The Georgia Agricultural Experiment Stations
College of Agricultural and Environmental Sciences
The University of Georgia
Research Bulletin Number 436
January, 1999

An Analysis of Foreign Direct Investment in the U.S. Fruit and Vegetable Industry

J. C. Burnham, Graduate Research Assistant
J. E. Epperson, Professor

Department of Agricultural and Applied Economics
Athens Experiment Station
Athens, Georgia

Gale A. Buchanan, Dean and Director
Gerald F. Arkin, Associate Director Northern Region
Jerry A. Cherry, Associate Dean and Senior Associate Director

ISSN 0072-128X

Contents

Introduction
U.S. Fruit and Vegetable Company Survey
Econometric Analysis
Summary
References

Introduction

The United States is a major producer of fruits and vegetables, valued at more than $20.3 billion in 1995 (U.S. Department of Agriculture, April and November 1995). The United States not only pro-duces for the domestic market, but is also a major exporter of fruits and vegetables. Leading export customers included Canada at $1.25 billion, Japan at $698 million, Western Europe at $155 million, and Hong Kong at $138 million in 1995 (U.S. Department of Agriculture, September/October 1995).

The U.S. fruit and vegetable industry has been largely seasonal in nature with respect to production. Production has occurred primarily in the warmer climates of the United States in the winter and early spring. With the passing of spring and into the summer, production advances northward. With the arrival of fall, the movement of production reverses course and moves southward (U.S. Department of Agriculture, Agricultural Marketing Service, 1983-1986).

In the past, U.S. imports of fruits and vegetables have been mainly supplemental. That is, in general, foreign-produced fruits and vegetables have entered the United States when production in the United States has not been in season or when prices have been sufficiently high to attract additional supplies from foreign sources. A primary example is the U.S. winter fresh vegetable market. During the winter season, large supplies have come primarily from Mexico (How).

Imports of fruits and vegetables in dollar value have risen steadily during the past 10 years from nearly $1.8 billion in 1986 to almost $2.7 billion in 1995 dollars (U.S. Department of Agriculture, September/October 1986 and 1995; U.S. Department of Commerce). Major foreign suppliers of fruits and vegetables to the United States in 1995 and their relative importance are shown in tables 1 and 2.

With the passage of time rapid advances in technology have occurred. Advances in information and communication systems, distribution systems, and large-scale production systems have changed markets dramatically, making them more global and less regional in scope. With such advances, distance between production and consumption locations becomes less and less of a barrier in the competition for markets (Cook). This changing phenomenon has an effect on the way fruit and vegetable firms compete in a widening market arena.

Moreover, artificial trade barriers between countries are falling. Important examples of such action are the General Agreement on Tariffs and Trade (GATT), the North American Free Trade Agreement (NAFTA), and the Caribbean Basin Initiative (Hillman; U.S. Department of Agriculture, 1994; U.S. International Trade Commission).

Trade negotiations typically involve negotiations on tariff and nontariff barriers with varying degrees of success. Tariffs are simply taxes on imports, while nontariff barriers come in many different forms. Nontariff barriers include production subsidies, licensing, strategic customs procedures, phytosanitary standards, packaging and labeling regulation, and credit restrictions (Hillman).

GATT was organized in 1948 and encompasses more than 90 countries. Several multilateral trade negotiation rounds - the most recent being the Uruguay Round - have been conducted under the auspices of the GATT (Stone).

Although much progress has been made under GATT, negotiations have been far more sweeping under NAFTA in the reduction of trade barriers. NAFTA, which is an agreement among Canada, Mexico, and the United States, was implemented on January 1, 1994. Under NAFTA, all nontariff barriers are to be converted to tariffs, and all tariffs are to be reduced to zero within 15 years, depending on the sensitivity of the product (American Farm Bureau, 1995). NAFTA also provides for the free movement of capital without restrictions on its ownership and control among the participating countries.

Thus, under NAFTA, fruit and vegetable firms in the participating countries are free to buy and sell produce without artificial barriers. Moreover, these firms are free to produce, build new plants and facilities, and buy or sell all or parts of other firms any-where within the participating countries.

Preliminary NAFTA negotiations drew the attention of other Latin and Caribbean countries. As a result, attitudes toward trade became more open, leading to the hope that they may at some point become a part of NAFTA. As an example, in December of 1994, the countries involved in NAFTA met with Chilean officials to begin talks that could lead to Chile's accession into NAFTA. Immediately, concerns were raised about the economic effects on the U.S. horticultural industry (U.S. Department of Agriculture, November 1995). Even without accession, horticultural trade between the United States and most of Central and South America has increased since the initiation of NAFTA (U.S. Department of Agriculture, September/October 1995). Combining easier foreign investment and decreased tariffs with technological and competitive innovations indicated earlier, changes in market conditions become even more important to fruit and vegetable producers in the United States.

Problem Statement

Historically, changes in international laws and treaties among countries have had significant impacts on trade and investment practices (Husted and Melvin). For example, when Mexico began liberalizing its laws against foreign ownership of property in the late 1980s, it spurred increased investment from the United States and other countries. U.S. investment in Mexico's agribusiness increased five-fold from 1987 to 1992, mainly because of changes in Mexico's land tenure and investment laws (Bolling and Valdes). With the initiation of NAFTA, further increases in investment and trade dominated the new economic relationship between the United States and Mexico (U.S. Department of Agriculture, 1994).

Foreign Direct Investment (FDI) is the investment by a company, group, or individual in new facilities, existing enterprises, a share of existing enterprise, or land or natural resources within another country (Bolling and Valdes). There are many reasons that companies engage in FDI. This study attempts to identify some of them within the U.S. fruit and vegetable industry. Of primary interest is determining whether changes in trade barriers have a significant effect on FDI affecting ownership and location of firms and competition within the produce industry.

