U.S. Agricultural Trade With Mexico: What is the Role of Trade Assistance with Our Southern Neighbor?

November 1998                                          SCSB# 390

TRADE, POLICY, AND COMPETITION:
FORCES SHAPING AMERICAN AGRICULTURE PROCEEDINGS

Chapter 7
U.S. Agricultural Trade With Mexico:
What is the Role of Trade Assistance with Our Southern Neighbor?

Glenn C.W. Ames, James E. Epperson, and Manlio Santillan

Introduction

Historically, the United States has been Mexico's primary source of agricultural imports. In 1994, Mexico alone accounted for approximately 10 percent of U.S. farm exports. In the last 20-year period of Mexican agricultural imports, the U.S. share has ranged from 51 to 88 percent of Mexican imports, Figure 7.1. Oilseeds, coarse grains, poultry, dairy products, feeds and fodders, animal fats and oils, vegetable oils, fruits, and preparations have accounted for the largest share of U.S. exports to Mexico (USDA-ERS 1979-1995).

FIGURE 7.1 MEXICAN AGRICULTURAL TRADE AND U.S. MARKET SHARE

Villa-Issa (1990) asserts that the increase in Mexican food imports during the last two decades is the result of past and current agricultural policy and economic conditions. Imports were favored during the first five years of the 1970s with an exchange rate policy that overvalued the Mexican currency. After the devaluation in 1976, there was a trade surplus until 1980 when the agricultural sector experienced a trade deficit. During the 1980s, Mexico's economic policy has mainly focused on controlling inflation, reducing the burden of foreign debt, and avoiding social unrest. Due to the poor performance of the rural sector, food imports have become a key factor in the national food balance.

In 1992, following a severe drought in which agricultural GDP fell 2.9 percent, the Mexican government instituted price controls which greatly impacted food production, processing, and consumption (Tanyeri-Abur and Rosson 1996). Price controls led to production shortfalls in many sectors including the dairy sector and to increased food imports from the United States and other suppliers.

CONASUPO (Compania Nacional de Subsistencias Populares-National Public Food Supply Company) controlled imports of basic food stuffs through a system of tariffs, quotas, and import licenses. Foodstuffs were purchased on the world market and sold at subsidized prices to low-income consumers. Since 1986 when Mexico entered the General Agreement on Tariffs and Trade (GATT), many trade barriers have been lowered. After joining GATT, Mexican tariffs fell from 100 percent in 1985 to 25 percent in 1992 (USDA-ERS, RS-93-2). Mexico eliminated official import and export reference prices, cut overall tariff rates, and reduced the number of commodities subject to import licensing. In the mid-1980s the government, through CONASUPO, continued to be the sole importer of bulk agricultural commodities which continued at the overvalued and controlled exchange rates (USDA-ERS, RS-86-7). By 1988 all price controls were eliminated except for retail fluid milk prices (Tanyeri-Abur and Rosson 1996). In the 1990s, CONASUPO's operations are limited to the purchase and distribution of corn and maize flour, dry beans, and non-fat dry milk at subsidized prices.

Since 1983, Mexico's agricultural imports have benefitted from U.S. agricultural export credit guarantee programs, (GSM-102) short-term credit guarantees, and (GSM-103) intermediate-term credit guarantees. Agricultural export commodities, which further the U.S. government's long-term market development objectives, have been eligible for short and intermediate term credit guarantees. This has boosted U.S. exports to Mexico.

Mexico signed the North American Free Trade Agreement (NAFTA) in 1992. With its implementation beginning in January 1994, Mexican trade liberalization continued. Import licensing restrictions have been replaced with either tariff-rate quotas (TRQs) or ordinary tariffs to be phased out within five to 15 years, depending upon the product (USDA-ERS, WRS-94-2). Other reforms include the publication of a new NAFTA compatible tariff rate schedule and guidelines to administer the TRQs. All of these factors have had an impact on the import demand for food in Mexico during the latter period of analysis.

Given the apparent importance of changes in Mexican economic and trade policy on its international trade, the purpose of this paper is to analyze the impact of these factors on U. S. agricultural exports to Mexico. Mexico was selected because of its relative importance as a major importer of U.S. agricultural goods and because it is now part of NAFTA. Thus, the Mexican market has implications for exports to other Latin countries which may join NAFTA.

