpluses.
On average, producer prices in industrial market economies are about 40 percent above comparable world prices, but there is considerable variation in protection rates by commodity and country. The average protection rate during 1980-82 in Japan, for example, was about 144 percent, and in the EC about 84 percent. By contrast, there is hardly any protection in New Zealand. Among commodities, the most protected items are sugar, dairy products, rice, and beef.
While protection rates tend to vary from year to year, the general trend has been sharply upward. Average protection rates in the mid-1950s were about 40 percent in Japan and only 16 percent in the EC. Agricultural protection in industrial countries in agriculture has never been higher—even compared to the 1930s. Moreover, those developing economies which have grown fast have begun to emulate the agricultural policies of industrial countries. For example, the Republic of Korea was taxing farmers in the mid-1950s, albeit at a rate much lower than typical in low-income countries today, but by the early 1980s its average agricultural protection rate had reached 166 percent.
Excessive support of farmers over such a long period has predictably increased yields and production, just as the excessive and sustained taxation of farmers in some parts of the developing world has produced the opposite effect. For example, the EC is now a large exporter of grains, although it was an importer when it initiated its Common Agricultural Policy. Thanks to its farm support policies, it is now second only to the United States as an agricultural exporter. It has captured large shares of world markets—especially in wheat and wheat flour, sugar, beef, and butter.
Support to farmers is wasteful and depresses national incomes; consumers and taxpayers always lose more than farmers gain. The 1986 World Development Report provides various estimates of the losses, drawing on the existing literature. It also presents the results of a special study done for the Report, which indicates that consumers and taxpayers in the OECD countries lose about $104 billion each year in order to support a sector of the economy that contributes only small shares of national income and employment. Farmers do gain, but by much less.
The gains that farmers realize from these support policies get capitalized in terms of higher land values. Moreover, policies to protect farmers have also become less necessary as farming has become more and more a part-time occupation. In the United States, net farm income as a proportion of farmers’ total income fell from 58 percent in 1960 to 36 percent in 1982. In Japan, where small-scale farming is dominant, farm households derived 75 percent of their income from nonfarm sources in 1980. The families of part-time farmers with permanent jobs outside farming were about 25 percent better off than families with one or more full-time farm workers.
As in developing countries, many governments in industrial countries are considering agricultural policy reforms. It has become clear that without policy changes to reduce protection, domestic costs will continue to rise in the years ahead, whatever means are chosen for handling growing excess supplies.
• Adding to stocks, as the EC and the United States have been doing for cereals and dairy products, will become increasingly costly and eventually unsustainable as stocks grow larger in relation to annual domestic use or the available storage capacity.
• Restricting output through direct intervention, such as the milk quotas in the EC or acreage restriction programs in the United States, is unattractive, economically and politically. Compulsory measures are unpopular with producers. If compliance with such measures is voluntary, US experience indicates that the budgetary and economic costs of obtaining even a modest reduction in output are great.
• Encouraging consumption domestically or abroad via subsidies will require even more budgetary outlays.
Consequences and priorities
As is clear from this discussion, industrial and developing countries tend to follow exactly opposite policies. Industrial countries produce too much, and developing countries produce too little. The patterns of world agricultural production and exports would change in favor of developing countries if both groups of countries followed efficient policies.
Just as a country gains by allocating its resources to where they can be most productive, the world as a whole would certainly gain if more were produced in developing countries for both domestic use and exports. This would require much freer trade and liberal domestic policies in all countries.
It is extremely difficult to make confident estimates of just how much the world might gain with such policies. Nonetheless, a review of the literature and the background work for the World Development Report suggest that the gains would indeed be very substantial. If industrial market economies and developing countries were simultaneously to liberalize their domestic policies and remove trade barriers, in temperate-zone products alone, industrial market economies would gain by about $46 billion annually and developing countries by about $18 billion. These estimated gains to the developing countries represent a large portion of their annual receipts of official development aid. If tropical-zone and processed agricultural products were also taken into account, the potential gains would probably be much larger.
Another benefit would be the savings and convenience that would result from a higher degree of stability in world prices. Free trade allows supply fluctuations in different parts of the world to offset each other. Insular policies, on the other hand, increase the variability of prices. Policies in both industrial and developing countries have been important factors behind the high volatility of world market prices of agricultural commodities, especially sugar. In the case of wheat, for example, studies have indicated that the amplitude of international price fluctuations would be reduced by 33 to 48 percent if free trade policies were instituted.
The difficulties of coordinating domestic and trade policies in various countries have thus far prevented the multilateral liberalization of agricultural trade. Various ad hoc measures have been proposed or undertaken to increase the benefits that developing countries receive from commodity trade. Examples are international commodity agreements, compensatory financing mechanisms, special trade preferences, and food aid.
The results of some of these measures have, however, fallen short of expectations. The commodity agreements have had very little success in stabilizing prices, and only two—for coffee and rubber—are currently active. Trade preference schemes have provided only small transfers to developing countries and have not significantly increased their agricultural exports. On the other hand, compensatory financing schemes have provided valuable assistance despite the fact that their operations have remained limited in scale. Food aid has also been helpful, especially for famine relief. Getting the right commodities to the people most in need, without retarding the growth of agricultural production in recipient countries, has, however, proved to be difficult.
In any case, the potential benefits of these measures fall far short of what could be achieved through agricultural trade liberalization. Progress toward liberalization of domestic and trade policies in a coordinated manner should clearly have priority on the current agenda of the world community. The prospective GATT multilateral trade negotiations are the most promising way of achieving such liberalization at present.