Overview

Andrew B. Stoeckel, David Vincent, and Sandy Cuthbertson

Summary

There is an acute and persistent agricultural trade problem. Everybody knows that. Large costs to national budgets, chronic dependence of export industries on subsidies, and agricultural products wasting away in stockpiles are the visible evidence of policy failure. Yet little reform has occurred. At the Uruguay Round of multilateral trade negotiations, which commenced at Punte del Este in September 1986, the various ministers said the right things—"to halt and reverse protection and to remove distortions to trade." But already these declarations appear to have been cosmetic: so far the words speak louder than the actions.

Why is there so little progress towards the patently better set of policies available to governments? In nearly every case, national objectives for agriculture, whatever they might be, could be achieved by other policies at lower cost. The reason for the lack of reform lies in governments' lack of political will to tackle the issue. Political will changes as the public's perception of the issues changes. But the evidence of the costs of agricultural policies has not yet convinced people to demand reform from their governments. The evidence available now is largely based on the assessment of the initial effects of agricultural policies, assessments which tend to focus on what happens in the agricultural sector alone. But protection from agricultural imports and subsidies to agricultural exports result in taxes on other sectors. Surprisingly, there has been little effort to measure these costs to other sectors in an economywide framework. There has been no attempt to identify the full economywide benefits and costs of agricultural policies.

The troubles with agricultural trade are not an outcome of some unfortunate configuration of world events. They are the consequences of domestic policies. Logically, it is these domestic policies that have to change. In turn, the best way to enhance domestic policy is to appeal to self-interest. This can be done through greater public awareness of the real economic benefits and costs of policies. While virtually all the expert commentators on trade agree on this need for greater policy transparency, only rarely is it provided. This is the starting point for our project.

The studies making up this project show something well known to economists but not to the public at large: just measuring and reporting only the initial effects of a policy can be very misleading. The second-round and indirect effects may be much different, even to the point of reversing initial impressions. Instead of jobs being saved by agricultural protection they are being lost; instead of countries losing from liberalization they would gain. It is only when the second-round and indirect effects are taken into account that protection from imports is shown to be a tax on exports. So large has the sum total of these trade distortions become, distortions which go well past agriculture, that the macroeconomic performance of countries has been greatly impaired.

Ten separate studies by researchers around the world—all adopting an economywide view of the effects of agricultural liberalization—are reported in this volume. These studies cover the three largest industrial groupings—the European Community (EC), Japan, and the United States; the developing world considered as six types of economies; Korea as an example of a rapidly industrializing country; and Australia as an example of a country where there are large trade restrictions on sectors other than agriculture. While the orientation of each of these studies is domestic, the possibility that one country's policy might affect the choices of another is examined in a final study of what appears to lead the United States and the EC to engage in an agricultural trade war (table 1.1).

Industrial countries' liberalization of agricultural trade could have a dramatic effect on developing countries. Real incomes could increase by $26 billion. Significantly, classifying the developing world into six representative economies (following definitions of the World Bank), the Loo and Tower study finds that agricultural importing countries could gain from agricultural trade liberalization. There would be terms of trade losses for agricultural importers but these would be more than offset by improved efficiency of domestic resource use and a reduction in distorting taxes made possible by the real income and revenue increases. These good things would happen largely because, on average, agriculture tends to be taxed in each of the six types of developing economies. The taxed sector uses resources relatively more efficiently than the heavily subsidized industrial sectors. Indeed, one dollar's worth of transfer to developing countries from agricultural liberalization is equal to or better than two dollars' worth of direct foreign aid. Trade in this case is two times better than aid.

The gain to developing countries from industrial country liberalization would allow a substantial reduction in their public external debt, which, at present levels, is a problem for the major world banks. Liberalization would allow the poorest countries to reduce their external debt by 5 percent per year.

Developing countries, exporters and importers alike, have a strong interest in full agricultural liberalization. They should therefore get behind the U.S. proposals in the Uruguay Round of trade talks. The results of studies reported here show that if the industrial world really intends to assist developing countries, it should liberalize agricultural trade. Not only would world incomes grow, but debt burdens could be significantly reduced. That is, there are other ways to address debt problems of the third world besides debt for equity swaps and the like. Agricultural liberalization by the industrial world would accomplish more now, and in the long run, and should be a point of discussion for bankers, industrialists, and governments. An assessment of the impact of liberalization on developing countries must involve a consideration of what liberalization would do for the efficiency of domestic resource use. Just looking at the terms of trade effects misses the impacts that matter.

