Agriculture in the Liberalization Process

Eight— Agriculture in the Liberalization Process
Eight—
Agriculture in the Liberalization Process
D. Gale Johnson
Liberalization of trade in agricultural products has a number of implications not commonly associated with liberalization of trade in manufactured products. Without attempting to order them in terms of significance or difficulty of achieving liberalization, several differences between trade liberalization for agricultural and manufactured products may be noted.
One difference is that trade liberalization for agricultural products means moving toward more market-oriented policies not only in international markets but in domestic ones as well. Though it is not as common for the manufacturing sector, virtually all countries have domestic policies that directly intervene in farm product markets. In the vast majority of these cases, trade policy becomes an adjunct of domestic agricultural policies. Thus it is commonplace for countries to argue that it is inappropriate to require that they negotiate about trade interventions that are required as a part of their domestic price support and subsidy policies. To permit negotiations on such trade policies would mean that domestic policies were subject to negotiation.
Another difference is that in the agricultural sector governmental interventions appear to be as likely to reduce farm product prices below international market levels as to increase the prices above them. Many countries follow a policy of low prices for consumers for important food products, either through lowering farm prices or by subsidizing the difference between producer and consumer prices. Thus liberalization of the agricultural sector is about as likely to result in higher producer and/ or consumer prices as in lower prices, and thus the interests affected by
― 284 ―
agricultural liberalization may take quite different positions than would be the case for liberalizing trade in most manufactured products, where the likely outcome is both lower producer and consumer prices.
Security of food supply is an argument often used as a rationalization for protecting domestic farm production. It is relatively easy to gain acquiescence, if not support, for maintaining a large domestic agriculture on the grounds that to do otherwise subjects the ordinary citizen to unacceptable and unnecessary risks of going hungry. Aside from the military sector, it is less easy to make similar arguments for nonfarm activities.
A final difference between agricultural and manufacturing policies is that in many countries, including all of the advanced industrial countries and the NICs (the newly industrializing countries such as South Korea and Taiwan), farm policies have significant income objectives for the farm population. The unavoidable fact is that as economic growth occurs, resources—especially labor—must be shifted out of agriculture into nonfarm sectors. This will generally result in somewhat lower income in farm areas than in urban areas. Most governments attempt to use price interventions to reduce such apparent income differentials, even though quite simple analysis indicates that such interventions have little or no long-run effect upon the returns to mobile factors of production, especially labor.
While this chapter concentrates on the issue of liberalization, it begins by providing an analytical framework for the overall role of agriculture in the development process. Though at unprecedented levels today, intervention in the agricultural sector is nothing new, and section 2 contains a historical summary of the worldwide pattern of agricultural protection. Looking at some of the studies that have been done, section 3 tries to explain why protection of the agricultural sector (both positively and negatively) is so prevalent, and why it varies so greatly from country to country. Section 4 then focuses on some of the institutional barriers to liberalization of this sector, and the chapter ends with a brief conclusion.
1—
Agriculture and Economic Development
At low levels of per capita income, most of a country‘s economically active population is engaged in agriculture. This is true in the low-income developing countries of the world today, just as it was true in the industrial countries in the not too distant past. In some industrial countries
― 285 ―
three-fourths of the economically active population were engaged in agriculture well into the present century. When such a large percentage of the population is required to produce a country‘s food and fiber, it is important that agricultural productivity increase. Increasing such productivity depends upon providing farm people with appropriate incentives to contribute the investments and effort needed to make both land and labor more productive.
A productive agricultural sector has three important contributions to make to economic growth and development. One is to provide an adequate supply of food and fiber at reasonable prices The second is to provide an export surplus so that foreign exchange will be available to provide for the import of capital goods and technology to modernize and expand the nonfarm sectors of the economy. The third is to increase productivity so that resources can be transferred to the rest of the economy. These transfers involve both labor and capital. In a rapidly growing economy there will be large and continuous transfers of labor from agriculture to urban areas. Eventually (as evidenced by the industrial countries today), the labor force in agriculture declines to less than 5 percent of the total labor force. The rate of the decline in the percentage of the labor force in agriculture is related both to the rate of productivity improvement in agriculture and the rate of growth of the nonfarm part of the economy. Where both changes occur at significant rates, agriculture‘s share of national employment can decline by one or more percentage points per year. This is approximately what occurred in upper-middle-income developing countries between I960 and 1980, when the percentage of the labor force in agriculture declined from 49 percent in 1960 to 30 percent in 1980 (World Bank 1982).
Capital transfers from agriculture to the rest of the economy are more difficult to document than is the labor transfer, yet it is generally agreed that such transfers do occur. How large the transfer will be depends upon institutional arrangements, such as the importance of landlords with large holdings who find greater expected profits from investments in industry than in agriculture, and the nature of the tax system. It is evident that over time agriculture‘s share of national investment declines.
The relative decline of agriculture as a source of national employment and output results from two important factors. One is that as real per capita incomes increase, the percentage of income spent upon farm products falls, declining from nearly 70 percent at income levels such as those prevailing in India to less than 20 percent in Western Europe and
― 286 ―
North America. The other factor is the increased productivity of agriculture. Despite the long-standing view that the supply of agricultural land is fixed and diminishing returns prevail, the growth of resource productivity in agriculture has in fact been sufficient to more than offset both. Through research, the use of nonfarm inputs such as fertilizer and machinery, and increased human capital, the agricultural-production frontier has been continuously pushed outward, more than compensating for whatever negative effect diminishing returns may have when the supply of land is limited.
The relative, and eventually absolute, decline of the farm labor force causes serious adjustment problems for rural communities, but the transfer of labor to the nonfarm economy makes it possible for farm people to participate in economic growth. If the transfer of farm workers to nonfarm jobs is inhibited, economic growth is slowed, and the disparities between farm and nonfarm incomes grow. Because labor must transfer out of agriculture as economic growth occurs, the returns to farm labor must be less than urban earnings to induce farm people to migrate. The relative earnings of farm people depend upon how easy it is for them to find nonfarm job opportunities.
A World Economic System for Agriculture
If there were free trade in agricultural products or even liberal trade regimes based upon low ad valorem tariffs, the world would produce its food at or near minimum cost. The arguments for liberal trade in farm products are the same as those for liberal trade in general: consumers everywhere would have access to a wide variety of foods at low cost, and efficient producers would find ready outlets for their production.
If the world‘s agricultural resources, both natural and human, were efficiently used, most present market interventions would have no place. Freedom of trade—domestic and international—is required if food is to be both abundant and low in cost and if farm people are to participate fully in economic growth. The levels of income of farm people depend primarily upon two things: the per capita incomes of the country in which they live and the access they have to nonfarm earning opportunities. Economic growth requires that farm employment decline, first relatively and then absolutely. Thus interventions that interfere with the smooth transfer of labor out of agriculture have an adverse effect upon the welfare of farm people.
However, not all government interventions or activities are inconsistent with domestic and international liberal trade. Many interventions
― 287 ―
are economically and socially productive—the education of farm youth, support of agricultural research and extension, and the creation and maintenance of the rural infrastructure (markets, highways, roads, and communication facilities) that lowers the cost of marketing farm products and helps to integrate farm people into the economy. These activities do not create distortions; they are basically public goods that cannot or will not be provided through the market.
In fact, of course, trade in farm products is subject to wide-scale governmental intervention. Some interventions increase farm output prices; others depress them. In the latter case, the price system is used to transfer income from farm people to the urban economy. The interventions can be measured through estimates of nominal protection coefficients (that is, estimates of the ratio of domestic farm prices to adjusted border prices, which reflect adjustments for marketing and transportation costs, for quality differences, and for overvaluation of the exchange rate). If the nominal protection coefficient is 1, then domestic and international prices for farm products are the same. If the coefficient exceeds 1, agriculture is receiving positive protection. If the coefficient is less than 1, domestic prices are below international market prices and agriculture is being taxed, or there is negative protection.
Subject to a number of qualifications that will be noted later, it is possible to indicate in a simple way the relationships between per capita income and the protection of agriculture. At low levels of per capita income, say below $1,000 per capita (1982 U.S. $), the protection coefficient is generally below 1. As per capita incomes approach $1,500 to $2,000, the protection coefficient becomes positive, and it increases gradually as incomes rise. As will be noted later, the explanation of the size of the nominal protection coefficients is complex and requires a considerable number of variables. The level of per capita income, or a proxy for it (the percentage of gross national product produced in agriculture), has a major influence on the size of the coefficient.
What Differences Do the Distortions Make?
The market distortions in agriculture have many effects. Some are in terms of income transfers; others lower productivity by encouraging inefficient use of resources.
Since the protection coefficients vary from country to country, the effects of the interventions upon the incomes of farmers and the welfare of consumers differ, depending upon the particular setting. Where the protection coefficient is positive, and the higher returns are achieved
― 288 ―
through raising market prices, income is transferred from consumers to farmers. When the protection coefficient is less than 1, and the lower domestic prices are achieved through export taxes, consumers have access to food at prices below world levels at the expense of farmers.
After a period of time the magnitude of the protection coefficients in excess of unity have little positive effect upon the returns to farm labor. The primary income effect of positive protection coefficients is to increase the price of land; the return to farm labor is determined primarily by the level of nonfarm wages and the access that farm people have to these jobs (Johnson 1973). Negative protection, on the other hand, may have serious adverse effects upon farm people. Low output prices (below international levels) will result in lower land prices than would prevail under free trade. And since negative protection occurs primarily in low-income developing countries, where rural areas are still generally isolated from urban areas, nonfarm jobs are not readily accessible to farm people. Access to such employment is also reduced by both limited education and restraints upon job availability in the formal sectors of the urban economies.
Agricultural price stability is often considered a desirable policy goal, and domestic prices can be stabilized in one of two ways: by varying imports and/or exports so that domestic demand and supply are made equal at a predetermined price, or by variations in stocks of particular commodities that absorb significant variations in world demand and supply. The first approach utilizes quantitative import quotas, or variable levies and variable export subsidies that make international market prices irrelevant to domestic producers and consumers. (Examples are the import quotas on dairy products and sugar in the United States; quotas for beef and oranges in Japan and state trading in Japan; and the variable levies on imports and subsidies on exports of the European Community‘s Common Agricultural Policy.) This form of protection affects the variabilty of international prices. Governments that stabilize domestic prices for their producers and consumers by adjusting international trade do so at the cost of increased instability of world market prices (Johnson 1975). The reason for this effect is, first, that when domestic prices are stabilized, neither consumers nor producers are required to adjust to domestic variations in demand and production, and second, that the markets in such countries do not respond to changes in production and demand in the rest of the world, since imports and/or exports respond not to external prices but only to variations in local demand and supply.
― 289 ―
The other method of domestic price stabilization need not lead to international price instability. By being willing to buy and sell grains, for example, at reasonably fixed and stable prices, and then storing any excess of purchases over sales, or selling from the large stocks that were accumulated, domestic support prices can still be established in excess of equilibrium prices—if this is the policy objective. (The United States and Canada have both used this technique.)
If the argument for high and stable prices of food is to provide for national food security (often defined in terms of self-sufficiency), then almost any level of prices and incentives can be justified.
Market interventions for agricultural products are also often used to provide high levels of protection for the processing of agricultural products. As will be discussed later, seemingly modest levels of tariffs on semiprocessed products—say, 10 percent—may lead to effective protection rates of 100 percent or more if the raw farm product is imported free.