Review of Literature

For some time, researchers have tried to explain the growth and decline in U.S. agricultural exports. Regarding trade, one of the more recent variables to be studied is that of exchange rates. Schuh, in 1974, concluded that the exchange rate has been an important omitted variable in the interpretation of U.S. agricultural development and trade problems. After Schuh's statement, numerous efforts were made to quantify the effects of the nominal exchange rate, the real exchange rate, and exchange rate risk on agricultural trade flows (Pick 1990).

In 1980, Collins, Meyers, and Bredahl investigated the effects of exchange rates on selected agricultural commodities. Their test included 37 foreign countries and four commodities: wheat, corn, soybeans, and cotton. Results indicated that exchange rate effects on real U.S. commodity prices are smaller under free trade and real price insulation policies, but rise substantially as nominal price insulation policies become more prevalent.

The decline of U.S. agricultural exports in the early 1980s was attributed to the appreciation of the dollar. Batten and Belongia (1984) addressed this problem, attributing part of the blame to the fundamental differences between nominal and real exchange rates. Using data for 1981-83, they found that real exchange rates were negatively related to exports, but their impact was dominated by the level of real GNP of the importing countries. In their conclusions, they suggested a weak link between U.S. money growth and real exchange rates, and indicated that foreign income has been the primary determinant of agricultural exports.

In 1988, Brada and Mendez, working on the effects of exchange rate regimes, found that bilateral trade flows among countries with floating exchange rates were higher than among those countries with fixed rates. Cushman (1988), studying U.S. bilateral trade flows and exchange rate risk during the period of floating rates, found a significant negative effect. This finding was consistent with Brada and Mendez's work in the sense that they observed that exchange rate uncertainty does lower the volume of trade among countries, regardless of the nature of their exchange rate regime. In 1982, Cushman investigated the effects of real exchange rate risk on international trade. This study showed significant negative effects on trade quantity from exchange risk in fourteen bilateral trade flows of industrialized countries.

Subsequent research by Batten and Belongia (1986) on monetary policy, real exchange rates, and U.S. agricultural exports, concluded that there was little question that the real value of the dollar reduced the volume of U.S. farm exports after 1981. They also found that the export volume appeared to be relatively insensitive to changes in U.S. farm prices but responded more to changes in foreign real GNP and the real exchange rate.

Recent studies on exchange rate risk by Pick (1990) and Bahmani-Oskooee and Ltaifa (1992) found comparable results. Pick analyzed the effect of exchange rate risk on U.S. agricultural trade flows for 10 countries. He observed that exchange rate risk was not significant in seven developed countries, but adversely affected U.S. agricultural exports to three developing countries included in the study. On the other side, Bahmani-Oskooee and Ltaifa, using the aggregate exports of 19 developed countries and 67 developing countries, noted that exchange rate uncertainty was detrimental to the exports of both developing and developed countries. However, exports of developed countries were less sensitive to exchange rate risk than that of developing countries. Moreover, within developing countries, those which fixed their exchange rates to one major currency were found to be subject to less risk than the other developing countries.

Shane (1990), using a general equilibrium model to simulate the effects of exchange rate devaluation on U.S. exports, found that a 1 percent depreciation leads to a greater than 1 percent rise in U.S. agricultural exports. His conclusions have important implications for U.S. trade in the 1990s since the dollar fluctuates in value against major world currencies.

The Mexican peso depreciated against the U.S. dollar after it was allowed to float in 1976. In the 1980s and early 1990s there were subsequent devaluations of the currency. The Mexican government introduced the new peso in an attempt to stabilize the currency and break inflationary expectations. These currency devaluations have impacted the country's ability to import food products. Thus, export assistance is important in Mexico's continued reliance on U.S. food exports.

U.S. Agricultural Trade Assistance and Trade Policy Reforms

The United States has assisted Mexican imports of food grains and oilseeds since 1980 through GSM-102, the primary export credit assistance program. GSM-102 guarantees export credit from private banks for periods of six months to three years. Interest rates are generally lower than commercial rates. In the case of default by importers, the United States covers the loss to the lender (Tweeten 1992). In 1992, Mexico purchased more than $1.1 billion worth of U.S. agricultural commodities under GSM-102, or 29 percent of total farm imports from the United States. By 1993, GSM-102 assisted exports to Mexico accounted for 33 percent of its $3.6 billion in agricultural trade with the United States (Figure 7.2). Between 1990 and 1994, Mexico received $6.041 billion in export credit guarantees (Table 7.1). This represents 30.6 percent of total allocations of GSM-102 to all countries during the 1990s.