The two studies on the European Community—one incorporating the four largest members, the other based on West Germany—show large macroeconomic effects from the Common Agricultural Policy (CAP) and supporting national government programs. These support policies have changed the EC from the world's largest importer of temperate zone agricultural products to the second largest exporter. The main cost has been on manufacturing output and exports and, given the rigidity of real wages in the community, on unemployment.

Manufacturing output is lower in each of the four largest members of the EC—Germany, France, Italy, and the United Kingdom—as a result of agricultural policies. By increasing the cost of living, these policies reduce the competitiveness of manufacturing exports, and jobs are lost. If agriculture was liberalized, output could increase by over 1 percent for manufacturing, a very large sector in the EC. Manufacturing exports to the rest of the world could increase by 5 percent.

Germany is potentially now the single most important force for reform. It is the largest contributor to the CAP, it is the largest economy in the EC, and of late, it has strongly resisted change in agricultural policy. Total public subsidies to German agriculture are now well over 20 billion DM, about 70 percent of the sector's gross value added at domestic prices. Under liberalization, manufacturing sectors, apart from food, expand. For example, net exports of mechanical and electrical engineering would rise by over 12 percent.

The effects of liberalization on employment are large. The EC suffers from high and persistent unemployment. Its experience with unemployment is different from that of Japan and especially the United States, where substantial increases in employment have occurred in recent times. Agricultural liberalization, along with a 10 percent increase in world agricultural prices (assumed to follow liberalization of world agricultural trade), would lead to significant increases in employment: up to 850,000 additional jobs in West Germany for instance. A reduction in the rate of unemployment from 9 percent to 5 percent could occur. Agricultural liberalization of West Germany alone would have a very big impact—over 500,000 jobs. The results from the German component of the multicountry model suggest that about one million jobs could be gained there from EC agricultural liberalization. These findings on employment depend on the assumption of rigid real wages. It could be argued that some flexibility does exist. But it would be wrong to assume complete real wage flexibility, given the large and persistent unemployment problem in the EC. Even if real wages were somewhat flexible, unemployment could increase by up to 3 million persons for the EC-10 as a whole as a consequence of supporting agriculture. This figure is higher than earlier assessments because national government supports to agriculture, which are as large again as those out of the EC budget, have been included in this new set of studies.

Japan is the world's second largest economy and the world's largest importer of agricultural products. Yet it also has the highest barriers to agricultural imports of any OECD country. The Japan study measures the extent to which this protection to domestic agriculture taxes the economic performance of other sectors. The study by Vincent shows that if agricultural protection were removed, manufacturing export earnings would expand by about 3 percent over the short term and manufacturing employment would grow by up to 4 percent in some sectors. In addition, the removal of agricultural protection would result in a rise in average real wages in the Japanese economy by about 2.5 percent, equivalent to over 100,000 yen per worker in 1985. Rural land prices would fall by about 68 percent. The Japan study observes that the tax effect on aggregate export earnings from agricultural protection is as great as, or greater than, the savings on aggregate import expenditures from such protection. Agricultural protection does not explain Japan's trade surplus.

In Korea, as in Japan, manufacturing growth has placed large adjustment pressures on domestic agriculture. The government has attempted to offset this by escalating agricultural protection. The results of the Korean study indicate that, following domestic agricultural liberalization, aggregate exports would increase by about 4,500 billion won in 1987, employment in processing and manufacturing sectors would increase by about 580,000 jobs, and average real wages would rise by about 250,000 won per worker. Rural land prices would fall by about 40 percent. In both the Japanese and Korean cases, agricultural support policies amount to a partial undoing of manufacturing industry growth.

In the first of the U.S. studies, a two-period economywide model is used to investigate the effects on the U.S. budget and trade deficits of elimination of U.S. agricultural subsidies. What this could do to the real rate of interest is important to determining the responses of private savers and investors and hence the outcome for the balance of trade. A multiperiod framework must be used to capture these interest rate effects. The results suggest that the removal of agricultural subsidies involving direct budgetary savings of $58 billion over 1985 and 1986 would actually have saved $74 billion through its impact on the tax base. Because net domestic private savings are increased in this scenario, the trade balance improves by more than the value of subsidies eliminated. In 1986 the elimination of $31 billion of subsidies would have improved the trade balance by $42 billion. The elimination of agricultural subsidies would therefore make a major contribution towards reducing the current budgetary and trade imbalances in the U.S. economy.