With the high rates of effective protection often provided for the processing of farm products, developing countries have placed export taxes on agricultural raw materials in order to encourage processing in the country where the raw product is produced. The high protection of processing in developed countries and the counterimposition of export taxes by the developing countries have two adverse effects. The first is that processing will not be done where it can be done at lowest cost; in fact, with the import duties and the export taxes, there is no way that the market can reveal where processing can be done at lowest cost. The second effect is that the farmers in the developing countries definitely lose through bearing the cost of the export tax, which often goes to support a relatively high-cost domestic processing industry. Once processing in the developing country is made possible by an export tax on the raw product, the finished products face import duties in the developed countries. Thus the farmers in the developing countries must pay for the processing wherever it is done.
2—
Patterns of Agricultureral Protection
Some Historical Background
It is important to note that significant protection of agriculture is nothing new. There has been but a brief interlude in modern history when there was something approximating free trade in agricultural products in Europe and North America. For Europe this period began with the abolition of the Corn Laws by the United Kingdom in 1846. Starting in
― 290 ―
1860 there was free trade in agricultural products in Western Europe. However, free trade was maintained for less than two decades—Italy reintroduced protection in 1878, Germany in 1879, and France in 1881. Only Denmark and the Netherlands, in addition to the United Kingdom, resisted the move to protection until the Great Depression of the 1930s.
The British Corn Laws, as they evolved over at least four centuries, were a complex set of laws, regulations, and interventions in domestic and international markets. They included every protective device in use today, save one. Included were variable import levies, quantitative restrictions upon imports (including seasonal quotas), export subsidies, export taxes, and, of course, tariffs. There were also special preferences for the use of British ships, a precursor of the U.S. cargo preference rules for farm products. The only twentieth-century innovation I have discovered is the use of direct payments to producers to meet a price or income commitment. Even this device is but a slight modification of various bounty schemes that have existed, at least briefly, in a number of countries.
The abandonment of free trade in farm products in Western Europe has been attributed to the Franco-German war of 1870–71 and the Great Depression of the last quarter of the nineteenth century. However, protection of agricultural products in Western Europe prior to World War I was quite modest compared to recent levels.
In many countries, primarily low-income developing countries, governments intervene to reduce the prices of agricultural products. But even negative protection of agriculture is not new. The British Corn Laws included provisions for an export tax on grain when internal prices exceeded a certain level. However, there is no historical precedent for the wide prevalence of negative protection extending over long periods of time such as prevails in many developing countries today. The difference in degree is great indeed.
Agricultural Protection Rates before 1940.
Estimates of the levels of tariff protection in Europe exist for 1913, 1927, and 1931 for three broad groups of products: foodstuffs, semimanufactures, and manufactures.[1]
― 291 ―
For the larger European countries, the tariff levels for agricultural products were generally in the range of 20 to 30 percent (see table 8.1).
The upper group of six countries in table 8.1 were industrial countries with significant imports of agricultural products. The bottom group consists of countries that as of 1913 were exporters of timber products (Sweden and Finland) and timber, agricultural, and other raw products, and were only modest importers of agricultural products. Countries in this second group had much higher rates of protection for agricultural products than did the countries in the upper group.
Protection of foodstuffs changed very little between 1913 and 1927 for the countries in table 8.1. Except for Belgium, there were sharp increases in protection between 1927 and 1931. Remember that three European countries—the United Kingdom, Denmark, and the Netherlands—had free trade in agricultural products in 1913 and until the early 1930s.
The United States did not protect its agriculture until 1890, and then protection remained at a relatively low level until the Smoot-Hawley tariff of 1929. Japan was nearly isolated from world trade prior to 1868 and did not move to a free-trade position for agriculture as it opened up to the rest of the world. South Korea was a part of the Japanese empire from 1910 to 1945, as was Taiwan for approximately the same period, and both were subject to about the same degree of protection from non-Japanese sources as was Japanese agriculture.
The earliest available Japanese rice price data are for mid 1880s. Comparisons of Thai export prices with Japanese producer prices indicate that in the 1880s there was a modest degree of protection for rice (table 8.2). For the five years centered on 1885, the producer price in Japan exceeded the Thai export price by 34 percent. Because of quality differences between Thai and Japanese rice, the 34 percent difference should not be taken too literally. However, comparisons of the same data series for the next half century indicate a sharp increase in the ratio of Japanese producer prices to the Thai export price, reaching a ratio of approximately 2.5 in 1920. The ratio declined to 1.8 in 1930 and then rose to 2.9 in the late 1930s.
The available price data indicate that the Korean export price was essentially the same as the Japanese producer price from 1915 to 1935.
― 292 ―
Table 8.1 Estimated Tariff Levels in Europe, 1913, 1927, and 1931
(in % of export prices, European products only)
Foodstuffs
Semimanufactured Goods
Manufactured Goods
1913
1927
1931
1913
1927
1931
1913
1927
1931
Germany
21.8
27.4
82.5
15.3
14.5
23.4
10.0
19.0
18.3
France
29.2
19.1
53.0
25.3
24.3
31.8
16.3
25.8
29.0
Italy
22.0
24.5
66.0
25.0
28.6
49.5
14.6
28.3
41.8
Belgium
25.5
11.8
23.7
7.6
10.5
15.5
9.5
11.6
13.0
Switzerland
14.7
21.5
42.2
7.3
11.5
15.2
9.3
17.6
22.0
Austria a
29.1
16.5
59.5
20.0
15.2
20.7
19.3
21.0
21.0
Czechoslovakia a
36.3
84.0
21.7
29.5
35.8
36.5
Sweden
24.2
21.5
39.0
25.3
18.0
18.0
24.5
20.8
23.5
Finland
49.0
57.5
102.0
18.8
20.2
20.0
37.6
17.8
22.7
Poland b
69.4
72.0
110.0
63.5
33.2
40.0
85.0
55.6
52.0
Romania
34.7
45.6
87.5
30.0
32.6
46.3
25.5
48.5
55.0
Hungary a
31.5
60.0
26.5
32.5
31.8
42.6
Yugoslavia
31.6
43.7
75.0
17.2
24.7
30.5
18.0
28.0
32.8
Bulgaria
24.7
79.0
133.0
24.2
49.5
65.0
19.5
75.0
90.0
Spain
41.5
45.2
80.5
26.0
39.2
49.5
42.5
62.7
75.5
SOURCE: Liepmann 1938, 413.
a In 1913 Hungary and the lands that subsequently formed Czechoslovakia were components of the Austrian Empire. The figures for Czechoslovakia and Hungary in 1913 are thus included in those for Austria.
b In 1913 most of the subsequent (pre-World War II) Polish state was part of the Russian Empire. Figures for 1913 Poland are thus equivalent to the Russian figures.
― 293 ―
Table 8.2 Estimated Protection of Rice in Japan,
1885 to 1938–39
Year a
Producer Price
(yen/kg)
(1)
Thai Export Price
(yen/kg)
(2)
Protection
Coefficient
(1) / (2)
1885
3.8
2.8
1.34
1890
4.5
2.9
1.54
1895
6.4
3.4
1.85
1900
8.6
4.9
1.77
1905
10.0
5.8
1.74
1910
11.5
7.3
1.57
1915
11.9
5.4
2.20
1920
26.7
10.7
2.49
1925
24.7
11.3
2.19
1930
15.8
8.9
1.78
1935
19.7
8.9
2.21
(1938–39)
27.5
9.5
2.89
SOURCE: Saxon and Anderson 1982, 5.
a Five-year averages centered on year shown except for 1938–39, which is a two-year average.
Since after 1920 rice from both Korea and Taiwan entered Japan without duty, the similarity in prices in the three areas is explained.
Agricultural Protection Levels since 1950.
Estimates of the degree of protection for agriculture at the end of the first decade after World War II and in the early 1960s for major Western European economies are given in table 8.3, along with similar estimates for the United States and Japan. The rates of protection for the 1950s and early 1960s now seem to be quite moderate, even in the case of Japan.
The estimates of the average protection rates for agriculture in tables 8.1 and 8.3 are not directly comparable, but the rates of protection in Western Europe in 1913 and the mid 1950s probably differed rather little. The Netherlands and Denmark remained virtual free traders, and Belgium had a very low level of protection. The United Kingdom, however, had adopted a protectionist stance with respect to producers, though not to consumers. The consumer had access to most foods at international market prices; farmers received protection through deficiency payments.
In the decade following 1956, the degree of protection of the Western European countries increased significantly, except for Ireland and Denmark. Prior to the formation of the European Community, the original
― 294 ―
TABLE 8.3 Estimated Price Support Costs
as a Percentage of Gross Agricultural Output
1956
1965–67
West Germany
22
54
France
18
47
Italy
16
64
Belgium
5
54
Netherlands
5
37
EEC (original six)
(16)
52
Ireland
4
3
United Kingdom
32
28
Denmark
3
5
Sweden
27
54
Japan
42a
76b
United States
2c
8d
SOURCES: McCrone 1962, 51; Howarth 1971, 29: and Saxon and Anderson 1982, 29. The McCrone and Howarth estimates have been adjusted to measure protection in terms of international prices instead of domestic prices. Sec Honma and Hayami 1984, table 1 (no page number).
NOTE: Valuation is at international market prices.
a 1955–59.
b 1965–69.
c 1955.
d 1965 only.
members had an average rate of protection of 16 percent; a decade later in the Community the average was 52 percent. The creation of the EC clearly increased the average rate of protection.
A comparison of the rates of nominal protection for agricultural products in the East Asian economies of Japan, South Korea, and Taiwan given in table 8.4 may be of interest. There are a number of parallels in the protection coefficients. Each country had relatively low levels of protection in the late 1950s, with South Korea and Taiwan having negative rates of 15 and 21 percent respectively. The rates of protection increased rapidly in Japan, reaching more than 100 percent by the early 1970s. In South Korea the level of protection remained modest until the early 1970s and then rapidly moved to the high Japanese level before the end of the decade. Taiwan followed a very different pattern during the period. The level of protection in Taiwan was modest through the early 1970s, gradually increasing until reaching the Japanese level of the late 1950s and the South Korean level of the early 1970s, but remaining much below the current protection levels of the other two.
― 295 ―
In just two decades South Korea changed from a slightly negative rate of protection to one that is among the highest in the world. Later I shall indicate possible explanations for changing patterns of protection and why there has been an increase in protection everywhere.
Generally, the estimates of protection have included only the added costs to consumers—basically the difference between domestic and international prices multiplied by the amounts consumed domestically. Costs to taxpayers have not been included. Failure to include taxpayer or governmental costs may distort the comparisons somewhat.
Estimates of the total costs of agricultural protection for the United States and the European Community as of 1979–80 are presented in table 8.5 on as comparable a basis as seems possible. This table gives the value of domestic human and industrial use at world prices, and the excess costs imposed upon consumers by the difference between domestic and world prices. For the EC the excess consumer costs were 45 percent of what consumer costs would have been under free trade. For the United States the excess consumer costs were 12 percent. Taxpayer costs were 20 percent of the world value of domestic use of EC farm output and 7.2 percent in the United States. Thus the total costs—consumer plus taxpayer—implied a cost of protection of 65 percent for the EC and 19 percent for the United States in 1979–80.
Fluctuations in exchange rates such as have occurred over the past few decades can have major effects upon agriculture and on the effects of particular agricultural policies. If real exchange rates increase, agricultural exports are adversely affected—as occurred from 1981 to 1985 in the United States. The EC‘s Common Agricultural Policy was financially viable after 1980 owing to the declining value of the European Currency Unit (ECU) in terms of the dollar. In 1980 the exchange rate was $1.39 per ECU, but by 1982–83 the exchange rate was $0.89 per ECU. This change meant that world market prices denominated in ECUs rose and the cost of export subsidies declined. Consequently, the EC could maintain its price supports at high levels during these years; had the value of the ECU not declined by more than a third, the financial costs would have increased substantially, perhaps to an untenable level.