FIGURE 7.2 U.S. AGRICULTURAL EXPORTS TO MEXICO

Mexico also benefitted from other subsidy programs. During the 1987-89 period, the Export Enhancement Program (EEP) played a role in expanding U.S. wheat exports to Mexico. The Vegetable Oil Export Enhancement Program and the Sunflowerseed Oil Assistance Program (SOAP) were also used to expand edible oil exports to Mexico (USDA- FAS, FOP-1-96).

GSM-102 export credit guarantees were mainly applied to imports of corn, sorghum, soybeans, soybean meal, wheat, rice, and tallow (Figure 7.3). In 1990, 85 percent of the credit was applied to these basic food commodities, 78 percent in 1991, 83 percent in 1992, 73 percent in 1993, and 78 percent in 1994 (USDA, FAS). The rest of the annual credit support was applied to a wide range of other agricultural commodities. Export credit guarantees supplied under the GSM-103 for Mexico accounted for $81.3 million, mainly applied to imported cattle for breeding purposes (USDA, Foreign Agricultural Service).

FIGURE 7.3 MAJOR U.S. COMMODITIES EXPORTED TO MEXICO
UNDER THE GSM-102 PROGRAM

During the 1990-94 period, the share of agricultural exports to Mexico ranged from 27 to 50 percent (Table 7.2). It is evident that the credit guarantee programs play an important role in maintaining and expanding U.S. agricultural exports to Mexico. The value of U.S. export credit guarantees in the estimation of Mexico's import demand for agricultural imports is included in the following analysis.

Conceptual Framework and the Model

A model that examines the impact of relative prices, exchange rates, levels of economic growth, export subsidies, and other relevant factors on U.S. agricultural exports may be specified as follows:

EX_t = f(RT_t, IN_t, EX_t-1, GSM_t, DF), (7.1)

where EX represents the real per capita value of U.S. exports to Mexico (Mex.$), RT is the Mexican–U.S. real exchange rate (Mex.$/U.S.$); IN is real per capita Mexican income (Mex.$), GSM represents the real per capita amount of interest rate guarantees allocated to Mexican imports under GSM 102 and 103, DF is a dummy variable that accounts for the change from a fixed to floating exchange rate system (1 is fixed, 0 otherwise), and t represents time in years. The base year for all real values is 1980.

The first two independent variables of the equation, exchange rate (RT) and income (IN), were included to account for relative prices and economic growth as per the previous literature. The variable for lagged exports (EX_t-1) was introduced in order to account for export adjustments to income and relative price changes over time (Miller and Fratiani 1974). The export credit subsidy (GSMt) was added to capture the impact of lower costs of imports due mainly to short-term credit guarantees. Finally, the dummy variable (DF) was used to indicate a fixed versus floating exchange rate since the first two years of the data set encompassed a fixed exchange rate regime.

Following the works of Batten and Belongia (1984 and 1986), Belongia (1986), Brada and Mendez (1988), and Bahmani-Oskooee and Ltaifa (1992), a log linear model was used to estimate the effects of exchange rates, income, and other relevant factors on U.S. agricultural exports to Mexico for the period 1974-1995 according to the specification of equation 7.1:

lnEX_t = B_o + B₁ ln(RT_t) + B₂ ln(IN_t) + B₃ ln(EX_t-1) B₄ln(GSM_t) + B₅DF + U_t. (7.2)

Ordinary Least Squares (OLS) was used to estimate the equation's parameters.

Data and Sources

Annual data from 1974 to 1994 were used in the empirical analysis. The value of U.S. agricultural exports to Mexico was obtained from Foreign Agricultural Trade of the United States, published by the U.S. Department of Agriculture. The Foreign Agricultural Service provided the data on GSM 102 and 103 by years and commodity (1984-1994).

Nominal exchange rate values of the Mexican peso with respect to the U.S. dollar were obtained from International Financial Statistics. The exchange rates were deflated with Mexican/U.S. CPI's (1980 base). The computation follows Madura (1992): S* = Sm/(Pm/Pu), where S* is the real exchange rate, Sm is the nominal exchange rate, Pm is the consumer price index in Mexico and Pu is the consumer price index in the United States. Mexican GDP data were obtained from International Financial Statistics. Indices for the U.S. and Mexican CPIs were also obtained from International Financial Statistics.

Results And Interpretation

The results of estimating equation (7.2) are presented in Table 7.3. Because of the log-linear specification, the coefficients are also partial elasticities which measure the percentage change of the dependent variable resulting from a 1 percent change in one of the independent variables, keeping all other independent variables constant.