Objectives

This research analyzes the incentives and factors affecting decisions to engage in FDI by U.S. fruit and vegetable firms. The primary objectives are to:

1. obtain data concerning the operations of U.S. produce firms, including importing and FDI activities;
2. analyze firm-level data regarding trends and patterns in decision making concerning factors that could affect FDI;
3. develop an econometric model estimating the probability that firms with certain characteristics and circumstances are engaged in FDI; and
4. summarize the findings and discern implications for the future of the U.S. produce industry.

There may also be some hidden barriers to trade and special characteristics that affect the FDI activities of some companies. Integrating this with the primary objectives, U.S. produce firms could benefit from using the findings in this study to assist them in future decision making.

Procedure

The data for this study were obtained through development and administration of a questionnaire to U.S. fruit and vegetable growers/importers. The questionnaire was designed to provide information directly beneficial to purely domestic firms, as well as others. The data were obtained through telephone interview, followed by facsimile, and additional telephone contacts.

The data set were analyzed and patterns observed. The observations were first differentiated on the basis of the presence of FDI. Special attention was taken with regard to possession of the characteristics expected from the literature. It was also expected that some unanticipated features would evolve.

The results are documented, followed by thorough investigation and reporting of the patterns identified. Next, special concerns that appear to be common and significant are addressed and their relevance to foreign investment discussed. Finally, an econometric model explaining the probability of FDI is developed.

Conclusions from the analysis are drawn. The relevance and implications of the results and conclusions for U.S. produce companies and policy makers are then presented and deliberated.

Literature Review

It is important to point out, as this research focuses on investment in agricultural production, that most of the literature is applicable to investment in both production and food processing. These two areas are classified as such when needed. While there is a limited amount of literature on agricultural foreign investment, there is an adequate amount of literature on related topics, such as foreign investment in general with specific countries, the effects of trade agreements such as NAFTA, and corporate taxation. A review of the literature in these areas provides a foundation for the upcoming analysis.

FDI in Mexico

An article by Bolling and Valdes gives an over-view of FDI by the United States in Mexico's agribusiness, describes several factors affecting FDI in Mexico, and gives a historical analysis of investment by the United States. Bolling and Valdes note several factors affecting U.S. investment in Mexico. These include "(1) Mexico's growing population and market for consumer goods; (2) Mexico's warmer climate, which allows off-season production and exports to the United States; and (3) developments in Mexico's macroeconomy, such as Mexico's devaluation of the peso and continued current account deficit."

They also note that Mexicans may be improving their diet and thus increasing the demand for higher valued foods.

A publication by Handy is mainly concerned with investment for the production of processed food and beverage products. Handy contends that investment in Mexico's processed food and beverage industry is mainly to satisfy local demand. The Mexican economy has achieved an average annual real growth rate of 3.8% from 1989 to 1992 with the liberalization of the nation's foreign investment rules, allowing 100% foreign ownership. Thus, Handy contends it is easy to see why food manufacturing firms would invest primarily to serve Mexico's domestic market. However, Handy is considering primarily processed instead of fresh fruits and vegetables.

The NAFTA and FDI

NAFTA: An Early Assessment, a report by the Economic Research Service (ERS), U.S. Department of Agriculture, lists some interesting details con-cerning NAFTA and foreign investment in Mexico and also gives a historical background of the increased investment in Mexico because of changes in investment laws. The report notes that while a significant increase in investment in 1994 may not all be directly related to NAFTA, "the lowering of trade and investment barriers in Mexico accounts for much of U.S. direct investment flows and exports to Mexico." Also, as Mexico's laws against foreign investment began to diminish, the report indicates that U.S. investment in Mexico tripled from 1987 to 1993. While easier investment laws are rather obvious factors of increased investment, the lowering of other trade barriers such as tariff rates and their affect on investment merits attention. The report by the ERS also says increased investment is likely to continue as NAFTA's schedule of tariff reductions and structural reforms take effect.

Along with decreasing tariff rates, NAFTA will make investment in Mexico even easier than it was after the changes in Mexican law in the late 1980s and early 1990s. A 1991 publication by the American Farm Bureau mentions how NAFTA has already begun to affect the decision making of multinational broccoli growers by stating, "it is of note that the industry has moved toward a structure of production-sharing, with plants in the United States mixing imported and domestic products." Rugman also addresses some of the economic effects of NAFTA as it relates to the international movement of capital. Rugman says, "NAFTA will encourage more liberal international investment flows in North America, as well as a more efficient allocation of productive resources throughout the continent."

Taxes and FDI

There is an abundance of literature on the subject of differences in corporate taxation between countries and its effect on the foreign investment practices of firms. Most literature concludes that multinational corporations (MNCs) do not make their decisions on where to locate based on a comparison of U.S. and foreign tax policies. However, there is a substantial amount of evidence in the literature that tax policy affects the capital flows once a foreign operation is established.

Jun, in his study, attempts to estimate the sensitivity of U.S. direct investment capital flows internationally to the U.S. net rate of return. Jun says a rational profit-maximizing firm will try to optimize over the capital allocation between the parent and the subsidiaries given different rates of return and sources of funds between countries. Jun integrates the subsidiary's foreign operation with its parent's domestic operation, thus allowing optimization over every relevant decision variable and allowing a relation between tax policy and a firm's marginal financing of foreign investment. He uses actual capital expenditures by subsidiaries, along with other data such as relative rates of return, to estimate the tax effects on investment flows using regression analysis.

Jun differentiates maximum potential net return from the net return on U.S. nonfinancial corporate capital to measure the effect of different tax policies on new marginal investment rather than existing capital. He concludes that changes in U.S. tax policy have direct, significant effects on MNC's foreign investment. The coefficient on the maximum potential net return variable was significant and larger than the coefficient measuring the return on existing capital. This suggests that the investment decision is made solely by the corporation, and it is better to measure the yield on new marginal investment rather than the existing capital.