Declines in the real exchange rate can encourage a country‘s exports of farm products, as was the case in the late 1970s and early 1980s in the United States. Such a growth in exports can result in unrealistic expectations concerning the prospective demand for exports. In other words, fluctuations in real exchange rates result in added uncertainty and may
― 296 ―
Table 8.4 Estimated Nominal Rates of Agricultural Protection in Japan, South Korea, and Taiwan
(%)
1955–59
1960–64
1965–69
1970–74
1975–79
1980–82
Japan
Rice
50 (51)
72 (62)
99 (69)
160 (116)
263 (207)
249 (204)
Wheat
37 (36)
62 (33)
97 (28)
127 (21)
276 (31)
278 (26)
Barley
39 (37)
66 (35)
101 (26)
129 (15)
312 (32)
399 (31)
Soybeans
37
43
53
124
173
287
Beef a
113
142
165
146
284
181
Pork
- 11
27
23
26
16
3
Chicken a
-12
27
21
23
20
5
Weighted average b
44
68
87
110
147
151
South Korea
Rice
- 14(- 12)
- 9 (- 6)
6 (4)
55 (46)
138 (130)
154 (163)
Wheat
- 22 (- 5)
- 8 (- 20)
18 (19)
16 (- 2)
47 (6)
128 (23)
Barley
- 14 (- 4)
7 (10)
- 6 (- 6)
35 (20)
77 (11)
77 (11)
Corn
__c
31
17
43
67
101
Soybeans
- 23
5
51
63
109
226
Beef
3
5
55
88
281
326
Pork
-11
- 5
82
111
113
208
Chicken
- 27
7
132
103
153
140
Weighted average b
-15
-  5
9
55
129
166
Taiwan
Rice
- 31 (- 28)
- 8 (- 5)
- 13 (- 8)
4 (11)
58 (64)
144 (162)
Wheat
48(50)
25 (16)
39 (30)
32 (15)
57 (- 2)
92 (0)
Barley
15 (—)c
73 (—)c
67 (—)c
33 (—)c
49 (- 18)
99 (- 3)
― 297 ―
Table 8.4 Estimated Nominal Rates of Agricultural Protection in Japan, South Korea, and Taiwan
(%)
1955–59
1960–64
1965–69
1970–74
1975–79
1980–82
Corn
2
21
37
29
51
91
Soybeans
69
47
37
13
16
56
Beef
-4
8
20
37
162
Pork
15
32
40
38
13
3
Chicken
- 50
- 2
21
27
29
36
Weighted average b
- 21
2
2
18
36
55
SOURCE: Tyers and Anderson 1984, 8.
NOTE: Defined as the percentage by which the domestic price exceeds the border price. The producer price is used in the case of grains and soybeans and so underestimates the rate of protection by the producer-to-wholesale marketing margin. The numbers in parentheses show the extent to which wholesale prices for food grains exceed border prices. The wholesale price is used for meats, which again provides an underestimate of the rate of protection to the extent of any producer subsidies. Thus these estimates are slightly biased downwards. In cases where no import or export price data are available because of lack of trade, proxy border prices have been used (for example. Hong Kong unit import values for some livestock products in the absence of Korean import data).
a The rates of protection to Japanese meat production (particularly pork and chicken) may be seriously underestimated, since they are based on the assumption that the average quality of domestic meats is the same as the extremely high quality of meats that Japan imports. Producer prices for cattle, pigs, and chickens in Japan in 1977–79, for example, were 5.8, 2.3, and 1.6 times producer prices in Australia (cattle) and the United States (pigs and chickens) according to FAO sources.
b Averages are derived using weights based on domestic production of the commodities listed valued at border prices.
c Not available.
― 298 ―
TABLE 8.5 U.S. and EC Taxpayer and Consumer Costs of Farm Price
Support and Income Stabilization Programs, 1979–80
United States
European Community
Million U.S. $
%
Million U.S. $
%
Value of domestic human and
industrial use at world prices
69,939
69,939
Excess consumer costs as % ofx
value at world prices
8,406
12
28,858
45
Taxpayer cost as % of value at
world prices
3,833a
5
12,945b
20
Total costs as % of value at world
prices
12,239
17
41,803
65
a Includes only direct expenditures on price support and supply management. Does not include expenditures for food and nutrition programs.
b Does not include expenditures by member states.
result in farmers and governmental officials permitting themselves to be misled about future prospects.
Our discussion of agricultural protection has thus far emphasized the established industrial countries and rapidly growing developing countries such as South Korea and Taiwan. The data on nominal protection indicates that during the 1950s and into the 1960s the latter two countries had negative nominal protection coefficients—domestic prices were lower than world market prices. On the assumption that they will spur urban and industrial activity, negative rates of protection are in fact common in low-income countries, one might say all too common. Data on 120 commodity protection coefficients for both high-and low-income countries show that there were 54 with positive coefficients. 59 with negative protection, and 7 with domestic prices equal to international prices (Miller). All of the commodities with negative protection were found in low-income countries.
Table 8.6 presents data on protection coefficients calculated by Binswanger and Scandizzo (1983) for fourteen developing countries. While I have biased the selection slightly by excluding two members of OPEC, it is worth noting that out of 57 coefficients, only 5 indicate positive levels of nominal protection. There were 13 instances in which the negative protection was 50 percent or greater—the price received by farmers was less than half the international price.
This review of the historical development and the current status of agriculture protection clearly shows that governmental intervention in agricultural markets has been, and is, pervasive. In fact, it is now quite
― 299 ―
TABLE 8.6 Protection Coefficients for Low-Income Countries
Cocoa
Coffee
Cotton
Groundnuts
Maize
Rice
Wheat
Malawi
—
—
0.72
0.62
1.50
—
—
Mali
—
—
0.48
0.48
—
—
—
Senegal
—
—
0.65
0.57
—
0.70
0.70
Sudan
—
—
0.48
0.53
—
—
0.95
Zambia
—
—
—
0.58
0.62
—
—
Pakistan
—
—
0.61
—
0.90
0.68
0.76
Thailand
—
—
0.91
—
0.81
0.58
—
Togo
0.30
0.26
0.57
—
—
—
—
Ivory Coast
0.39
0.43
0.76
0.73
0.83
0.97
—
Egypt
—
—
0.44
—
0.67
0.34
0.76
India
—
—
—
—
0.80
0.65
0.80
Brazil
—
0.43
0.65
—
0.87
0.57
—
Kenya
—
0.74
0.85
—
0.91
1.30
1.13
Philippines
—
—
—
—
0.72
0.73
—
Cameroon
0.32
0.40
0.62
—
—
0.50
—
Ghana
0.05
—
—
—
—
0.06
—
Turkey
—
—
0.83
—
1.20
1.50
0.94
SOURCE: Binswanger and Scandizzo 1983, 20–23.
NOTE: The coefficient of less than unity means that protection is negative; farmers receive less than the international market price. A coefficient greater than 1 means protection is positive. A coefficient of 0.72 means that the farm price is 28 percent less than it would be if farmers were permitted to export their product without governmental intervention. A coefficient of 1.30 means that the domestic farm price exceeds the international market price by 30 percent. The international market prices are adjusted to reflect potential farm gate prices by reflecting marketing and transportation costs and, where possible, quality differentials.
clear that it is inappropriate to always describe the interventions as protection—at least as that term is normally used. In far too many cases, governments intervene to reduce prices received by farmers.
There is a way to reduce consumer prices without necessarily resulting in intervention adverse to farmers. This is the policy of reducing prices paid by consumers not only below international market prices, but also below the prices actually paid to farmers. Such policies are common in the Centrally Planned Economies (CPEs), with the USSR having the most expensive such effort. There, meat and milk prices have remained constant in nominal terms since 1963, and some other prices since the mid 1950s. As of 1989 all of the Eastern European CPEs and China had significant consumer subsidies on food products. Some low-income countries such as Egypt and Sri Lanka have subsidized consumer prices or have held food prices at low levels through periods of greatly overvalued currencies (World Bank 1982).
― 300 ―
In light of the foregoing discussion, it is clear that liberalization of agricultural markets would have quite different effects from country to country, and within countries. Let us now turn to what we know about why there is such a wide array of interventions in agriculture.
3—
Why Agricultural Protection Varies
The information contained in tables 8.1 and 8.6 reveals enormous variation in the degree of protection. The rates of protection vary widely by country and by commodity within countries. In recent years there have been a number of studies that have attempted to explain the variations in protection rates. While much remains unexplained, some important results have been obtained.[2]
The studies of protection that I summarize can be said to assume that there is a market for protection. There is a demand for protection, which may be strongest when agriculture loses its comparative advantage. There is a supply of protection as the politicians respond to effective political power. As the studies indicate, effective political power is not measured by the number of farmers or their relative importance in the population, but rather by certain characteristics that permit them to press their demands upon politicians effectively and obtain a response. In these studies an implicit assumption is that the actors in the political process are rational in pursuit of their objectives and that they use the resources available to them in an effective manner. Thus politicians supply protection as one part of their effort to maintain their position of influence and authority or, to put it more crassly, to stay in office. The motivations influencing the supply of protection appear to be similar for various forms of political authority.
Binswanger and Scandizzo (1983) tested a model in which they included nominal protection coefficients for a cross section of 151 commodities in 33 countries. The protection coefficients ranged from the highly negative (0.44 for cotton in Egypt) to the highly positive (2.81 for wheat in Japan). Both low-income and high-income countries were included; however, four major exporters were not included—Australia, Canada, New Zealand, and the United States. In Binswanger and Scandizzo‘s analysis, country characteristics (per capita income, agriculture‘s share of GDP, farmland per capita) and dummy variables for tropical beverages, exportable commodities, and importable commodities were
― 301 ―
included. Regressing the logarithms of the nominal protection coefficients on the variables resulted in R-squares of 0.43 to 0.48, depending on the particular regression. Per capita income, agriculture‘s share of GDP, and farmland per capita, and the tropical beverage dummy were important in explaining the protection coefficients.[3]
Honma and Hayami (1984) undertook to explain differences in agricultural protection levels in industrial countries, utilizing data for ten countries and six periods between 1955 and 1980. Their analysis was intended to determine if they could explain both differences over time and between countries in the amount of protection afforded agriculture. They included the following variables: a measure of comparative advantage, the share of agriculture in the labor force, the international terms of trade between agricultural and industrial commodities, and two dummy variables—one for the formation of the European Economic Community and its Common Agricultural Policy (CAP) and the other for a country maintaining self-reliance (food self-sufficiency) to support a policy of neutrality. This dummy variable was positive for Sweden and Switzerland. Two measures of comparative advantage were used. One was the ratio of labor productivity in agriculture to labor productivity in the entire economy. The U.S. 1975 value of the ratio was set equal to 100 and the ratios for other countries in specific years were compared to that ratio; the U.S. ratio for other years was also indexed to 1975 as a base. The other measure of comparative advantage was the ratio of the average amount of farm land per farm worker to the average capital endowment per worker in the entire economy. Lacking a measure of the capital endowment per worker, the average per capita GDP in 1975 constant prices converted into U.S. dollars by purchasing power parities was used. Again the U.S. value in 1975 was set equal to 100.
The following hypotheses were supported by the empirical analysis:
(1) In the process of economic growth, the comparative advantage of agriculture declines or shifts from agriculture to industry. This shift increases the demand for protection of agriculture.
