Source: OHIO STATE UNIVERSITY submitted to NRP
USING THE GRAVITY MODEL TO EXPLAIN TRADE IN FOOD AND AGRICULTURAL PRODUCTS
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
0198383
Grant No.
(N/A)
Cumulative Award Amt.
(N/A)
Proposal No.
(N/A)
Multistate No.
(N/A)
Project Start Date
Oct 1, 2003
Project End Date
Sep 30, 2009
Grant Year
(N/A)
Program Code
[(N/A)]- (N/A)
Recipient Organization
OHIO STATE UNIVERSITY
1680 MADISON AVENUE
WOOSTER,OH 44691
Performing Department
AGRICULTURE, ENVIRONMENTAL AND DEVELOPMENT ECONOMICS
Non Technical Summary
Agricultural trade has often been analyzed with the neo-classical model, which may be less useful in analyzing processed food trade which has accounted for nearly 70 percent of world trade in food and agriculture since the early-1990s. The project will use the gravity model to analyze food and agricultural trade in order to establish which theories best explain such trade flows.
Animal Health Component
50%
Research Effort Categories
Basic
50%
Applied
50%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
60661203010100%
Goals / Objectives
This project will use the gravity model to analyze bilateral trade in food and agricultural products over the period 1961 to 1999 using panel econometric methods. The results of the proposed study will establish which trade theories best explain bilateral trade for a data set comprised of bilateral food and agricultural trade flows for a sample of 10 developed countries including the United States. Trade in the food and agricultural sector has typically been examined using the neo-classical model of trade, based on the assumptions of competition, constant returns to scale, and homogeneous goods. While this is likely an appropriate model for understanding trade in agricultural commodities, it may be less useful in analyzing trade in processed food products which have accounted for nearly 70 percent of world trade in food and agriculture since the early-1990s. There is now considerable evidence that the food-processing sector has a concentrated market structure, with firms selling highly differentiated food products. Consequently, if more recent theories of trade are based on the premise that firms sell differentiated products under increasing returns, it is appropriate to test the validity of these theories in the context of food and agricultural trade. The so-called gravity equation of trade predicts that the volume of trade between two countries will be proportional to their GDPs and inversely related to any trade barriers between them. Modern versions of the gravity model, however, allow for much more explicit tests of a wider variety of trade models. In analyzing bilateral trade data, there appears to be a model identification problem: the gravity equation works well empirically for both differentiated and homogeneous goods. It has been argued, however, that the gravity model can nest both the increasing returns/product differentiation story as well as a more conventional homogeneous goods/relative factor abundance story. With appropriate theoretical restrictions on the income parameter, it is possible to use the gravity model of bilateral trade to discriminate between different theories of international trade. In this context, there are two key objectives of this project. The first is to divide the different industries in the food and agricultural sector into different groups to reflect differentiated versus homogeneous goods, and also country pairings to account for differences in relative factor endowments. By separating bilateral trade flows into differentiated and homogeneous goods, we can explicitly identify whether the increasing returns/product differentiation story can explain processed food and agricultural trade, and whether the homogeneous goods/relative factor abundance story can explain trade in agricultural commodities. The second objective is to use bilateral trade data over the period 1961-1999 for 10 countries, and panel econometric methods to estimate four versions of the gravity model.
Project Methods
Even though it is unlikely that any observed trade flows are solely determined by any one of version of the gravity model, it is expected that different trade models will account for a large part of such flows if the data set is carefully divided up on the basis of well-defined criteria. We will first divide our sample up into groups of differentiated good and homogeneous goods. This can be done by applying the standard Grubel and Lloyd index to measure the share of intra-industry trade in total trade using 3-digit SITC categories. Given that intra-industry is typically associated with increasing returns/differentiated products, this index is used to identify a sample where the increasing returns gravity equation is most likely to apply. For each 3-digit SITC category in food and agriculture, we can calculate the average value of the index for each country pairing over the time-period 1961-1999. These values will be used to split the bilateral trade flow data, into high and low intra-industry trade samples. The high intra-industry trade sample will be used to test for perfect specialization based on increasing returns, while the low intra-industry sample will be used to test for perfect specialization based on relative factor endowments. Second, in a sample of country pairs where there is little or no specialization due to increasing returns, but the absolute difference between the two countries' factor proportions is large, it would be expected that the standard trade model applies. Consequently we also divide the sample of countries up into those that have low (large) absolute differences in their capital-labor ratios. We will test four versions of the gravity model using a fixed effects model with country fixed effects, and a time fixed effect. Version 1 will be the gravity equation for perfect product specialization based on increasing returns to scale, the sample being based on bilateral trade flows characterized by high values of the Grubel and Lloyd index. Version 2 will be the gravity equation for the multi-cone Heckscher-Ohlin (H-O) model, the sample being bilateral trade flows characterized by low values of the Grubel and Lloyd index, where the sample is also divided by country pairings based on low (large) absolute differences in their capital-labor ratios. Version 3 will be the gravity equation for the increasing returns/unicone H-O model, the sample being bilateral trade flows characterized by high values of the Grubel and Lloyd index, and taking account of whether one country in the pairing is relatively capital abundant compared to the other country. Finally, version 4 will be the gravity equation for the uni-cone H-O model, based on the sample of bilateral trade flows characterized by low values of the Grubel and Lloyd index, and taking account of whether one country in the pairing is relatively capital abundant compared to the other country.