Grubert and Mutti analyze capital allocations across borders among MNCs and how it is affected by taxes. They conclude that the incentive to invest in a manufacturing affiliate outside of the United States becomes larger at lower tax rates. After regressing stocks of net plant and equipment against several control variables and tax terms, they state that "a reduction in the host country tax rate from 20% to 10% is projected to increase U.S. affiliates' net plant and equipment in the country by 65%." Grubert and Mutti defend the size of this estimate by stating that this comparison has an effect relating the cost of imports to domestic production and a "low-tax country attracts investment not only to serve its own market, but other markets as well through exports." In conclusion, Grubert and Mutti note that taxes and tariffs often have very similar effects on the FDI practices of MNCs.

U.S. Fruit and Vegetable Company Survey

The preceding literature on foreign investment and how it relates to the fruit and vegetable industry in the United States has provided a framework for the analysis and has given a general guideline for the type of information needed. The data for the study were obtained through a questionnaire administered to U.S. fruit and vegetable growers/importers. The section explains how the survey sample was obtained and explains the survey techniques, as well as summarizes, discusses, and presents implications for the results of the survey. Further analysis via econometrics will be presented in the next section.

Survey Population

The types of fruit and vegetable firms targeted in the survey were firms that produce or obtain the same commodity both domestically and internationally. Firms with this characteristic are labeled produce growers/importers. These firms were targeted because they were deemed most likely to be involved in foreign investment.

According to the Produce Reporter Company, there were almost 480 produce importers in the United States in 1995. At least 80 of these were importers only. This status nullifies their inclusion leaving a maximum of 400 growers/importers (Produce Reporter Company).

Survey Techniques

A firm-level, cross-sectional analysis was done, giving specific information concerning the possible ownership and globalization, as well as the locational advantages of FDI. The questionnaire solicited annual figures for the most recent year. Because of seasonality, most fruit and vegetable firms in the United States begin their year at a different time than January, the beginning of the calender year. Information was typically obtained from a particular month in 1994 to the same month in 1995.

The questionnaire included items such as a firm's total sales, total imports, origin of imports and whether they are the result of purchases or investment, and the costs of obtaining commodities domestically and internationally. Most questions, except items such as total sales, were in percentage form. This allowed the participants to answer the questions more easily, and it provided clearer responses.

The questionnaire was administered through telephone interviews with follow-ups via facsimile and telephone. The participants were assured of confidentiality, as the analysis does not reveal details about any particular firm.

Survey Results

More than 90% of the population of approximately 400 U.S. fruit and vegetable growers/ importers was contacted. Complete questionnaires and, therefore, usable observations, were obtained from 81 firms. The data are clear, concise, and easy to interpret allowing quantification and analysis.

The fruit and vegetable firms participating in the survey had average annual gross sales of $26.9 million. Firms involved in FDI reported average sales of $29.6 million and non-investors reported an average of $22.8 million in annual gross sales.

The average percentage of total sales domestically was 88% for all survey participants. Of the 81 firms surveyed, 57 sold at least 85% of their total output in the United States. The remaining 24 firms reported an average of 71% in domestic sales. The average percentage of domestic sales as a result of imports was 51.5%. The foreign investors indicated that imports accounted for an average of 60.7% of total domestic sales, as opposed to 37.4% for non-investors.

The survey requested information concerning the participants' most relevant sources of foreign supplies. A majority of the imports from 49 of the 81 companies surveyed were from Mexico. Of these 49 firms, 30 were engaged in foreign investment in Mexico, and the remaining 19 were strictly importers. Most of the imports from 12 of the 81 companies came from Chile. Of the 12, three were investors, and nine were strictly importers. A majority of the imports from 7 of the 81 firms came from Guatemala. All seven of these firms were engaged in FDI. The countries from which nine firms reported receiving a majority of their imports were Costa Rica, Nicaragua, the Dominican Republic, the Bahamas, Honduras, and Colombia. Seven of the nine firms were investors, while only two strictly imported from these countries. Two firms imported from Ecuador because of foreign investment, while one company imported strictly from Canada, and another one from India only. Table 3 summarizes the results.

Aggregation reveals that firms reported receiving an average of 81.7% of their total imports from the countries indicated in Table 3. The firms involved in foreign investment in those countries received an average of 88% of their imports from these sources, and those that strictly imported received an average of 66% of their imports from their most relevant country. Further aggregation indicates that 65 of the 81 companies surveyed got most of their imports from countries that were generally charged a tariff. Of the 65, 36 were investors and 29 were strictly importers. Of the 81 companies, 16 received imports from countries that were not charged a tariff. Of the 16, 14 were investors in their country of interest and only two were strictly importers.

The survey revealed that an average of 75.7% of the participants' total sales in countries other than the United States resulted from U.S. supplies exported to the relevant country. There were only minor differences in the responses from foreign investors and non-investors, as they indicated that averages of 73.9% and 79.3%, respectively, of their international sales were from U.S. production.

Some 49 firms indicated they were engaged in FDI, while 32 reported they strictly imported fruits and/or vegetables and were not engaged in FDI of any kind. All 49 firms were involved specifically in foreign production practices, and some also contracted for production.

Most of the foreign-produced supplies for 29 firms came from Mexico. Seven firms indicated that most of their foreign production took place in Guatemala, and three reported Chile. Another three indicated Costa Rica as their most important country for FDI. Two firms indicated Ecuador, while Honduras, Colombia, Nicaragua, the Bahamas, and Venezuela were the most relevant locations of foreign production for one firm each. Table 4 summarizes the results.

The survey revealed that an average of about 53% of the total supplies from firms engaged in foreign investment resulted from this activity in 1994-1995. Of the firms that practiced FDI in 1994 and 1995, more than 95% of those surveyed cited cheaper labor and, to an extent, seasonality as the most important factors influencing their historical FDI practices. Three firms included risk reduction as one of the reasons for their FDI, while only one firm reported that demand in the target country influenced its decision to produce internationally.

Survey participants who were engaged in FDI reported that an average of 82.4% of their supplies received from their most relevant source country resulted from their investment there, and the remaining 17.6% was purchased from local producers in the source country. Regarding foreign ownership, all of the firms engaging in FDI indicated that, to their knowledge, they could legally own 100% of their operation in all of the countries included in the survey.