― 302 ―
(2) The relative contraction of the agricultural sector (as measured by agriculture‘s share of GDP) makes it easier for farmers to organize to press their interests (owing to greater ease of organizing smaller numbers of farmers) and the reduction in the burden upon the nonfarm population of a given degree of protection.
Tracy Miller, in his Ph.D. dissertation (1985), applied interest group theory to the analysis of differences in rates of protection. He undertook to explain two things. One was whether nominal protection coefficients were negative or positive. If domestic prices were below border prices, nominal protection coefficients were negative; if domestic prices were exactly equal to border prices, the coefficient was zero and if the domestic prices exceeded border prices, the coefficients were positive.
The other aim was to explain the size of the nominal protection coefficients. Miller‘s model was quite robust in explaining the magnitude of the nominal protection coefficients, since he applied it to all commodity-country combinations, including both negative and positive protection coefficients. In this summary, only the results based on nominal protection coefficients adjusted for overvaluations of currencies (the adjusted nominal protection coefficients) will be presented.[4]
In explaining the sign of the protection coefficient, the most important variable was national per capita gross domestic product. With the addition of one of two variables, the correct classification of the nominal protection coefficients occurred 82 percent of the time (there were more than 100 coefficients in the analysis). One additional variable was exports of agricultural products per capita; the other was the amount of agricultural land per capita. In the analysis, imports are counted as negative exports. Given the level of per capita GDP, "The smaller the imports or the greater the exports of a commodity, the more likely that it will be taxed and the less likely that it will be subsidized" (Miller 1985, 82). The amount of agricultural land per capita had a negative coefficient in the model; that is, the less land there is per capita, the more likely it is that the protection coefficient will be positive. With little land per capita, a country is likely to be a net importer of agricultural products.
― 303 ―
As these studies indicate, there seems to be little mystery as to why in some economies (generally the more developed ones) agriculture is taxed and in others (usually LDCs) it is subsidized. At least there seems to be considerable agreement concerning the country and commodity characteristics associated with the differences. Differences in either agriculture‘s share of GDP or per capita GDP, in net exports per capita, and in land per capita seem to explain most of the observed differences. The interest-group analysis as applied by Miller adds something to the explanation in terms of specifying what interests are affected by the degree of protection and what variables influence whether the various groups will find it worthwhile to organize and push their interests in the political process.
Robert H. Bates presents a picture of governmental policies toward agriculture that is generally consistent with the results just presented. Based on his studies of Africa, he argues that as the number of large farmers increases, "the farming community will tend to grow politically more assertive." Bates adds:
Countries with greater numbers of larger farmers will tend to have agricultural policies that offer more favorable prices to farmers. The Ivory Coast and Kenya are cases in point. Planters, large farmers, and agribusiness in the two countries have secured public policies that are highly favorable by comparison with those in other nations. Elsewhere the agrarian sector is better blessed by the relative absence of inequality. But it is also deprived of the collective benefits which inequality, ironically, can bring. (Bates 1981, 95)
Bates summarizes his results by indicating that there were three factors that were important in determining the degree and nature of governmental intervention in the agricultural markets: the nature of the product (whether an export crop or a food crop), the structure of production (large versus small farms), and the degree of relative advantage that producers have in the production and marketing of the crop. The last point implies that the stronger the comparative advantage of a particular crop, the greater the potential for exploitation through low prices, since the high degree of relative advantage means farmers can continue to produce at very low prices for a considerable period of time.
The role of the structure of production is related to the political influence of large farmers relative to many small farmers. "Large farmers . . . often possess close social and political ties with governing elites. . . . One consequence is that crops whose production is dominated by large farmers tend to be less heavily taxed" (Bates 1981, 126). This
― 304 ―
result occurs because it is easier for a small number of large farmers to organize to protect their interests than it is for a large number of very small farmers. It is this phenomenon that explains the increased political influence of farmers in industrial countries as the number of farmers declines. Miller‘s analysis gives weak supports for this conclusion, but the reason for the weak support may be the large negative correlation between the number of farmers and the level of GDP per capita or agriculture‘s share of GDP.
Protection of Processing of Farm, Products
Not only is agricultural protection divorced from general trade policies, but similar differentiation often applies to the processing of agricultural products. Countries that import raw agricultural products that they do not produce often have zero tariffs on such products but have positive tariffs on processed product. Similarly, in response to the importers‘ actions, exporters often impose an export tax upon the raw product but no tax upon the products derived from it. Soybeans can be used to illustrate both phenomena: Brazil has had an export tax on soybeans of approximately 10 percent, with nil or much lower export taxes on meal and oil derived from the soybeans. As an importer, the EC permits the free entry of soybeans, but imposes duties on the importation of oil and meal. In both instances, the rate of effective protection of the processing activity is very high—probably on the order of 100 percent. And it is the Brazilian soybean producer that pays for the protection of the processing sectors.
Both the EC and the United States have used subsidies to increase their exports of flour. Flour milling is not exactly a high-technology undertaking and is performed in many economies at much different levels of income and technology. Except to serve the interests of domestic flour millers, there is no reason industrial countries should subsidize the export of flour at a rate that differs from the subsidy, if any, on wheat.
Examples of the differences in LDC export taxes and developed country tariffs on agricultural products by the degree of processing are given in table 8.7. These data are for the early or mid 1970s, but the Tokyo Round did little to change the tariffs and obviously had no influence on the LDC export taxes.
To some degree the structure of LDC export taxes is a response to the structure of developed country import tariffs. If the industrial
― 305 ―
TABLE 8.7 Average LDC Export Taxes and DC Tariffs
(in %)
LDC Export Tax
DC Tariffs
Copra
6.0
0
Coconut ml
4.0
10.1
Natural rubber
6.3
0
Rubber articles
0
7.6
Cocoa beans
26.5
3.2
Cocoa butter and powder
5.6
10.2
Raw cotton
11.9
.7
Cotton yam and fabric
0
9.9
Raw wool
11.8
1.3
Wool yam and fabric
0
11.4
Hides and skins
23.4
0
Leather
13.3
6.9
Logs
11.3
0
Sawn logs
4.0
1.1
Coffee beans
30.0
3.3
Soluble coffee
6.0
8.9
SOURCES: The average tariffs were in all but one case (coffee) calculated from Yeats‘s 1976 tabulation. For coffee, the tariffs were obtained from tariff tables for the EEC, the United States, and Japan. The export-tax information was obtained from a U.S. government interagency report on export restrictions (United States 1976). Export taxes and tariffs were averaged across countries using weights based on 1973 trade flows.
NOTE: The above table and accompanying information originally appeared in Golub and Finger 1979, 560.
countries have no tariff on a raw product, such as copra, and 10 percent duty on coconut oil, the LDC that produces the copra is precluded from extracting the oil from the copra unless the LDC imposes a tax on the export of the raw product or controls the amount of the raw product exported.
Golub and Finger have presented estimates of the gain in LDC export revenue if the escalation of tariffs by the degree of processing were abolished in the industrial countries. For eight commodities as of the mid 1970s the elimination of developed country tariffs on processing would have resulted in an annual increase of LDC export revenue of $1.6 billion. If both the tariffs and export taxes were eliminated simultaneously, the gain in LDC export earnings was projected to be $1.2 billion. The authors compare this gain in annual revenue to projected increases from
― 306 ―
the Generalized System of Preferences of approximately $500 million (Golub and Finger 1979, 573).
The magnitude of the barrier confronting the LDC processor of raw farm products from what appear to be very low rates of tariff turns out to be surprisingly high. Assume that the industrial country permits the import of a raw product, such as soybeans, free of duty, but has a tariff on oilmeal and oil at the seemingly low rate of 10 percent. If the raw product accounts for 90 percent of the value of the two processed products, the effective protection of the processing turns out to be in excess of 100 percent. Assume that soybeans have a landed price of $250 per ton and that the value of the oilmeal and oil derived from a ton of soybeans is $275 per ton. This means that processing adds 10 percent or $25 per ton to the value of the soybeans. For the present example, it is assumed that the $25 per ton is the value added from processing; actually, the value added will be significantly less than this because of the costs of power and certain materials required for proper handling of the processed products. As noted, the tariff on the products processed from soybeans is just 10 percent. A developing country that attempted to export the oilmeal and oil from a ton of soybeans would have to pay a tariff of $27.50 per ton of the original raw materials. The tariff paid would be more than the value added in processing, and the effective protection of the processing would be a minimum of 110 percent. This is a remarkable effect from a seemingly modest tariff of 10 percent. It also explains why much of the processing of agricultural products occurs in industrial countries.
Alexander Yeats has compared the nominal and effective rates of protection for several processed farm products. A few examples will suffice. The EC had an 11 percent tariff on soybean oil; Yeats estimates that the rate of effective protection—the protection of the value added in processing—was 148 percent. For the same product, the Japanese had a tariff rate of 25.4 percent, which generated an effective protection rate of 268 percent. Somewhat less extreme, there was a Japanese tariff of 7.2 percent on palm kernel oil, which gave an effective rate of protection of 49 percent. The United States had a tariff rate of a mere 2.6 percent on cocoa powder and butter, but the effective protection turned out to be 22 percent (Yeats 1981, 6–7).
In the industrial countries farmers have an ally in the firms that process agricultural products, especially from raw farm products that are imported. It is unlikely, however, that most farmers realize that their allies are receiving a much higher degree of protection than they are.
― 307 ―
An interesting case of mutuality of interest between producers of a farm crop and processors is to be found in U.S. sugar policy. During the period of time of the U.S. sugar acts—from 1934 to 1974, and since 1978—essentially only raw sugar could be imported into the United States for consumption. Thus producers agreed to support a monopoly for domestic processors of imported raw sugar. In recent years there has been a rather different confluence of mutual interests. This has arisen because of technological changes that reduced the cost of producing sugar from corn. It is in the interests of the producers of high-fructose sugar to have a very high price support for cane and beet sugar in the United States. The domestic raw sugar price since 1981 has been approximately 20 cents per pound. The production of high-fructose sugar is very profitable at a price significantly less than 20 cents per pound dry weight equivalent. As a result, the share of the U.S. sweetener market captured by high-fructose sugar increased rapidly and in 1985 accounted for 40 percent of all caloric sweeteners consumed in the United States. One consequence was that U.S. imports of raw sugar declined from approximately 5 million short tons in 1980 to approximately 2 million tons in 1985 to as low as 1 million tons in 1989. During 1985 the U.S. sugar price was at least four times the international market price and for some months as much as six times.
There are now fewer than 12,000 producers of sugar cane and beets in the United States. The domestic production of beet and cane sugar has remained approximately constant over the past several years. Since there are no controls over the production of these sources of sugar, this means that the current high price of sugar is not especially profitable to the producers. In other words, the farmers are receiving a net income from producing sugar that on the average is approximately what they could obtain from producing the next best alternative farm product. The primary beneficiaries of the existing policies are the processors of cane and beet sugar and the producers of products that substitute for sugar, such as high-fructose sugar and the noncaloric sweeteners such as aspartame. Even the refiners of cane sugar are beginning to feel the pinch of reduced imports of raw cane sugar.
Food Security
Except for Honma and Hayami 1984, none of the studies of protection reviewed have included food security as a persuasive argument for positive protection. Even the Honma-Hayami analysis is rather ad hoc, since it is achieved by introducing dummy variables for Sweden and
― 308 ―
Switzerland and obtaining highly significant coefficients. But in the case of Sweden, at least, it may be that other explanations for the high level of protection are as relevant. The Swedish protection levels are not all that high compared to either Italy or Germany and are significantly lower than the Japanese ones.