Progress 10/01/03 to 09/30/09

Outputs
OUTPUTS: OUTPUTS: In the lifetime of this project, three key activities have contributed to its goals and objectives. First, we have conducted a very thorough review of the extant theoretical and empirical literature pertaining to the use of the gravity model in explaining international trade in the food and agricultural sector, with particular reference to the impact of exchange rate uncertainty. Second, compared to previous research in this field, rather than focusing on the impact of short-run exchange rate volatility on aggregate trade in the agricultural and food sector, we have focused specifically on measuring the impact of medium to long-run exchange rate uncertainty on trade in specific agricultural products. This is critical, as there are instruments available for hedging against short-run volatility, but not long-run exchange rate uncertainty. Third, in order to conduct our econometric analysis, we have participated in a very careful collection of two large panel data sets for two trade in two different agricultural commodities, and two different panels of countries. These data-sets are available for both updating and further analysis by other researchers in this field. In collaboration with a graduate student at Ohio State University, and an economist at the Economic Research Service, USDA, we have completed our analysis of these data-sets, and the results have been written up as paper, which once revised, will be prepared for submission to a peer-reviewed applied economics journal. DISSEMINATION: The results of this research have been widely disseminated at USDA through our collaboration with researchers at the Economic Research Service, who were also invaluable in helping to build the panel data set that underlies the research. PARTICIPANTS: PARTICIPANTS: At Ohio State, Ian Sheldon (PI) worked on the project in collaboration with Stan Thompson, and Shruti Khadka Mishra(graduate student), and also worked with Daniel Pick from USDA/ERS. The graduate student was able to base her MA Thesis on part of the research conducted through this project. TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
In order to analyze the effect of exchange rate uncertainty, we developed and applied an empirical gravity equation to two sets of US bilateral trade data: fresh fruit over the period 1976-99 for a panel of 26 countries; and fresh vegetables over the period 1976-2006 for a panel of 9 countries. These products were chosen for analysis because of their contribution to US farm income receipts, their increasing importance in the diets of both US and other countries' consumers, and also their growing share of international trade in the agricultural sector. In addition, in its report, "Global Trade Patterns in Fruits and Vegetables", the Economic Research Service of USDA (2006) had raised the important issue that trade in these products might be significantly affected by exchange rate uncertainty. Based on fixed and random effects panel estimation methods, and using the Peree and Steinheer (1989) measure of exchange rate uncertainty, the results show that US bilateral fresh fruit trade has been negatively affected by exchange rate uncertainty. We also found that the exchange rate between the US dollar and the currencies of Latin American trading partners account for most of the impact of exchange rate uncertainty on bilateral trade flows in fresh fruit. In contrast, while gravity-type variables such as GDP and distance have the expected significant impact on trade, we have found no statistically significant evidence for any negative effect of exchange rate uncertainty on US bilateral fresh vegetable trade. Unlike fresh fruit trade, US bilateral trade in vegetables, and notwithstanding the growing importance of Chinese vegetable exports, is clearly dominated by trade with its NAFTA partners, Canada and Mexico.