Of the supplies produced internationally because of FDI, an average of 10% sold locally, an average of 85.6% was exported to the United States, and an average of 4.4% was exported to another country, usually Canada or countries in Europe.

Survey participants reported imports on average 14.4% cheaper than produce obtained domestically in the United States. Foreign investors perceived U.S. imports to be an average of 17.7% cheaper than U.S. domestic sources, as opposed to 9.4% for firms strictly importing. Table 5 summarizes all relevant responses from foreign investors and those from firms that strictly imported fruits and vegetables.

Several companies indicated their participation in the practice of providing start-up funds to determine whether certain fruits and vegetables could be grown profitably internationally. Table 6 summarizes the results.

Part of the survey requested information concerning whether the participants imported their fruits and vegetables seasonally or year around. Of the 81 firms surveyed, 37 indicated they imported fruits and vegetables year around, and the remaining 44 firms imported seasonally, typically when their commodities were not available domestically. Of the 37 firms that imported year around, 29 were involved in FDI, and the remaining eight were strictly importers. Of the 43 seasonal importers, 20 were foreign investors, and 23 were strictly importers and thus not involved in foreign production.

The survey participants indicated tariffs as the most important trade barrier they faced. They cited phytosanitary regulations as the second most common problem. Some 10% of the investors listed this restriction as their most important trade barrier, while 16% of the firms that strictly imported cited this as their most prevalent concern. Other rules and regulations set by the U.S. Department of Agriculture and the Environmental Protection Agency were indicated as the third most common trade barrier, followed by paperwork and other policies and procedures. Table 7 summarizes these results.

The responses were almost identical from foreign investors and non-investors concerning expectations that NAFTA would directly affect their operations. Of the firms engaged in FDI, 48% indicated they would be affected by NAFTA, and 50% of the firms not engaged in foreign production reported they would be affected. Specifically cited reasons included lower tariffs, easier investment, cheaper labor, and easier trucking.

Of the firms surveyed not involved in foreign production, 38.3% reported they would consider FDI because of NAFTA or a similar trade agreement that would involve their country of interest. Specific reasons cited included lower tariffs and easier investment.

Special Concerns and Details

Concerning the international trade relationship between Mexico and the United States, many com-panies noted changes that have already begun to affect their operation because of NAFTA. Several fruit and vegetable firms indicated NAFTA has improved international transportation between Mexico and the United States. However, several other firms cited problems with exporting to and from Mexico because of NAFTA. The most common problems included increases in the amount of paperwork required as well as general "red tape."

Several firms expressed concern that U.S. and Mexican authorities often find it hard to collaborate their efforts at the border in an efficient and effective manner.

Many firms provided additional details concerning their foreign production efforts. Fruit and vegetable producers primarily from Texas and Arizona reported that the availability of water is an incentive to produce in Mexico. With the future of the water supply in south Texas and Arizona in question, these firms have found it less risky to produce in Mexico. Firms reporting the least expensive foreign production were involved in FDI in Ecuador. However, transportation and other costs minimized this benefit. Most of the respondents producing in Mexico, Honduras, and Guatemala indicated they import almost all their capital from the United States. Finally, some firms indicated they engaged in foreign production primarily for diversification and other risk reduction purposes.

Discussion of Results

More than 60% of the firms surveyed were primarily involved in Mexico, and 61% of these were foreign investors. This involvement can probably be attributed to Mexico's proximity to the United States, its size relative to other fruit- and vegetable-producing Latin American countries, as well as the availability of labor. Concerning FDI, these incentives are locational.

In the second most popular country, Chile, the low percentage of investors (25%) is probably because most commodities produced in Chile were charged a relatively high tariff upon entering the United States. The average U.S. tariff on Chilean imports was 9.23 cents/kg. This is significantly higher than the average U.S. import tariff of 2.46 cents/kg on the survey participants' most relevant commodity. Chile is locationally disadvantaged as well. In contrast, 100% of the firms involved in Guatemala were involved in FDI. Imports from Guatemala were not charged a tariff because of the Caribbean Basin Initiative (CBI) (U.S. International Trade Commission, 1995). Furthermore, more than 77% of the firms that imported from other countries not charged a tariff by the United States, because of the CBI, were investors. These results support the hypothesis that tariff rates may significantly affect FDI.

The amount of the import tariff charged each commodity upon entering the United States is a locational factor that not only separates countries, but commodities as well, according to foreign production practices. Among survey participants that specialized in production and importation of grapes - the commodity with the highest tariff rate - only 17% indicated they were involved in FDI. Companies concerned mainly with green onions and berries - the two commodities with approximately the lowest average tariff rates - reported 93% and 86%, respectively, involvement in FDI for production of these commodities. These results provide further support of the hypothesis that FDI responds in a negative way to tariff rates within the U.S. fruit and vegetable market. The significance of the relationship between tariff rates and the probability of FDI are evaluated further in the next section.

The survey revealed an additional locational incentive. Firms engaged in foreign production reported receiving their product an average of 8.3% cheaper from the country of origin than firms that strictly imported. Most of the firms indicated this stemmed primarily from labor costs. Also, seasonality adds another locational incentive.

It is clear the decision to engage in FDI is not to satisfy demand in the target country. The firms reported an average of 88% of their sales were in the U.S. domestic market and also indicated an average of 85.6% of their product produced internationally was exported to the United States. Furthermore, more than half of all U.S. domestic sales resulted from imported product, firms reported. Foreign investors indicated an average 23% more U.S. domestic sales from imports than did non-investors. This difference for investors versus non-investors indicates a globalization incentive for FDI. These results show most firms participating in the survey were not interested in Latin America as a market for fresh fruits and vegetables. This situation eliminated a majority of FDI incentives associated with superior marketing strategies, superior technology for product and process differentiation purposes, and other monopolistic advantages associated with gaining market share in a foreign country.