The supporters of agricultural protection in Japan have made persuasive and effective use of the food security objective. The Ministry of Agriculture, Forestry and Fisheries (MAFF), with the support of politicians from all of the major political parties, has been an active propagandist for a high level of food security. The cynic might well argue that the emphasis upon food security is designed primarily to support the heavy taxpayer and consumer costs of Japan‘s agricultural policies. Public opinion polls indicate, however, that a large percentage of the Japanese people favor a high degree of food self-sufficiency.
The self-serving nature of MAFF‘s promotion of food security as an important national objective has been its emphasis upon the results of the "Projection for the World Food Supply and Demand Model" study. In June 1982 MAFF summarized the findings of this study in Farm Product Imports, Present State of Agriculture and Direction of Agricultural Policy . Included was the categorical statement that in the year 2000, if prices were constant, "a shortage of 53 million tons [of grain] will occur in the case of normal crop and a shortage of 198 million tons in case both the United States and the Soviet Union are simultaneously hit by crop failures. There will also be considerable shortages of livestock products." In terms of 1978 real prices, and assuming a normal crop and "a 2-to-3 percent increase in the prices of fertilizers, the prices will turn upward in the latter half of the 1980s. And in the year 2000, the prices of grains and soybeans will be about 1.7-fold to 1.8-fold and the prices of meat will be about 1.3-fold." And in a scenario in which the United States and the USSR were hit by simultaneous crop failures in 1985 and 1986, "the peak prices of wheat and coarse grains will be about 4-fold and those of rice and soybeans 2-fold to 3-fold (of the 1978 real prices)" (MAFF 1982, 24–26).
The first section of the report containing these statements is entitled "Problems about Farm Product Import Liberalization." The quoted material is from a section entitled "The Necessity for Maintenance of Domestic Agricultural Production." I have seen nothing to indicate that MAFF has repudiated the rather wild price projections or supplanted the earlier study by one more consistent with other reputable projections of future world demand for and supply of food.
― 309 ―
A speech given by an official of MAFF in 1980 painted a gloomy picture of world food demand and supply in an effort to justify the maintenance of the existing amount of land in rice paddy:
If Japan‘s agricultural production falls sharply, necessitating increasing grain and soybean imports, problems may arise in connection with the world demand and supply of grains. The world population will continue to grow, especially in developing countries. It will take a long time for those countries to improve their domestic agricultural production to a point approaching self-sufficiency. In addition, demand for feed grains for livestock production will rise steadily in developed and semi-developed countries. In the meantime, while demand for grain will increase in centrally planned economies, production remains somewhat unstable. In view of these tendencies, it is expected that the world demand-supply situation of feed grains and soybeans will become very tight, and the fragility of the demand-supply balance with regard to good or poor harvests will appear more frequently, thus throwing prices into instability. (Matsuura 1980, 6–7)
Later in the same speech, Matsuura declared that as an Asian nation, Japan should follow an import policy that would not adversely affect "those countries, particularly those in Southeast Asia, where people are suffering from serious food shortages" (8).
The view that food prices in world markets will be unstable and display an upward trend persists in the face of overwhelming evidence to the contrary. In the summary of the 1983 agricultural white paper, released in April 1984, this position was taken by MAFF: "When we also consider the irregular weather conditions in recent years and the growth of the world population, it appears that the international supply-demand situation leaves little room for optimism in either mid- or long-term projections" (MAFF 1984, 24).
Essentially the same point was made in a handout at a small meeting with officials of MAFF in late 1984, with added comment about concerns of the population that have been reinforced, perhaps consciously, by ministry officials: "Under the probable unstability [sic] of world food supply and demand in the future, concerns over the low level of food self-sufficiency rate have been growing among people."
Among the benefits that MAFF claims for its agricultural and trade policies is that they contribute to stability of world markets. Again, in the handout referred to above: "We think our country has been contributing to the stability of world markts through the stable imports of agricultural commodities." This is an incorrect conclusion; japan does not adjust its imports either when world prices decline or when they increase,
― 310 ―
thus leaving consumers and producers elsewhere in the world to cope with the variability of supply and demand.
Given the projections and speeches by important public officials that point to the possible disastrous consequences of further declines in the self-sufficiency ratio, it is not surprising that opinion polls indicate that the majority of the Japanese population support efforts to maintain or increase food production in Japan. Kenzo Hemmi reports the results of surveys made in 1975 and 1980 that asked whether domestic production of food should be increased whenever possible. In 1975 affirmative responses were given by 71 percent; in 1980 the affirmative responses increased to 75 percent. Hemmi also reports the results of a poll by a major newspaper (no date) in which 65 percent of the respondents agreed with the statement that "import liberalization of agricultural commodities must be promoted if it is not harmful to domestic producers" (emphasis added) (Hemmi 1983, 324).
For Japanese who have any memory of World War II and the years immediately thereafter, it is relatively easy to arouse fears related to food security. In those years many Japanese went hungry. But it is somewhat ironic that Japan was nearly self-sufficient in rice, and in food generally, before 1940—yet this did not provide food security. Before 1940 Japan itself produced approximately 85 percent of the rice it consumed (the rest came from Taiwan and Korea) and imported hardly any other food products.
The lesson that could be emphasized from the World War II experience is that a high level of domestic food supply is not enough to provide security. And what was true four decades ago is even truer today. Japan today is self-sufficient in the capacity to produce rice, but if Japan were blockaded or its trade with the rest of the world significantly reduced, its dependence upon fertilizer produced from imported materials would rapidly result in a sharp decline in rice production. This could readily occur in one year. There would be other negative output effects of a sharp reduction in energy imports.
I do not want to leave the impression that national food security is an inappropriate objective of national policy. But even rather casual analysis shows convincingly that most countries have made no effort to devise a food security policy that would, in fact, protect their citizens against the more likely adverse circumstances. Nor can it be said that the particular policies that are followed are carried out with minimum cost. For example, assuming that the objective of rice self-sufficiency were appropriate for the circumstances that Japan faces, its current rice policy is not an efficient means. The recent and current levels of rice
― 311 ―
prices have resulted in more rice production than can be consumed domestically. The solution to that embarrassment has been an expensive land diversion program. There is clearly a lower rice price that would be consistent with rice self-sufficiency without the costly land diversion effort. The current program obviously pursues objectives other than food security, presumably increasing the incomes of farmers who produce rice. These, it may be noted, are primarily part-time and not fulltime farmers.
But there are other approaches to security of rice supplies. Domestic use is about 10.5 million tons and declining. One approach would be to set rice prices at a level required to produce about 5 million tons or about half domestic use and to continuously hold a beginning-of-the-year stock of 11 to 12 million tons. Against the types of interruption of supply that might adversely affect Japan, this policy could well provide a greater degree of food security than the current policy and at a significantly lower consumer and taxpayer cost. The cost of the storage program would be no more than a fifth of the budgetary costs of the present rice price policy.
This discussion of food security and the manner in which governmental and politcal groups have created and used the fears of the Japanese population for the benefit of the majority party is not meant to imply that what is done is in any way unique to Japan. Politicians everywhere make those expenditures and commitments that they believe will maximize their own interests.
In the European Community, for example, the CAP has been supported on several grounds that are without substantial foundation. The major appeal to consumers has been one related to security and stability. The EC‘s Guidelines for European Agriculture justifies the high and stable prices as follows:
Comparisons with world market prices may easily lead to misleading conclusions. It is highly unlikely that European consumers could be supplied for long at low and stable world prices if Community supply, because of reduction in production, would depend to a greater extent on imports. World market prices are notoriously volatile because the quantities involved in international trade are often marginal in relation to total production (e.g., sugar, cereals, dairy products) and may reflect short-term fluctuations in production. For several products (e.g., beef, wine, tobacco) there is no real world market and prices vary according to the destination of exports.
Therefore the Commission is convinced that a generalized and systematic alignment to world market prices would not be a practical policy guideline. (Commission of the European Communities 1981b, 8)
― 312 ―
The fact that the CAP is a major source of price instability in world markets is entirely ignored by the Commission of the European Communities. Also ignored is the available empirical and analytical evidence that with liberal trade among OECD countries there would be adequate supplies at prices much more stable than now prevail in the world markets. The commission does not recognize that the internal price stability resulting from the CAP is bought at the expense of greater instability of prices for many developing countries and consumers and producers in countries that permit domestic prices to vary with international prices. But failure to recognize the effect upon others outside the borders of the European Community is not unexpected, since there is no direct loss to EC policymakers from the costs imposed upon those who have no voice in EC policy determination.
4—
Why There Has Been Modest Success in Liberalizing Agricultural Trade
By explaining why countries choose to intervene in agriculture (be it positively or negatively), the previous section has gone a considerable distance in suggesting why trade liberalization has made so little progress. The low-income countries could not carry out many of their urban-oriented or urban-biased policies if trade in agricultural products were liberalized. And in the high-income countries, the political influence of agricultural interests, including the processors and distributors of farm products, has become so great that the political process apparently does not have the capacity to deal with the issues involved.
But there are aspects of the General Agreement on Tariffs and Trade (GATT) rules and principles for agricultural products that also have responsibility for the lack of success. I have dealt with these issues in some detail elsewhere (Johnson 1950, 1984) and will only summarize the main main points here.
The general GATT principles for international trade were that both quantitative restrictions and subsidies, including but not restricted to export subsidies, were prohibited. However, exceptions to these principles were made for primary products at the insistence of the United States.
Article XI, paragraph 2, exempts certain quantitative restrictions from the general ban on all "restrictions other than duties, taxes or other charges:
(c) import restrictions on agricultural or fisheries products, imported in any form, necessary to the enforcement of governmental measures which operate:
― 313 ―
(i) to restrict the quantities of the like domestic product permitted to be produced . . . ; or
(ii) to remove a temporary surplus of the like product . . . by making the surplus available to certain groups free of charge or at prices below the current market level; or
(iii) to restrict the quantities permitted to be produced of any animal product the production of which is directly dependent, wholly or mainly, on the imported commodity, if the domestic production of that commodity is relatively negligible."[5]
The following important guideline was indicated: "Moreover, any restrictions applied under (i) above shall not be such as will reduce the total of imports relative to the total of domestic production, as compared with the proportion which might reasonably be expected to rule between the two in the absence of restrictions. In determining this proportion, the contracting party shall pay due regard to the proportion prevailing during a previous representative period and to any special factors which may have affected or may be affecting the trade in the product concerned."
But these exceptions, inserted by the United States to placate Congress, were not enough to permit the United States to carry out some farm programs.[6] To a considerable degree the failure of GATT to be an appropriate instrument for dealing with agricultural protection stems
― 314 ―
from the unwillingness of the United States to abide by the exceptions that it designed and insisted upon.[7]
In the late 1940s and early 1950s unwillingness by the United States to subject its farm programs to the discipline of international trade or serious international negotiations resulted in the insertion into the GATT of exceptions for agricultural subsidies, including export subsidies. Article XVI of GATT contains a general provision against subsidies of all kind, including price and income supports, that might result in increasing exports. After January 1, 1958, subsidies on all products other than primary products were prohibited.
An attempt was made to include a clause that might hold subsidies on agricultural products in check, but it was written in such ambiguous language that it is without any effect. The export subsidy provision for primary products was that for any subsidy "which operates to increase the export of any primary product from its territory, such subsidy shall not be applied in a manner which results in the contracting party having more than an equitable share of world export trade in that product, account being taken of the shares of the contracting parties in such trade in the product during a previous representative period, and any special factors which may have affected or may be affecting such trade in the product."