Publications

  • Sheldon,I., Shruti Khadka M., Pick, D. and Thompson, S. (2009). "Exchange Rate Uncertainty and US Bilateral Fresh Fruit and Fresh Vegetable Trade: An Application of the Gravity Model." Working Paper.


Progress 01/01/08 to 12/31/08

Outputs
OUTPUTS: Progress has been made on applying a gravity model to US bilateral trade in fresh vegetables to establish the effects of exchange rate uncertainty, using a ten-country panel data set for the period 1976-2006. Initial application of the model was disappointing, with no evidence that exchange rate uncertainty has had an impact on US bilateral trade in fresh vegetables. However, closer inspection of the results indicated a problem with heteroskedasticity and autocorrelation involving US bilateral trade with China. After adjusting for these problems, econometric results were derived based on Ordinary Least Squares (OLS) and fixed effects panel estimation methods. The OLS results indicate that the gravity model works pretty well, with GDP per capita and membership in a free trade agreement, having a significant positive impact on US bilateral trade in fresh vegetables, while distance has a somewhat significant negative impact on bilateral trade. There is also evidence that exchange rate uncertainty has a statistically negative impact on bilateral trade over periods of 3 and 5 years out. In the one way fixed effects model, GDP per capita and membership of a free trade agreement again have a statistically significant positive impact on US bilateral trade in fresh vegetables, and exchange rate uncertainty has a statistically negative effect over periods of 2 and 3 years out. Distance is no longer included in the model as it is perfectly correlated with the country fixed effects. Also, introducing a time fixed effect results in bilateral trade being negatively correlated with GDP per capita, so this was dropped, GDP per capita capturing any time effects in the model. At present the focus is on estimating the model with random effects panel methods. The results for the gravity model applied to US bilateral trade in fresh vegetables are very similar to those derived previously for US bilateral trade in fresh fruit. The key difference seems to be the strong free trade agreement effect for fresh vegetable trade, which is not surprising given the dominance of Mexico and Canada in the data, whereas there is a strong regional effect for fresh fruit trade. PARTICIPANTS: Work on project conducted in collaboration with a graduate research assistant at Ohio State University, Misashruti Khadka, and Daniel Pick at the Economic Research Service, USDA. TARGET AUDIENCES: The target audience for this research includes other academics working on agricultural trade issues, as well as economists working in USDA. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
Gravity models have rarely been applied to specific commodities such as fresh fruit and vegetables therefore this research makes a contribution to empirical discussion in discipline. While short-term exchange rate risk can be hedged in financial markets, uncertainty beyond a one-year time horizon cannot be hedged at low cost. It is important, therefore, as global integration of markets increases, to establish whether it is short-run or long-run movements in exchange rates that matter for international trade in agricultural sector.

Publications

  • No publications reported this period


Progress 01/01/07 to 12/31/07

Outputs
OUTPUTS: An empirical gravity equation has been applied to US bilateral trade data for fresh fruit over the period 1976-99 for a panel of 26 countries in order to analyze the effect of exchange rate uncertainty. The results of this research have been shared and discussed with other researchers at the Economic Research Service of USDA through publication of a working paper. PARTICIPANTS: Ian Sheldon, faculty at Ohio State, supervised the research, and wrote up the analysis and results. Shruti Khadka Mishra, research assistant at Ohio State helped complete empirical analysis. Daniel Pick, Branch Chief at Economic Research Service, USDA provided data and feedback on the model and empirical analysis. Stanley Thompson, faculty at Ohio State, provided input into discussion of empirical analysis. TARGET AUDIENCES: Key target audiences are other academic researchers, as well as informing researchers working in USDA of the impact of exchange rate uncertainty on agricultural trade. PROJECT MODIFICATIONS: Refined application of panel methods in applying the gravity equation.