The exchange rate was expected to have a negative effect on U.S. agricultural exports to Mexico. A higher exchange rate (more pesos per dollar) makes U.S. goods more expensive in pesos resulting in less imports from the United States and vice versa. While, the coefficient for the exchange-rate variable RT_t was not significant, the coefficient had the expected negative sign. In addition, CONASUPO imported bulk commodities at overvalued exchange rates that could also lessen the impact of exchange rates on imports.

As expected, the income coefficient was positive and significant (a = 0.01). The coefficient indicates that a 1 percent increase in the real per capita income leads to an increase in U.S. agricultural exports to Mexico by 2.55 percent. This finding is similar to that found by Pick (1990) for some countries. He worked on bilateral trade with quarterly data. Estimated income elasticities were 2.97, 2.74, 2.86 for United Kingdom, Australia, and Japan, respectively. These results differ in magnitude to those found for Batten and Belongia (1984 and 1986) and Bahamani-Oskooee and Ltaifa (1992). The results of Batten and Belongia, also working with quarterly data, were 1.32 and 1.325. The results of Bahamanni-Oskooee and Ltaifa, who worked with cross-sectional data, were around 1.25 for developed countries and less than 1.00 for developing countries. As the magnitude of the partial elasticity coefficient is relatively large (elastic), Mexican food imports appear very sensitive to real income growth.

The coefficient for the lagged export variable (EX_t-1) was not significant. Seemingly the trade adjustment process is unimportant in this case or is being dominated by other factors.

The coefficient of the export credit system (GSM) was positive as expected and also significant (a = 0.01). The coefficient indicates that a 1 percent increase in the real per capita allocation of export credit guarantees leads to an increase in U.S. exports to Mexico by 0.11 percent. This was expected because export credit guarantees would lessen the cost of food imports vis-a-vis lower interest rates.

The coefficient for the dummy variable (DF) was not significant. Apparently, only two observations with fixed exchange rates were insufficient to materially affect the results of the analysis.

An R² of 0.56 is an indicator that other forces are also at work regarding U.S. agricultural exports to Mexico. It may well be that long-term U.S. policies affecting Mexico have a positive impact on Mexican food imports.

Conclusions

The study examined the effects of exchange rates, levels of economic growth, and export credit subsidies on U.S. agricultural exports to Mexico. According to the empirical results, the effect of the exchange rate was not significant in the flow of trade during the 22-year period included in the analysis. This finding is similar to some previous studies in which the exchange rate variable was dominated by real income. Indeed, the income variable in this study was significant. The income elasticity suggested that a 1 percent increase in real per capita income in Mexico would lead to an increase of 2.55 percent in U.S. agricultural exports.

The results are also consistent with an agribusiness survey reported by Conley and Le Boulanger (1996). After the peso devaluation of December 20, 1994, they surveyed executives of agribusiness firms exporting grain, soybeans products, livestock feed, farm supplies, and food items to Mexico. These firms managed their foreign exchange exposure in different ways. Only the largest firm accepted pesos for payment. It immediately hedged its exposure in the foreign exchange market. The smaller firms accepted payment only in U.S. dollars and "...obtained [U.S.] government credit to guarantee the payment of the sales" (Conley and Le Boulanger 1996, p. 3). Thus, devaluation had almost no real negative consequences for these firms.

In addition, the credit guarantee program eliminated payment risks. Exports to Mexico recovered quickly (within six months) after the devaluation. These firms intended to continue exporting to Mexico but remain cautious and cognizant of the risks. Thus, the GSM program enhanced U.S. food and fiber exports to Mexico as this research indicates.

The results of this analysis reflect the importance of economic development, and rising income in developing countries, for growth in U.S. exports of agricultural commodities. Further, export credit guarantees appear to eliminate some of the financial risk in dealing with the Mexican market.

References

Document Prepared by:
Leigh H. Stribling, lstribli@acesag.auburn.edu
Alabama Agricultural Experiment Station
Auburn University

Chapter 7 U.S. Agricultural Trade With Mexico: What is the Role of Trade Assistance with Our Southern Neighbor?

Introduction

Review of Literature

U.S. Agricultural Trade Assistance and Trade Policy Reforms

Conceptual Framework and the Model

Data and Sources

Results And Interpretation

Conclusions

References

return to top of page return to contents

Chapter 7
U.S. Agricultural Trade With Mexico:
What is the Role of Trade Assistance with Our Southern Neighbor?

return to top of page
return to contents