This exception, which was designed to permit the United States to hold some of its farm prices above world market prices, especially those of cotton and wheat during the 1950s, is now being used by others for the same objective. Had the United States agreed to the prohibition on all subsidies after 1957, the shape of the Common Agricultural Policy might now be quite different. This point was well put in a recent book on the CAP
Ironically the Community‘s creation of the CAP in its current form, with its use of variable import levies and export refunds as its principal agricultural trade measures, was only possible as a result of earlier actions by, principally, the USA. Thus, in 1955 the US achieved a formal waiver from GATT provisions so that it could continue to use import quotas and fees to the extent necessary to prevent material interference with its domestic agricultural support programmes, so legitimizing the primacy of such programmes over international trade obligations. Then, in 1958, the USA was foremost among those countries which refused to endorse an absolute prohibition on the use of export subsidies. As a result, GATT allowed export subsidies to continue to be used for primary products, Sub-
― 315 ―
ject to the condition that they did not allow a country "more than an equitable share of world trade" (Article XVI of the GATT). Hence the EC has been able to use export refunds as a principal CAP policy instrument. (Harris, Swinbank, and Wilkinson 1983, 275)
As long as the United States retains the 1955 GATT waiver that permits it to use import restrictions as it sees fit, there can be no significant progress toward liberalization of agricultural trade. Generally speaking, the United States makes no pretense of abiding by the GATT exceptions for quantitative import restrictions—exceptions the United States wrote. We have done nothing to bring our dairy price support program into conformity with the GATT exceptions. Our current sugar program is in clear violation of the GATT principles, as are our beef import restraints. The fact that we use "voluntary" export restraints by our suppliers rather than import quotas is beside the point. The voluntary restraint procedure is one designed to keep exporters from complaining, since we cooperate with them to exploit the American consumer. The same principle applies to our sugar quotas: for the diminishing amounts of the quotas, we permit the exporters to receive the U.S. domestic price, currently several times the world price.
The United States has never taken seriously the requirement that quotas are not to be used to reduce imports more than domestic production is limited. The GATT provisions were designed to assure importers that their share of a market would not be reduced by the use of quantitative restrictions. We have never seriously accepted this obligation. Thus it is not surprising that the EC has never accepted responsibility to administer the variable levies—which are comparable to quantitative import restrictions—in such a way as to assure exporters a constant share of the EC market.
The United States, so far as I know, has never attempted to explain how the deficiency payments that result from our target prices are consistent with the mild GATT provisions on subsidies. The deficiency payments affect U.S. farm production and thus inevitably the volume of our exports. It is true that the deficiency payments are usually associated with acreage diversions, but this is not always the case. Nor has there been any public indication of whether the output-increasing effects of the deficiency payments outweigh the output-reducing effects of acreage diversions.
The United States and the EC continue to talk past each other, each accusing the other of various violations of appropriate behavior. Japan continues grudgingly to resist opening its markets to agricultural
― 316 ―
products. It makes minor adjustments, though it has so far completely resisted any significant modification of its expensive rice policy. Until these three entities decide to approach trade negotiations in a serious manner, with the positive objective of reducing barriers to trade in agricultural products, there is little chance that there will be significant trade liberalization during this century.
5—
Costs and Benefits of Agricultural Liberalization
The unwillingness to embrace liberalization of agriculture is almost universally attributed to the effect that it would have upon the incomes of farm people. There is also the associated argument, frequently noted in Japan and the European Community, that liberalization would drastically reduce the number of persons engaged in agriculture and would greatly diminish the role of the rural community in the nation‘s political and social life.
Let us consider first the effect of trade liberalization upon the incomes of farm people. The fear that trade liberalization would adversely affect farm incomes is derived from the proposition that the level of farm prices has a significant long-run effect upon farm incomes. The argument for protection and prices above market equilibrium is that farm incomes are enhanced thereby. This conclusion rests upon the proposition that incomes are determined by demand. If you increase the price of a product by a government commitment to purchase at the higher price, you increase the demand for that product and, in turn, the demand for the labor, capital, and land used to produce the product.
What this line of argument ignores is that prices of resources are determined by the interaction of demand and supply of those resources. The position or shape of demand is not sufficient to determine the price of any resource. True, increasing the prices of farm products does shift the demand for farm resources and does increase the amount of such resources demanded at any given price. But how large the effect of the increase in demand for labor will be from any increase in the price of a product depends upon its supply response to a change in wage. In the extreme, if the elasticity of supply of the resource is very high, an increase in demand for it will have almost no effect upon its price; the primary effect will be upon the quantity employed.
Increasing the demand for farm inputs, such as labor and capital, has no measurable long-run effect upon the return to a farm input or re-
― 317 ―
source unless the supply of the input can be limited. Since it is not possible to limit the amount of labor or capital engaged in farming, any increase in the return to labor through higher output prices will be quite temporary, because adjustment in supply of both labor and capital occurs quite promptly, especially in countries where agriculture is well integrated into the national economy. The lesson that the returns to labor and capital in agriculture are little affected by the level of farm prices is one that is very difficult for the layman, and I include politicians in this category, to learn and to accept. It seems self-evident that higher prices mean higher returns to labor and capital. But it just isn‘t so.
The supply of labor to agriculture in the industrial economies and in the newly industrializing economies (NICs) is very elastic. In other words, any increase in the returns to labor, other things being the same, results in an increase in amount of labor offered (Johnson 1973). This refers to both family and operator labor and to hired labor. As demand increases, whether owing to economic growth or to governmental intervention through higher output prices, the main consequence is to increase farm employment (compared to what it otherwise would have been) and not in any significant increase in the returns to farm families for their labor. There are a number of statistical studies that show that the supply of labor to agriculture is very responsive to differences in the returns to farm labor compared to nonfarm labor (Trychniewicz and Schuh 1969).
In the United States between 1960 and 1980, farm output prices declined by 15 percent relative to the prices of farm production inputs. Yet during the same period the incomes of farm people relative to nonfarm people increased from 50 to 80 percent. Dale Hathaway, a prominent agricultural economist who became under secretary of agriculture, strongly supported the view that adjustments in the labor market and not commodity programs were the source of the improvement in the relative income position of farm families that occurred after 1940.
During the war, the differential began to narrow. In the last half of the 1940‘s, per capita income of farm people averaged 60 percent of the per capita income of nonfarm people from all sources.
In the 1950‘s, however, farm income per person again fell behind—remaining mostly static while the per capita income of nonfarm people rose by more than a third. In the last half of the 1950‘s, the per capita income of farm people was only one-half the per capita income received by people living off the farm.
In the early 1960‘s, we could see the beginning of adjustment. By the end of the decade, per capita income on farms averaged above $2,000
― 318 ―
compared with around $3,000 for nonfarm people. For the 5 years 1965 through 1969 people living on farms averaged 71 percent of the per capita income of people living off the farm.
In more recent years, this percentage has risen to 85 or more— although this of course varies from year to year.
So—the labor market did adjust. But the adjustments were difficult for many. Despite government efforts to deal with these difficulties, it appears in retrospect that no government policy or program was significant in aiding the adjustment or softening the pain of adjustment for farm people. (Hathaway 1980)
What, then, does determine the incomes of farm people if it is not the level of output prices? The principal determinants of the return for the work of farm people are the amount of human capital (education) they possess, the level of wages in the rest of the economy, and the access that farm people have to available nonfarm employment. One need only look at the experience of the industrial economies, especially the European Community, the United States, Canada, and japan, to verify this.
There are large differences in farm incomes within the EC—regional differences of 5 or 6 to 1. It is worth quoting the Commission of the European Communities on the wide regional differences in farm incomes and the reasons why they exist:
During the period from 1964/65 to 1976/77, regional disparities in agricultural incomes (as measured by gross value-added per agricultural worker) increased in the Community. The ratio between the regions with the highest agricultural incomes and those with the lowest rose from 5:1 to 6:1.
Generally speaking, the regions with an above-average-level of agricultural income are to be found in a favourable general economic context; the converse is true of regions with a low level of agricultural incomes. (Commission of the European Communities 1981a, 52)
What the commission is saying is that high farm output prices, even when maintained over a long period of time, have no significant effect upon the relative income positions of the low-income farm communities. Clearly the large differences in farm incomes cannot be explained by differences in output or input prices, since such prices are more or less the same in all countries. What does explain the differences are the opportunities for nonfarm work, as noted by the commission, and the amount of education farm people have acquired, something the commission tends to ignore.
If you find the previous line of reasoning unconvincing, compare data on the level of national income per capita with the percentage of the population engaged in agriculture. It is evident that the incomes
― 319 ―
of farm families are not going to exceed those of nonfarm families, except in a few high-income countries where farm people own much more wealth than nonfarm families. As of the early 1980s, countries with less than $400 per capita gross national product had about 70 percent of their labor force engaged in agriculture, those with incomes of about $1,000 per capita had about 50 percent; those with incomes of $2,000, about 30 percent; and those with per capita incomes of $11,000, about 6 percent (World Bank 1985). A similar relationship has prevailed for decades, either across countries or in the same countries over time.
It is easy to exaggerate the potential effects of the level of output prices upon farm employment. One effect of economic growth— increasing real per capita incomes—upon agriculture cannot be modified or offset. As economic growth occurs, agriculture‘s share of both national income produced and of national employment declines. There is no escape. As people‘s incomes increase, they spend a smaller and smaller fraction of that income upon food. And agriculture has shown the capacity to increase productivity, at least labor productivity, as rapidly as the nonfarm sectors of the economy have. In fact, in the industrial countries labor productivity in agriculture has outpaced labor productivity in manufacturing and in the remaining nonfarm sectors. Japan is the only important exception to this rule.
The only effect that higher output prices could have upon farm employment is to slow down the rate of decline. But a comparison of the relationships between the levels of farms prices for wheat and corn and the decline in farm employment since the mid 1950s shows that the level of farm prices seems not to have had a significant effect upon the rate of farm employment decline.
Tables 8.8 and 8.9 provide simple comparisons between the levels of grain prices and the decline in farm employment. Employment change is presented for three time periods from 1955 to 1979; the grain prices are for 1970 and 1979. However, the relative grain prices in 1960 were about the same as in 1970. As noted, there is no apparent effect of high output prices in slowing down the decline in farm employment.
International Price Effects of Agricultural Protectionism
Before there can be consideration of the welfare and income-distribution effects of trade liberalization, it is necessary to have estimates of the effects of existing forms of agricultural protectionism upon international market prices. There have been a number of studies made of such effects, and several more are now in process.
― 320 ―
TABLE 8.8 Prices Received by Farmers in Major
Industrial Countries
(U.S. $per metric ton)
1970
1979
Wheat
Coarse Grain
Wheat
Coarse Grain
Belgium
100
86
212
214
Denmark
—
62
224
214
France
84
75
232
214
Germany
99
89
274
247
Italy
110
93
231
210
Netherlands
102
—
212a
202a
United Kingdom
74
68
160
—
Ireland
76
—
—
181
Canada
61
—
117
100
Japan
162
131
770
582
United States
48
52
118
95
SOURCE: Johnson 1982, 359.
a 1978.