The survey participants involved in FDI received an average of 22% more of their product from a particular source - country indicated as their most relevant source of imports - than participants not involved in FDI. This result indicates that firms receiving a majority of their imports from one place may have more of an incentive to engage in foreign production there. This could stem from non-locational incentives, such as marketing, economies of scale, and other ownership-specific advantages.

Another ownership-specific advantage, that of firm size, may be important. Average gross annual sales for firms involved in FDI were $6.8 million more than for firms not involved in FDI. Larger firms appear to have more incentive to engage in FDI.

Production of different types of produce could result in ownership-specific and globalization incentives as well. It may be more favorable to diversify locations of production for certain commodities because of risk characteristics and other factors, indicating globalization incentives. Ownership-specific advantages may manifest through international production of certain types of fruits and vegetables, allowing economies of size, competitive advantage, or increased efficiency related to the types of commodities produced.

Implications

The survey results indicated many characteristics present among U.S. fruit and vegetable producers engaged in FDI that were not as prevalent among firms that strictly imported. For example, firms engaged in FDI indicated higher average annual gross sales, a larger amount of domestic sales from imports, a larger percentage of imports received from one particular place, lower tariff payments, and less expensive imports than firms that strictly imported. Meanwhile, a large percentage of firms reported starting or expanding their international presence because of NAFTA and a more globally oriented marketplace. Thus, firms engaging in FDI have ownership-specific, globalization, and locational advantages over those that do not participate.

Produce firms exhibiting one or more of the profile characteristics listed above could possibly realize substantial gains through FDI. This is especially important for producers, domestic and international, that are facing increased competition from less expensive imports because of NAFTA. A large percentage of the survey participants indicated this was a major concern. The econometric analysis to follow provides quantitative discrimination of factors affecting FDI, as well as additional information affecting the foreign production practices of U.S. fruit and vegetable firms.

Econometric Analysis

The survey results previously presented indicated several patterns that reflect possible incentives among U.S. fruit and vegetable firms to engage in FDI and indicated that the objective of such investment in agricultural production was primarily to satisfy demand in the United States. The econometric analysis that follows explains the probability that a firm will engage in FDI given the explanatory variables derived from the survey results and the tariff rates associated with the most important commodities. The probit model estimates the probability that a firm is engaged in FDI.

Model Specification

Table 8 summarizes the variables included in the probit model. The variables are discussed in turn.

The dependent variable (INV) is binary in form. Firms reporting that they strictly imported fruits and/or vegetables in 1994-1995 and were not involved in FDI for the production of their commodities were assigned a value of zero. Firms reporting involvement in foreign production of fruits and/or vegetables during this time period were assigned a value of one. Measures of the levels of investment were not obtainable. Most firms found it difficult to measure the percentage of their total assets used in foreign production.

The explanatory variables, as indicated earlier, result from the survey responses previously discussed as well as tariff rates (LTAR) indicating the import tariff associated with a firm's most important commodity. LTAR is an explanatory variable designed to capture the incentive for a firm to engage in FDI given lower tariffs charged on imports into the United States. This was expected to be a significant variable, given the survey results indicating that U.S. fruit and vegetable firms conducted most of their FDI to export goods to the United States. The impact of this variable was expected to be negative, indicating that higher tariff rates are associated with a lower probability that a firm is engaged in FDI. This is a locational advantage for firms investing in countries that are not charged, or charged a very low tariff, on commodities upon entering the United States.

Annual gross sales from a month in 1994 to the same month in 1995 (SAL) represented another explanatory variable designed to account for firm size. It was expected that higher levels of SAL would be associated with a higher probability that a firm is engaged in FDI. This is an ownership-specific advantage concerning foreign production. Obviously, a large firm can finance a foreign operation easier than a small firm.

Also included as an explanatory variable was the percentage of U.S. sales resulting from imports (FRIMP). FRIMP was designed to account for a firm's reliance upon imports to satisfy domestic demand. Clearly, there is a higher probability for FDI by a firm relying heavily upon imports to satisfy U.S. demand than one that does not. Reliance on imports is a globalization incentive for FDI and was expected to have a positive association.

The percentage of total imports from the source country from which most of the foreign supplies are received (TOTIMP) also was included in the model to help explain the probability of FDI. TOTIMP is an ownership-specific incentive for FDI and was included in the model to capture the incentive for firms concentrating, in one location, commodity sourcing interests to engage in FDI. The probability for a firm to practice FDI was expected to be higher for the firm receiving most of its imports from a single country, thus a positive sign was expected for the coefficient of this variable.

Including a variable (CHEAP) to account for the percentage savings a firm receives from foreign-produced fruits and vegetables, as opposed to domestically obtained product, offers further explanation for the probability of FDI. The coefficient for this locational variable was therefore expected to be positive, indicating a higher probability of FDI for firms with cheaper sourcing globally.

The model also includes a dummy variable indicating whether the firm imported year around or seasonally (SEAS), where 1 indicates seasonally and 0 otherwise. SEAS is a locational incentive for FDI in the U.S. fruit and vegetable industry, given more suitable climates for production in Latin America. As noted in the introduction, the United States heavily imports fruits and vegetables on a seasonal basis. However, given the post-NAFTA direction of this study and a more globally oriented marketplace in the future, this variable was not expected to be positively related to the probability of FDI. Rather, for greater use of investment costs, seasonal shipping, as opposed to year-round shipping, to the United States should be negatively related to the likelihood of FDI.

Results

Probit analysis included estimation of coefficients of explanatory variables, the likelihood ratio index determining the "goodness of fit," and the likelihood ratio test determining the overall significance of the model. Table 9 presents the coefficients, standard errors, t-ratios, p-values, and marginal probabilities for the explanatory variables. All variables except SEAS (seasonal imports) were significant, as expected, at the 0.01 probability level. All variables had the expected sign. Though the negative coefficient for SEAS was not significant at the 0.01 level, it was significant at the 0.15 level.

The "goodness of fit" of the model, as indicated by the likelihood ratio index, is 0.449. The likelihood ratio test determining the overall significance of the variables associated with the probability of FDI yielded a test statistic, distributed as a chi-square, of 12.592. This figure indicates that the model is significant and explains about half of the incentive for FDI.