The results of this study illustrate that unless the effects on real interest rates and hence private savings and investment are taken into account, the likely improvement in the trade deficit from agricultural liberalization will be understated.

In the second U.S. study a ten-sector economywide model is used to explore the macroeconomic and sectoral implications of U.S. agricultural trade liberalization over the longer term. The conclusions indicate that U.S. real GNP would increase as would domestic investment and consumption. The budget deficit would fall and employment and output losses in agriculture would be offset by increased economic activity and employment in manufacturing sectors, particularly those supplying investment goods.

The third U.S. study considers both the short- and long-term effects on the U.S. economy of a unilateral liberalization of U.S. agricultural policies. A feature of the economywide framework used in this study is its comprehensive treatment of relationships between different parts of U.S. agriculture. The results demonstrate how U.S. farm programs assist some parts of U.S. agriculture at the expense of other parts; also, by encouraging excessive use of scarce resources in agriculture, these programs have penalized the economic performance of other sectors. The United States as a whole would be unambiguously better off were it to unilaterally eliminate agricultural subsidies. Gains would accrue in the form of higher nonfarm output, reduced treasury outlays, higher real domestic income, and lower food costs to consumers.

The same study also concludes that the gains from agricultural trade liberalization are large enough to compensate U.S. farmers. This compensation could maintain the farmers' incomes at existing levels without the waste that occurs with current programs.

Agriculture can also be a "taxed" sector in the sense that other sectors, with which it competes for domestic resources, receive more assistance. This is the case for Australian agriculture, which receives relatively low assistance when compared with Australian manufacturing. The Australian study shows that when this taxation effect on agriculture (of assistance to manufacturing) is removed, agricultural output expands by about 3 percent over a two-year period, agricultural exports expand by up to 6 percent, and real net farm incomes improve by about 9 percent or $A2300 per farm in 1988.

The combined effect of the removal of assistance to agriculture and manufacturing is a fall in domestic prices, an improvement in international competitiveness, and an expansion in exports and imports. Real GDP, agricultural output, exports and incomes, aggregate employment, and the net revenue position of the government sector all expand.

The results of the Australian study highlight the pitfalls of using summary measures of agricultural assistance, such as producer subsidy equivalents (PSE), in international trade negotiations. In Australia, New Zealand, and many developing countries where agriculture is taxed relative to manufacturing, reducing agricultural assistance as part of a coordinated cross-country reduction in PSEs would lead to worse outcomes for domestic economic efficiency and real income. To achieve unambiguous gains in domestic economic welfare, assistance reform must begin at home and must apply to all sectors.

The final study in this volume examines the possible interactions of policies in the context of an agricultural trade war between the EC and the United States. In this paper, unlike all the others commissioned, self-interest is not defined independently of the possibility of strategic retaliation by other countries. Game theory is used to predict the outcome in each country, assuming both the United States and EC decide policies on the basis of expectations about each other's behavior.

Whether or not it is in the interests of countries such as the United States and EC to participate in a trade war is seen to depend entirely on the payoff to each that unilateral changes in protection would bring. Assuming that both the EC and United States can significantly influence, by their own actions, world agricultural prices, then both could gain by increasing border protection. Hence, an escalation of a trade war is a likely result. However, the weight of evidence from many studies shows that the payoff from unilateral increases in protection is negative. Certainly the results of all the other studies of this project show significant gains from unilateral liberalization. Under these circumstances a trade war would indeed be silly. Even so, the original question is valid: the actions of one country, under certain circumstances, could influence the choice of options of another.

In summary, the support to agriculture around the world imposes large costs on OECD economies and developing countries alike. These costs extend well beyond the direct and obvious budgetary cost and transfers from consumers. The reduction in manufacturing output, employment, and exports that occurs in countries heavily subsidizing their agriculture is most significant. A more rational use of resources in both industrial and developing economies would lead to substantial macroeconomic benefits—benefits large enough to have a bearing on contemporary macroeconomic problems such as the large U.S. trade deficit, the high unemployment in Europe, and the international debt problem. Microeconomic reform is just as important for the health of the world economy as changes in macroeconomic policies.