TABLE 8.9 Annual Rates of Decline in Farm Employment
(% per year)
1955-60
1960-70
1970-79
European Community
- 3.2
- 4.6
- 3.4
Belgium
- 3.7
- 4.9
- 4.6
Denmark
- 2.1
- 3.1
- 2.7
France
- 3.7
- 3.7
- 4.6
Germany
- 3.3
- 4.6
- 4.8
Ireland
- 2.5
- 3.0
- 3.1
Italy
- 3.2
- 5.8
- 2.2
Netherlands
- 2.7
- 3.4
- 1.8
United Kingdom
- 2.5
- 3.8
- 1.0
Canada
- 3.6
- 2.7
- 0.7
Japan
- 2.6
- 4.0
- 4.1
United States
- 3.3
- 4.5
- 0.5
SOURCE:  Johnson 1982, 359.
Protectionism has two effects upon international market prices: one is upon the average level of prices, and the other is upon the variability of prices. The studies that I now summarize generally include both.
It is important to note that the effects of the trade interventions in agriculture upon international market prices depend upon the level of
― 321 ―
protection that existed for the time period covered by the estimates. Some studies are based on the levels of protection that existed in 1975–77; others are based on 1978–80. Protection levels in the first period were significantly less than in the second one. As a result, the international price effects based on the 1975–77 period are the smaller of the two periods.
The studies undertaken by Tyers and Anderson (1984) and Chisholm and Tyers (1985) have been the most extensive. These studies included five commodities or groups of commodities: rice, wheat, coarse grain, ruminant meat, and nonruminant meat. In one exercise it was assumed that there was free trade in the six principal market economies, four NICs, and eight LDCs as of 1978–80. The results are presented in table 8.10 in terms of changes in projected prices. The 1990 estimates, which allow sufficient time for all production and consumption adjustments to occur, indicate that the expected level of wheat and coarse grain prices under liberalization would differ from the prices under continuation of present policies by an increase of 20 percent for wheat and 16 percent for coarse grains. The increase for nonruminant meat was projected at 2 percent. Liberalization was estimated to result in a 27 percent increase in the price of ruminant meat, but part of this difference may be owing to the difficulty of making appropriate adjustments for quality differences.
Valdes and Zietz have estimated the effects of the agricultural protection of the OECD countries upon the exports of farm products by the developing countries (mimeographed paper). These estimates, based upon 1975–77 levels of protection, involved projecting the price effects of the trade restrictions. Based on reducing protection levels by 50 percent, Valdes and Zietz obtained results comparable to those reported in table 8.10, though not as large, owing to the smaller degree of liberalization. Their study included 99 commodities, with the price increases resulting from reducing the barriers to trade reported for 47. There were only 4 commodities with price increases greater than 10 percent— and all of these were processed products: roast coffee, cocoa paste and powder, malt, and wine. Other price increases were 2 percent for maize, 4 percent for wheat, 8 percent for raw sugar, 7 percent for beef, and 9 percent for pork. (The degree of protection for sugar from 1975–77 was much lower than in subsequent years.)
Ulrich Koester, using the model and data bases developed by Valdes and Zietz, estimated the effect of removing the grain protection by the European Community upon the level of international market prices for
― 322 ―
TABLE 8.10 Projected Outcomes of Hypothetical 1980 Agricultural Trade Liberalization
Rice
Wheat
Coarse Grains
Ruminant
Meat
Nonruminant
Meat
World Price a (% change)
LDCs, 1981
- 8
+ 10
+ 1
+ 1
+ 1
(- 71)
(- 40)
(- 7)
(- 9)
(- 8)
LDCs, 1990
- 10
- 5
+ 2
+ 2
- 0.1
(- 71)
(- 55)
(- 12)
(- 12)
(- 14)
OECD/NICs, 1981
+ 45
+ 36
+ 28
+ 38
+ 23
(- 10)
(- 37)
(- 27)
(- 36)
(- 18)
OECD/NICs, 1990
+ 17
+ 21
+ 16
+ 26
+ 3
(- 12)
(- 46)
(- 26)
(- 22)
(+ 3)
All eighteen countries, c 1981
+ 8
+ 32
+ 26
+ 37
+22
(- 71)
(- 54)
(- 30)
(- 39)
(- 23)
All eighteen countries, c 1990
+ 6
+ 20
+ 16
+ 27
+ 2
(- 70)
(- 63)
(- 32)
(- 27)
(- 11)
World Trade b (difference and % change)
LDCs, 1990
+ 1.4
+ 1.5
+ 4.1
0
0
(+ 10)
(+ 2)
(+ 4)
(0)
(0)
OECD/NICs, 1990
+2.4
- 4.7
+ 27.6
+ 2.5
3.3
(+ 18)
(- 6)
(+ 27)
(+ 73)
(+ 61)
All eighteen countries, c 1990
+ 3.3
- 3.8
+ 28.3
+ 2.4
+ 3.2
(+ 25)
(- 5)
(+ 27)
(+ 72)
(+ 60)
SOURCE: Chisholm and Tyers 1985.
a World prices are percentage changes in forecast means, with changes in standard deviations in parentheses (based on 100 simulations), for the first year (1981) and the tenth year (1990) after trade liberalization respectively.
b World trade volumes are quantity changes as of 1990, measured in million metric tons, with percentage changes in parentheses.
c i.e., the eight LDCs (Bangladesh, Burma, India, Indonesia, Pakistan, the Philippines, Sri Lanka, and Thailand), the four NICs (South Korea, Malaysia, Singapore, and Taiwan), and the six OECD members (Australia, Canada, EC, Japan, New Zealand, and the United States).
― 323 ―
grain (mimeographed paper, 1982). The projected increases in world grain market prices ranged from less than 1 percent for millet and sorghum to almost 20 percent for oats. For wheat the projected increase was 9.6 percent and for maize, 2.2 percent. The price increase for barley was projected at 14.3 percent. If the grains are weighted by the value of world exports in 1975–77, the average increase in price would have been 6.7 percent.
Maurice Schiff (1983) has estimated a model of the world wheat market, based on econometric estimates of his own. Free trade is assumed for the European Community, the United States, Canada, Australia, Japan, and Argentina. The model includes estimates of the wheat trade functions of the USSR and the rest of the world for continuation of existing policies. Schiff estimates that if there had been free trade in wheat in the designated countries from 1964 to 1978, the average increase in world wheat price would have been 15 percent. He also estimates that if there had been free trade in the European Community only, with all other countries continuing their actual policies, the world market price of wheat for the same period would have been 17 percent higher. This result may seem somewhat surprising until it is remembered that during most of the years included in the analysis the major exporters, especially the United States and Canada, had limited the output of wheat by domestic supply-management programs. If there had been universal free trade, exports of wheat by the major exporters would have been somewhat higher than they actually were.
Stefan Tangerman and Wolfgang Krostitz (1982) have estimated the effects of trade liberalization on the beef sector and calculated the implicit tariff equivalent of the restraints on trade that existed during 1977–79. Elasticities of supply and demand were also estimated. With this information plus the actual levels of production and consumption of beef in each country or region, changes in production, consumption, and net trade were made for reductions in the implicit tariffs of 25, 50, and 100 percent. They estimate that with full trade liberalization, the international market price for beef would increase by 47 percent.
One very interesting result is that no one of the three degrees of reduction of the implicit tariffs would have any noticeable effect upon domestic prices of beef in the United States or Canada. The reason for this rather striking result is that the increases in world market prices would be approximately equal to the reduction in the implicit tariff for each of the three cases: 25, 50, or 100 percent. For example, if the United States reduced its implicit tariff by 50 percent, this would have amounted to
― 324 ―
a decrease in the tariff by $230 per ton (slaughter weight). However, if all countries reduced their implicit tariffs by 50 percent, the world price would increase by $220 per ton. For the European Community a reduction of its implicit tariff of 118 percent by 50 percent would have resulted in a decrease in the domestic price of 15 percent. The decline in the domestic price in Japan was projected to be 28 percent, or $163 per ton.
Roy Allen, Claudia Dodge, and Andrew Schmitz (1983) arrive at a much more modest estimate of the effect of the voluntary export constraints for beef on beef prices in the United States. For 1976–77, when there were voluntary restraints on beef exports to the United States, the U.S. price of frozen boneless beef was increased by about $85 per ton, or about 8 percent of the free-trade price. However, the price increase for all U.S. beef would be significantly smaller than the estimated 8 percent, since beef of the quality that is imported accounts for no more than a quarter of U.S. beef consumption. An interesting result of the study was that the average price received by the exporters was slightly higher than it would have been under free trade. Under the voluntary quotas the exporters realized the price gain from the reduced level of U.S. imports. It may be noted that there were no voluntary restraints in effect during 1980, 1981, and 1982, though such restraints were imposed for the second half of 1983.
The trade intervention policies not only affect the average level of international market prices but also influence the variability of prices. The figures in parentheses in table 8.10 are measures of the price variability under current policies and free trade. The measure is the coefficient of variation, which is in percentage terms and represents the relationship between the standard deviation of prices and the average prices. For wheat and rice the estimates indicate that current policies have substantially increased the variability of international market prices.
The much greater variability of international market prices under current policies than under free trade results from the nature of agricultural protection that prevails in many countries. Agricultural protection per se need not result in increasing price variability in international markets and in the countries where international prices are directly reflected in domestic prices. It is the form of protection that causes the increased variability. Protection of agriculture that functions by stabilizing domestic prices by varying imports and exports of commodities destabilizes international market prices. It does so by using import and export changes to meet any variations in domestic supply-and-demand variabil-
― 325 ―
ity and by preventing internal price changes that would absorb at least part of the variability.
Consider the following example. A country has a fixed price of $100 per ton for grain, average use is 55 million tons and average production is 50 million tons. The price of $100 is maintained by varying the amount imported, with average annual imports being 5 to 10 million tons. The internal price remains at $100. In the next year production is 55 million tons and there are no imports, since none are required to keep the price at $100 if demand has remained unchanged. As this example indicates, all of the variability in domestic production is imposed upon the international market. None of the variability is absorbed by changing domestic use; since the price remains at $100, the users have no incentive to change. Nor are producers encouraged to increase their production in the year following the short crop. In effect, the country achieves domestic price stability by exporting its instability through varying imports to exactly offset production departures from the average.
Welfare Gains and Losses from Agricultural Liberalization
Chisholm and Tyers (1985) have provided estimates of the worldwide welfare impacts of full agricultural trade liberalization by the eighteen countries included in table 8.10. Their results are presented in table 8.11. Despite the likely increases in international market prices of many commodities, the welfare gains from liberalization are very large for Japan and the EC, two entities with high rates of agricultural protection. The per capita gains are also large for New Zealand, owing to the impact of greater market access at improved prices for farm products, in which it has a great comparative advantage. Australia also gains, while the United States is estimated to have a very modest gain.
Among the NICs, South Korea has a large gain at $158 per capita, and Taiwan has a significant, though smaller, gain in welfare. Most of the other developing countries are little affected by free trade in agricultural products as measured by per capita welfare effects. Bangladesh is an important importer of cereals, for example, and would lose as a result of higher international market prices. Some of the modest effects for the developing countries may result from the limited range of commodities included in the exercise.