The coefficients on the explanatory variables are related to an index and thus, not directly related to the dependent variable, INV. The marginal probability concept is useful in determining the relative strength of the variables and predicting the effects of change in each of the explanatory variables on the probability of FDI activities undertaken by a fruit and vegetable firm. Derivatives of a probability function evaluated at the mean values of explanatory variables can be calculated by multiplying the coefficients by the standard normal probability density function of the probit model (Maddala). Because these derivatives embody the marginal concept, the change in probability of a fruit and vegetable firm engaging in FDI can be calculated by multiplying the amount of change in the explanatory variable by the derivative of that variable (Epperson et al.). Table 9 presents these marginal probabilities.

The marginal probability on SAL indicates the probability of a firm being engaged in FDI increases by approximately 0.007 as its gross sales increase by $1 million. The marginal probability for FRIMP indicates that as the composition of a firm's domestic sales resulting from imports increases by 1%, the probability that the firm is engaged in FDI increases by about 0.008. The marginal probability for TOTIMP indicates that as a firm's interest in one particular country increases by 1%, in terms of its total imports, the probability that this firm engages in FDI increases by about 0.010. The marginal probability on LTAR reveals that as tariff levels in the United States increase by 1 cent/kg, the probability that a firm engages in FDI decreases by about 0.076. The marginal probability on CHEAP suggests that for each 1% savings in cost from international production, the probability that a firm engages in FDI increases by about 0.016.

Discussion of Results

The likelihood ratio index presented earlier suggests that the model fits the data fairly well given that the data are cross-sectional. This index is similar to the R² of a model with a continuous dependent variable (Kmenta). The likelihood ratio test provided

further support that the model significantly explains much of the proclivity for FDI in the U.S. fruit and vegetable industry.

The significance of LTAR corroborates many of the results and expectations of the survey. LTAR indicated that firms facing lower U.S. tariff rates have more of an incentive to engage in FDI than firms facing higher U.S. tariffs upon importing their most important commodity. First, U.S. tariff rates are important because most of the firms engaging in FDI reported that they export a majority of their output produced internationally to the United States. Clearly, firms interested primarily in the international marketplace would be less concerned with U.S. import tariffs. Second, a majority of U.S. fruit and vegetable importers obtain their commodities primarily from Mexico. With NAFTA and diminish-ing tariff levels, this effect could have significant implications for the future of the U.S. fruit and vegetable industry.

The significance of SAL (gross annual sales) and FRIMP (percentage of U.S. sales from imports) suggests that large fruit and vegetable firms, where imports constitute a majority of their U.S. domestic sales, have a higher probability of being engaged in FDI. The importance of firm size and the presence of global activity provides direct support for the validity of the model.

TOTIMP (percentage of U.S. imports from the most important source country) and CHEAP (percentage savings from foreign sources for a firm's most important commodity) indicate that firms interested in one particular foreign location where their commodity can be produced cheaper than in the United States have an incentive to engage in FDI. These factors combined with the importance of tariff rates (LTAR) and a more globally oriented marketplace could bring significant changes to the U.S. fruit and vegetable industry.

The negative sign on the coefficient for SEAS perhaps indicates that firms interested in FDI are making their decisions based on factors other than seasonality. It may be that greater utilization of foreign investment in fixed factors of production is considered important. This seems likely in a post-NAFTA environment, as previously hypothesized.

Implications

The variable LTAR captures perhaps one of the most significant implications to be derived from the econometric model. It indicates that U.S. fruit and vegetable firms face changing market conditions given NAFTA and decreased tariffs. Firms importing a commodity from Mexico under tariff may profit from FDI as tariffs diminish. U.S. fruit and vegetable firms can benefit from adapting their present operations to changes in tariff levels. As this barrier between markets diminishes, both inputs and outputs can move across international borders less expensively. This can create new opportunities for U.S. domestic producers who become involved in FDI or expand such activities.

Policy makers should not only consider the effect of tariff levels on imports and exports but also contemplate the effect on FDI. The significance of LTAR implies that firms alter production strategies globally in reaction to trade policies. Consideration of welfare gains from such policies merit the inclusion of this effect when considering international trade legislation.

The variables SAL, FRIMP, and TOTIMP have significant implications concerning the type of U.S. fruit and vegetable firm that can remain competitive in the future. Large U.S. firms receiving large amounts of imports during the year from one source region have an incentive to engage in foreign production in that area. Smaller, strictly domestic firms can be seriously affected by this competitive trend.

NAFTA has engendered many complaints from domestic fruit and vegetable producers, typically small in size and without global interest, according to specific comments extracted from the survey results. These firms are concerned with decreased tariff rates bringing cheaper imports into the domestic market, making it difficult to compete with foreign firms. The results of this analysis suggest that U.S.-owned foreign production facilities may benefit greatly from NAFTA and pose an even greater threat to strictly domestic firms. Concerning firm type and size, these results imply that a global marketplace is of increasing interest and that domestic firms must alter operations if they expect to survive.

Smaller firms that strictly import fruits and vegetables likely cannot enjoy the returns to size and minimization of transactions costs that are characteristic of successful international firms. Further, U.S. fruit and vegetable firms that buy from foreign sources may be at a cost disadvantage relative to U.S. firms that produce internationally for the U.S. domestic market.

Given these results, policy makers face an increasing array of consequences when implementing trade agreements. Decreased trade barriers among countries may affect different firms in different ways. Strictly domestic firms may face increased competition and lower returns because of imports, and international firms may enjoy greater returns because of decreased transaction costs. Policy makers should consider that costs and benefits of decreased trade barriers affect companies differently, depending on firm size and type.

That less expensive international commodity sourcing was a significant decision factor for firms considering FDI implies that U.S. domestic firms are interested, and have been successful, in capturing gains associated with foreign fruit and vegetable production. This has implications for domestic as well as foreign firms and markets.