Impacts
Based on fixed and random effects panel estimation methods, and using the Peree and Steinheer (1989) measure of exchange rate uncertainty, results show that US bilateral fresh fruit trade has been negatively affected by exchange rate uncertainty. It was also found that the exchange rate between the US dollar and the currencies of Latin American trading partners accounts for most of the impact of exchange rate uncertainty on bilateral trade flows in fresh fruit. This research is now being extended to examining US bilateral trade in fresh vegetables, funded by a new cooperative agreement with the Economic Research Service of USDA. While short-term exchange rate risk can be hedged in financial markets, uncertainty beyond a one-year time horizon cannot be hedged at low cost. This results of this research indicate that as global integration of markets increases, it is important to establish whether it is short-run or long-run movements in exchange rates that matter for international trade in agricultural sector.

Publications

  • Sheldon, I., Khadka Mishra, S., Pick, D. and Thompson, S. (2007). Exchange Rate Uncertainty and US Bilateral Fresh Fruit Trade: An Application of the Gravity Model. Working Paper, Ohio State University.(Pending)


Progress 01/01/06 to 12/31/06

Outputs
Conducted further refinements of analysis of impact of medium to long-term exchange rate uncertainty on US fruit trade using an empirical gravity model and panel econometric methods. Focusing on US fruit trade with its twenty-six top fruit trade partners over the period 1976-199, the analysis shows that fruit trade is negatively affected by exchange rate uncertainty as measured by a moving average index originally developed by Peree and Steinherr (European Economic Review, 1989). Initial results based on a pooled ordinary least squares (OLS) regression, indicate that the elasticity of US bilateral fruit trade is highest with respect to exchange rate uncertainty over a two-year time horizon, but the elasticity decreases with an increase in the moving average period to three and five years. Further analysis using fixed and random effects panel methods find similar results, implying that risk-averse traders are affected by exchange rate uncertainty. The analysis also segregated the effects of exchange rate uncertainty on US fruit trade with developed versus developing countries, but found no difference between these two groups of countries. Currently, analysis being extended to examine whether there are regional differences in the effect of exchange rate uncertainty. The basis for testing the monopolistic competition model of trade has also been further developed drawing on recent theoretical developments in the international economics literature. Specifically, it has been shown that the monopolistic competition model can be nested within a broad framework based on the gravity model, and that with specific parameter restrictions, tests of this and other theories of trade can be conducted using the appropriate agricultural trade data sets. Currently, a panel data set is being developed to test this model.

Impacts
In studying the effects of exchange rate uncertainty on trade, it is important to distinguish between short- and medium/long-term changes in exchange rates. While short-term exchange rate risk can be hedged in financial markets, uncertainty beyond a one-year time horizon cannot be hedged at low cost. It is important, therefore, as global integration of markets increases, to establish whether it is short-run or long-run movements in exchange rates that matter for international trade in the agricultural sector. This is particularly relevant in the fresh fruit sector which does not receive US government subsidies in the form of price and income supports.

Publications

  • Sheldon, I.M. (2006). Monopolistic competition and trade: does the theory carry any empirical weight? Journal of International Agricultural Trade and Development, 2(1): 1-31.