The welfare gains and losses are measured as the sum of the changes in consumer surplus and producer surplus minus the net change in government revenue. I have argued elsewhere (Johnson 1973) that the
― 326 ―
TABLE 8.11 Projected Welfare Impacts in 1990 of Hypothesized Complete Agricultural Trade
Liberalization in 1980
Free Trade by LDCs
Free Trade by OECD-NICs
Free Trade by Eighteen countries
Total
(mill. U.S. $)
Per Capita
(U.S. $)
Total
(mill. U.S. $)
Per Capita
(U.S. $)
Total
(mill. U.S. $)
Per Capita
(U.S. $)
OECD
Australia
- 27
- 2
1,098
67
1,126
69
Canada
- 87
- 3
650
25
697
27
EC
- 426
- 2
34,157
129
34,156
129
Japan
240
2
30,923
247
30,896
246
New Zealand
40
13
607
193
637
202
United States
356
1
3,805
16
4,158
17
Total
96
71,240
71,670
NICs
South Korea
- 11
0
6,959
156
7,036
158
Malaysia
- 46
- 3
- 269
- 16
- 289
- 17
Singapore
6
2
- 44
16
- 36
- 13
Taiwan
30
1
1,291
61
1,229
58
Totals
- 21
7,940
7,940
7,940
LDCs
Bangladesh
- 33
0
- 106
- 1
- 150
- 1
Burma
- 19
0
5
0
3
0
India
- 346
0
- 1,647
- 2
- 1,830
- 2
Indonesia
374
2
- 337
- 2
- 66
0
Pakistan
- 77
- 1
41
0
- 54
0
Philippines
- 93
- 2
- 93
- 2
- 166
- 3
― 327 ―
TABLE 8.11 Projected Welfare Impacts in 1990 of Hypothesized Complete Agricultural Trade
Liberalization in 1980
Free Trade by LDCs
Free Trade by OECD-NICs
Free Trade by Eighteen countries
Total
(mill. U.S. $)
Per Capita
(U.S. $)
Total
(mill. U.S. $)
Per Capita
(U.S. $)
Total
(mill. U.S. $)
Per Capita
(U.S. $)
Sri Lanka
7
0
- 13
- 1
- 12
- 1
Thailand
72
1
28
0
234
4
Total
- 114
2,122
2,041
Aggregates
Africa
231
0
- 1,264
- 2
1,025
- 2
China
- 60
0
- 66
0
181
0
Central & South America
64
0
- 1,053
- 2
- 1,016
- 2
USSR
79
0
- 492
- 2
- 424
- 1
Middle East and other
Asian countries
142
0
- 1,157
- 4
- 1,027
- 3
Other European
countries
- 46
0
- 170
0
- 62
0
Total
410
4,202
3,735
SOURCE: Chisholm and Tyers 1985.
NOTE: Figures represent changes in total and per capita mean annual welfare measured in 1980 U.S. dollars.
― 328 ―
conventionally measured welfare costs of market intervention are only a part—perhaps a minor part—of the economic effects. These policies frequently involve very large income transfers from one group to another within each of the economies. In the industrial economies the income transfers are from consumers to producers, whereas in many, though not all, developing countries, the transfers are from producers to consumers. What social and economic benefits are realized by such transfers is almost never asked. In the industrial countries, the transfer from consumers to farmers results in net transfers from at least half of the consumers to at most a fifth of the farmers, who have significantly higher incomes than the median consumer. Even where governmental expenditures are involved, there is a significant transfer from low-income taxpayers to farmers with much higher incomes.
At the time when Great Britain was considering joining the European Community, estimates were made of the distribution of costs and benefits of the EC agricultural policy compared to free trade (Josling et al. 1972). This study indicated that consumers with incomes in the lowest six quintiles would pay £ 581 million more for their food with the EC prices than under free trade. Of this total £ 366 million went to farmers who had incomes higher than the designated group of consumers. Farmers in the same income group as the consumers would have received £ 77 million in extra income if the EC prices had prevailed. These results are comparable to what occurs in industrial countries today when most of the transfer to producers is through higher prices.
6—
Concluding Comments
What does the future hold? There was agreement among the OECD countries to include agriculture in the Uruguay round of trade negotiations. True, agriculture was also included in past GATT negotiations, though in the prior negotiations none of the participants were willing to seriously discuss the changes in their domestic programs that would be required to achieve significant trade liberalization. Nor has it been possible to achieve a modification of GATT rules governing the use of quantitative restrictions on trade or reasonable restraints upon the use of subsidies, including, but not restricted to, export subsides.
The current round of GATT negotiations could show more positive results. The financial costs of the farm programs of the EC and the United States and other OECD countries are enormous, with a total annual cost to consumers and taxpayers of more than $200 billion.
― 329 ―
The hope for positive results rests upon the combined effects of the large budgetary costs and the growing recognition that the needed changes in domestic farm policies can only be made if all the industrial countries act in unison. It is generally agreed that no country can unilaterally reduce its subsidies and trade restrictions and permit more liberal trade. To do so would result in the flooding of its market with large quantities of highly subsidized farm products. Politically none of the industrial countries can accept this outcome. But if in unison each gradually reduced the protective measures, including domestic subsidies as well as border measures, then the scope of the required adjustments would be reduced for all. This would be true because the gradual reduction in protection would result in increases in international market prices.
There may also be greater recognition among politicians that the current policy measures have not been effective in guarding farmers against the loss of income and financial difficulties. In neither the EC nor the United States have the high costs of the current programs been adequate to maintain the level of farm incomes, to say nothing about increasing such incomes. As of the mid 1980s in the United States, the total governmental expenditures upon farm income and price supports have, in some years, exceeded the net income of all farm operators from their farm operations. In the EC the sum of governmental expenditures plus consumer costs exceeds net farm income. One can have some hope that the combination of high costs and a recognition that these costs are not translating into increased farm incomes may lead the political process to try other means for improving the economic situation of farm families. Yet there is a significant probability that the Uruguay round will fail owing to the inability to devise an acceptable agreement for reducing agricultural subsidies.
References
Allen, Roy, Claudia Dodge, and Andrew Schmitz. 1983. "Voluntary Export Restraints as Protection Policy: The U.S. Beef Case." American journal of Agricultural Economics 65 (May): 291–95.
Bates, Robert H. 1981. Markets and States in Tropical Africa: The Political Basis of Agricultural Policies . Berkeley: University of California Press.
Binswanger, Hans P., and Pasquale L. Scandizzo. 1983. Patterns of Agricultural Protection . Washington, D.C.: World Bank.
Chisholm, Anthony H., and Rodney Tyers. 1985. "Agricultural Protection and Market Insulation Policies: Applications of a Dynamic Multisectoral Policy."
― 330 ―
In J. Whalley and J. Piggott, eds., New Developments in Applied General Equilibrium Analysis . Cambridge: Cambridge University Press, 1985.
Commission of the European Communities. 1981a. The Agricultural Situation in the Community : 1980 Report. Luxembourg: Official Publications of the European Community.
———. 1981b. Guidelines for European Agriculture . COM (81) 608 final. Brussels.
Dam, Kenneth W. 1970. The GATT: Law and International Economic Organization . Chicago: University of Chicago Press, 1970.
Golub, Stephen S., and J. M. Finger. 1979. "The Processing of Primary Commodities: Effects of Developed-Country Tariff Escalation and Developing-Country Export Taxes." Journal of Political Economy 67 (June): 559–77.
Harris, Simon, Alan Swinbank, and Guy Wilkinson. 1983. The Food and Farm Policies of the European Community . New York: John Wiley & Sons.
Hathaway, Dale E. 1980. "Shifting Markets, Government Policies, and the Implications for Agricultural Financing." Speech at Farm Credit/Universities Conference, Louisville, Kentucky, April 26, 1980. U.S. Department of Agriculture Press Release.
Hemmi, Kenzo. 1983. "A Japanese Perspective." In John W. Rosenblum, ed., Agriculture in the Twenty-First Century. New York: John Wiley & Sons.
Hillman, Jimmye S. 1978. Nontariff Agricultural Trade Barriers . Lincoln: University of Nebraska Press.
Honma Masayoshi and Hayami Yujiro[*] . 1984. "Structure of Agricultural Protection in Industrial Countries." Unpubished paper. Tokyo Metropolitan University.
Howarth, Richard W. 1971. Agricultural Support in Western Europe . London: Institute of Economic Affairs.
Japan. Ministry of Agriculture, Forestry, and Fisheries. 1982. Farm Product Imports, Present State of Agriculture, and Direction of Agricultural Policy . Tokyo: MAFF.
———. 1984. The State of Japan‘s Agriculture, 1983 A Summary Report . Tokyo: MAFF.
Johnson, D. Gale. 1950. Trade and Agriculture: A Study of Inconsistent Policies . New York: John Wiley & Sons.
———. 1973. World Agriculture in Disarray . London: Macmillan & Co.
———. 1975. "World Agriculture, Commodity Policy, and Price Variability." American Journal of Agricultural Economics 57 (December): 823–28.
———. 1982. "International Trade and Agricultural Labor Markets: Farm Policy as Quasi-Adjustment Policy." American Journal of Agricultural Economics 64 (Mar): 355–61.
———. 1984. "Domestic Agricultural Policy in an International Environment: Effects of Other Countries‘ Policies on the United States." American Journal of Agricultural Economics 66 (December): 735–44.
Josling, T. E., Brian Davey, Alister McFarquhar, A. C. Honnale, and Donna Hamway. 1972. Borders and Benefits of Farm-Support Policies. Agricultural Trade Research Paper no. 1. London: Trade Policy Research Centre.
Liepmann, H. 1938. Tariff Levels and Economic Unity of Europe . London: Allen & Unwin.
― 331 ―
McCrone, Gavin. 1962. The Economics of Subsidizing Agriculture . London: Allen & Unwin.
Matsuura Akira. 1980. "Japan‘s Agricultural Policy." Japan‘s Agricultural Review 80 (February): 1–11.
Miller, Tracy C. 1985. "Explaining Agricultural Price Policy across Countries and across Commodities Using Political Interest Group Theory." Ph.D. diss., University of Chicago.
Ogura Takekazu. 1980. Can Japanese Agriculture Survive ? 2d ed. Tokyo: Agriculture Policy Research Center.
Saxon, Eric, and Kym Anderson. 1982. Japanese Agricultural Protection in Historical Perspective . Australia-Japan Research Center, Pacific Economic Paper no. 92. Canberra: Australian National University.
Schiff, Maurice. 1983. "Information, Expectations and Policies: A Study of the World Wheat Market." Ph.D. diss., University of Chicago.
Tangerman, Stefan, and Wolfgang Krositz. 1982. Protectionism in the Livestock Sector with Particular Reference to the International Beef Trade . Göttingen: Institut für Agrarökonomie der Universität Göttingen.
Trychniewicz, E. W., and G. E. Schuh. 1969. "Econometric Analysis of the Agricultural Labor Market." American Journal of Agricultural Economics 51 (November): 770–87.
Tyers, Rodney. 1982. "Effects on ASEAN of Food Trade Liberalization in Industrial Countries." Paper presented to the Second Western Pacific Food Trade Workshop, Jakarta, 22–23 August.
Tyers, Rodney, and Kym Anderson. 1984. Price, Trade and Welfare Effects of Agricultural Protection: The Case of East Asia. Australia-Japan Research Center, Pacific Economic Paper no. 109. Canberra: Australian National University.
United States. 1976. "Export Taxes." An interagency report prepared for submission to the GATT. Pt. 4, sec. G. Mimeographed. Washington, D.C.: Department of the Treasury.
World Bank. 1982. World Development, 1982 . Washington, D.C.: World Bank.
———. 1985. World Development, 1985 . Washington, D.C.: World Bank.
Yeats, A. J. 1976. "Effective Protection for Processed Agricultural Commodities: A Comparison of Industrial Countries." Journal of Economics and Business 29 (Fall): 31–39.
———. 1981. "Agricultural Protectionism: An Analysis of Its International Economic Effects and Options for Institutional Reform." Trade and Development: An UNCTAD Review 3 (Winter): 1–29.
― 332 ―

Eight— Agriculture in the Liberalization Process