U.S. domestic firms may experience increasing returns to factors of production if they engage in FDI. Factors such as labor and land were important to firms participating in the survey, and the significance of CHEAP (less expensive foreign sources) in the econometric model supports their relevance. However, an exception was found regarding factors of production: Many firms indicated importing machinery from the United States to the site of FDI. Even with this added burden, FDI appears to be economically worthwhile.

Foreign firms may face increasing production costs as international firms enter their countries and begin to demand factors of production such as land and labor. However, the affected foreign countries may benefit from increased international investment in terms of job creation and capital formation.

In any event, the significance of less expensive foreign sources of produce (CHEAP) implies that U.S. fruit and vegetable firms are interested in capturing gains associated with foreign production. The significance of concentrated sourcing in a particular foreign country (TOTIMP) suggests that U.S. firms interested in FDI centralize their inter-national interests. Thus, the competition for resources is likely to be concentrated in specific locations in specific foreign countries.

The negative sign for the coefficient of the dummy variable indicating seasonal importing to the United States (SEAS) may have important implications for the future of the U.S. fruit and vegetable industry. SEAS indicates that firms are making FDI decisions based on production cost advantages and a more globally oriented marketplace for greater use of investment costs. This implies that rational decision making requires firms to consider shifting production to the most cost-effective and efficient environments whether the commodity is in or out of season in the United States.

Summary

The United States is a major producer of fruits and vegetables, valued at more than $20.3 billion in 1995. The U.S. fruit and vegetable industry has been largely seasonal in nature with respect to production. In the past, U.S. imports of fruits and vegetables have been mainly supplemental. That is, in general, foreign-produced fruits and vegetables have entered the United States when production in the United States has not been in season, or when prices have been sufficiently high to attract additional supplies from foreign sources.

Rapid advances in technology have occurred. Advances in information and communication systems, distribution systems, and large-scale production systems have changed markets dramatically, making them more global and less regional in scope. With such advances, distance between production and consumption locations becomes less and less of a barrier in the competition for markets. Moreover, artificial trade barriers between countries are falling. Important examples of such action are the General Agreement on Tariffs and Trade (GATT) and the North American Free Trade Agreement (NAFTA). Historically, changes in international laws and treaties among countries have had significant effects on trade and investment practices. This research analyzed the effect of free trade agreements, as well as other decreased barriers to trade, on the foreign investment practices of U.S. fruit and vegetable firms. Of primary interest was whether changes in trade barriers have a significant effect on FDI, determining ownership and location of firms and competition within the produce industry.

Firms have many reasons for establishing or changing the geographical area in which they choose to produce, and the potential for greater returns can result from many different factors. The decision to produce internationally is based on one or more of these factors. The most widely accepted incentives for firms to engage in FDI include ownership-specific as well as other firm-level globalization incentives and locational advantages.

This study analyzed the incentives and factors affecting decisions to engage in FDI by U.S. fruit and vegetable firms. The data for the study were obtained through a questionnaire administered to U.S. fruit and vegetable growers/importers. The types of fruit and vegetable firms targeted in the survey were those that produce or obtain the same commodity both domestically and internationally. Firms with this characteristic were labeled produce growers/ importers. These firms were targeted because they were deemed most likely to be involved in foreign investment.

More than 90% of the entire population of approximately 400 U.S. fruit and vegetable growers/importers were contacted. Complete questionnaires, and therefore usable observations, were obtained from 81 firms. The fruit and vegetable firms participating in the survey had average annual gross sales of $26.9 million. Firms involved in FDI reported average sales of $29.6 million, and non-investors reported an average of $22.8 million in annual gross sales.

The survey participants indicated that an average of about half of their domestic sales resulted from imports. This average for foreign investors was almost twice that of firms that strictly imported. Mexico was the leading source for imports, followed by Chile and Guatemala. More than half of all survey participants received most of their imports from Mexico. Concerning international sales, more than three-fourths of the participants' total sales in countries other than the United States resulted from U.S. supplies exported to the relevant country.

More than half of the firms surveyed reported involvement in foreign production in 1994-1995. Mexico was the leading destination for FDI among the survey participants, followed by Guatemala and Chile. An average of more than half the total supplies from firms engaged in FDI resulted from their investment in 1994-1995. Almost all the firms surveyed cited cheaper labor and, to an extent, seasonality as the most important factors influencing their historical FDI practices. Survey participants indicated that most supplies they received from their primary source country resulted from their investment there. Participants purchased very few supplies from local producers. Most of the supplies produced internationally because of FDI were exported to the United States, while very few were sold locally and/or exported to another country.

Survey participants said imports were significantly cheaper on average than produce obtained domestically in the United States. The foreign investors indicated that U.S. imports were much less expensive than that indicated by firms strictly im-porting fruits and vegetables in 1994-1995. Several firms indicated provision for start-up funds to determine whether certain fruits and vegetables could be grown profitably internationally. Mexico was the most popular place for this practice. Almost half the firms indicated they imported fruits and vegetables year around and a little over half, seasonally.

Survey participants reported tariffs as the most important trade barrier followed by phytosanitary regulations and other rules and regulations set by the U.S. Department of Agriculture and the Environmental Protection Agency. About half the firms surveyed indicated expectations that NAFTA would directly affect their operation. The responses (from foreign investors and non-investors) concerning NAFTA were almost identical. Of the firms surveyed that were not involved in foreign production, more than a third indicated they would consider FDI because of NAFTA or a similar trade agreement that would involve their country of interest.

The econometric analysis provided estimates of the power of explanatory variables from the survey in explaining incentives for foreign production. The probit model estimated the probability that a firm is engaged in FDI. Analysis determined that larger fruit and vegetable firms receiving a large amount of imports, interested in one particular source region internationally, and facing a relatively small U.S. import tariff had a higher probability of being involved in FDI.

The relationship between U.S. import tariffs on fruits and vegetables and FDI in Latin America by U.S. fruit and vegetable firms provides perhaps the most significant conclusion from this analysis. U.S. tariff rates are important because most of the firms engaging in FDI reported they export a majority of their output produced internationally to the United States. A majority of U.S. fruit and vegetable importers obtain their commodities primarily from Mexico. With NAFTA and diminishing tariff levels, this effect could have significant implications for the future of the U.S. fruit and vegetable industry.