Progress 01/01/05 to 12/31/05

Outputs
Analysis has been conducted examining the impact of medium to long-term exchange rate uncertainty on US fruit trade using the gravity model and panel econometric methods. Focusing on US fruit trade with its twenty-six top fruit trade partners over the period 1976-199, the analysis shows that fruit trade is negatively affected by exchange rate uncertainty as measured by a moving standard deviation measure and an index originally developed by Peree and Steinherr. The results indicate that the elasticity of US bilateral fruit trade is highest with respect to exchange rate uncertainty over a two-year time horizon, but the elasticity decreases with an increase in the moving average period to three and five years, and becomes statistically insignificant in the case of a ten year moving average. These results imply that risk-avers traders are affected by exchange rate uncertainty. The analysis also segregated the effects of exchange rate uncertainty on US fruit trade with developed versus developing countries. The results indicate that exchange rate uncertainty is an important factor affecting US fruit trade with developing countries, but does not impact trade with developed countries. The basis for testing the monopolistic competition model of trade has also been developed drawing on recent theoretical developments in the international economics literature. Specifically, it has been shown that the monopolistic competition model can be nested within a broad framework based on the gravity model, and that with specific parameter restrictions, tests of this and other theories of trade can be conducted using the appropriate agricultural trade data sets. Currently, a panel data set is being developed to test this model.

Impacts
In studying the effects of exchange rate uncertainty on trade, it is important to distinguish between short- and medium/long-term changes in exchange rates. While short-term exchange rate risk can be hedged in financial markets, uncertainty beyond a one-year time horizon cannot be hedged at low cost. It is important, therefore, as global integration of markets increases, to establish whether it is short-run or long-run movements in exchange rates that matter for international trade in the agricultural sector.

Publications

  • Mishra, S.K. 2005. Exchange rate uncertainty and US fruit trade: application of the gravity model. Unpublished Master's Thesis, The Ohio State University.
  • Sheldon, I.M. 2005. Monopolistic competition and trade: does the theory carry any empirical weight?, in New Dimensions in Modeling Food and Agricultural Markets, International Agricultural Trade Research Consortium.


Progress 01/01/04 to 12/31/04

Outputs
Currently examining impact of exchange rate uncertainty on trade in fruit and vegetables, part of the project developed and funded as a Cooperative Agreement with the Economic Research Service of USDA. This project is examining the effect of both short-run and medium to long-run exchange rate uncertainty on US trade in fruits and vegetables. This is being done by constructing a gravity model and applying it to a panel of US bilateral trade flows in fruits and vegetables over the period 1978-2000. Currently the panel data set is being built, focusing on collection of data for the US's 20 largest bilateral trading partners in fruit and vegetables, and also data on nominal, real and effective exchange rates for these pairings. In addition, data on country characteristics is being collected from Andrew Rose's web-site at the Haas Business School, UC-Berkeley.

Impacts
In studying the effects of exchange rate uncertainty on trade, it is important to distinguish between short- and medium/long-term changes in exchange rates. While short-term exchange rate risk can be hedged in financial markets, uncertainty beyond a one-year time horizon cannot be hedged at low cost. It is important, therefore, as global integration of markets increases, to establish whether it is short-run or long-run movements in exchange rates that matter for international trade in the agricultural sector.

Publications

  • No publications reported this period


Progress 01/01/03 to 12/31/03

Outputs
Using recent theoretical analysis of the gravity model, and panel econometric methods, developed a model to test which trade theories best explain bilateral trade for a data set comprised of bilateral food and agricultural trade flows for a sample of 10 developed countries including the United States. This draws on the fact that the new theoretical literature on the gravity model shows how restrictions can be placed on the country income elasticity parameters, allowing one to discriminate between different trade theories. Proposal to examine impact of exchange rate uncertainty on trade in fruit and vegetables developed and subsequently funded as a Cooperative Agreement with the Economic Research Service of USDA. This project will examine the effect of both short-run and medium to long-run exchange rate uncertainty on US trade in fruits and vegetables. This will be done by constructing a gravity model and applying it to a panel of US bilateral trade flows in fruits and vegetables over the period 1978-2000.

Impacts
In studying the effects of exchange rate uncertainty on trade, it is important to distinguish between short- and medium/long-term changes in exchange rates. While short-term exchange rate risk can be hedged in financial markets, uncertainty beyond a one-year time horizon cannot be hedged at low cost. It is important, therefore, as global integration of markets increases, to establish whether it is short-run or long-run movements in exchange rates that matter for international trade in the agricultural sector.

Publications

  • No publications reported this period