Analysis shows that less expensive international commodity sourcing was a significant decision factor for firms considering FDI. This implies that U.S. domestic firms are interested, and have been success-ful, in capturing gains associated with foreign fruit and vegetable production.

The negative relationship between seasonal importing and the probability of FDI may have important implications for the future of the U.S. fruit and vegetable industry. This relationship indicated that firms are making their FDI decisions based on production cost advantages and a more globally oriented marketplace for greater use of investment costs. Rational decision making requires firms to consider shifting production to the most cost-effective and efficient environments whether the commodity is in or out of season in the United States.

The findings presented in this analysis are significant given plausible trends in the U.S. fruit and vegetable industry. First, large firms benefitting from economies of size internationally should have the best chance of surviving in the future. Second, given that firms involved in FDI are typically interested in particular source countries outside of the United States and given that they reported receiving product cheaper on average from these areas implies that U.S. firms are already allocating more resources globally as it proves profitable.

Further, because U.S. fruit and vegetable firms are investing internationally in countries that are charged a low or zero U.S. tariff rate, NAFTA can be expected to increase U.S. FDI in Mexican fruit and vegetable production for the purpose of exporting to the United States. The behavior of the firms participating in the analysis is consistent with expected post-NAFTA behavior.

This analysis supports the notion that as barriers to entry and exit between markets diminish, factors of production will flow between these markets until marginal productivities of these factors equilibrate. Furthermore, it reveals the types of firms that are expected to be most successful as barriers to trade diminish. Firms engaged in FDI seem to have different ideas than those that are not, and they appear to visualize the U.S. fruit and vegetable market from a more global perspective.

References

American Farm Bureau. NAFTA, Effects on Agriculture, Volume 4, An American Farm Bureau Research Foundation Project. Washington, DC: American Farm Bureau Research Foundation, 1991.

American Farm Bureau. GATT and NAFTA: Major Provisions, Promises and Threats. Volume 1, Final Report. Washington, DC: American Farm Bureau Federation, November 1995.

Bolling, C. and C. Valdes. The U.S. Presence in Mexico's Agribusiness. U.S. Department of Agriculture, ERS. Foreign Agricultural Economic Report Number 253. Washington, DC: U.S. Government Printing Office: July 1994.

Cook, R. L. "From Competition to Coordination in Vegetable Trade: The Case of Mexico and California." Vegetable Markets in the Western Hemisphere. eds. R. A. Lopez and L. C. Polopolus, pp. 129-145. Ames, Iowa: Iowa State University Press, 1992.

Epperson, J.E., M.H. So, S.M. Fletcher and D.H. Carley. "Profile Analysis of Market Participants' Attitudes Toward Selected Market Alternatives for US Shelled Peanuts." Agribusiness. 4(1988): 433-452.

Grubert, H. and J. Mutti. "Taxes, Tariffs, and Transfer Pricing in Multinational Corporate Decision Making." The Review of Economics and Statistics 1991, pp. 285-293. Harvard University, 1991.

Handy, C. R. U.S. Foreign Investment and Trade Prospects in Mexico's Agribusiness Sector. Organization and Performance of World Food Systems: NC-194, OP-51, October 1993.

Hillman, J. S. "Nontariff Barriers: Major Problem in Agricultural Trade." American Journal of Agricultural Economics. 60(1978): 491-501.

How, R. B. Marketing Fresh Fruits and Vegetables. New York: Van Nostrand Reinhold, 1991.

Husted, S. and M. Melvin. International Economics. New York: Harper & Row, Publishers, Inc., 1990.

Jun, J. "U.S. Tax Policy and Direct Investment Abroad." Taxation in the Global Economy. The University of Chicago Press, 1990.

Kmenta, J. Elements of Econometrics. New York: Macmillan Publishing Company, 1986.

Maddala, G. S. Limited-Dependent and Qualitative Variables in Econometrics. New York: Cambridge University Press, 1983.

Produce Reporter Company. Blue Book, Spring 1994. Fruit and Vegetable Credit and Marketing Service. Carol Stream, Illinois: Produce Reporter Company, 1994.

Rugman, A. M. Foreign Investment and NAFTA. Columbia, South Carolina: University of South Carolina, 1994.

Stone, F. GATT: From the Tokyo Round to the Uruguay Round. Ottowa: International Economics Program, 1988.

U.S. Department of Agriculture. Agricultural Marketing Service. Fresh Fruit and Vegetable Shipments. Fruit and Vegetable Division, Washington, DC, 1983-1986.

U.S. Department of Agriculture. Economic Research Service. Foreign Agricultural Trade of the United States. An Economic Research Service Re-port. Washington, DC: U.S. Government Print-ing Office, September/October 1986 & 1995.

U.S. Department of Agriculture. Economic Research Service. NAFTA Economic Monitoring Task Force. NAFTA: An Early Assessment. Washing-ton, DC: U.S. Government Printing Office: December 1994.

U.S. Department of Agriculture. Economic Research Service. Situation and Outlook Report, Vegetables and Specialties, Report No. VGS-265 April 1995 and Report No. VGS-267, November 1995. Washington, DC: U.S. Government Printing Office, 1995.

U.S. Department of Commerce, Bureau of Eco-nomic Analysis. Survey of Current Business, Vol. 70, No. 8. Washington, DC: U.S. Government Printing Office, 1993.

U.S. International Trade Commission. Annual Report on the Impact of the Caribbean Basin Economic Recovery Act on U.S. Industries and Consumers. Ninth Report, Investigation No. 332-227. ed. H. Troupos. Washington, DC: U.S. Government Printing Office, March 1995.

The University of Georgia is an equal opportunity/affirmative action institution. Information contained in Georgia Agricultural Experiment Station publications is available to everyone, without regard to race, color, national origin, sex, age, or handicap.