This paper analyzes the European Union (EU) import demand for in-shell peanuts from three sources: the United States, China, and the rest of the world. We find that peanuts from different sources are differentiated by EU consumers. The expenditure elasticity is elastic for U.S. in-shell peanuts, which is associated with their higher quality. The conditional own price elasticities are more elastic for U.S. and Chinese in-shell peanuts. These findings have at least two implications. First, U.S. producers and exporters should direct efforts to ensure that in-shell peanuts exported to the EU are of the best possible quality, and, second, promotion efforts should stress the quality of U.S peanuts as an advertising tool.
Key Words: European Union import demand, in-shell peanuts, nonlinear SAIDS
JEL Classifications: D12, Q11, Q17
The European Union (EU1) is the largest importer of peanuts in the world. In 2005, its peanut imports accounted for around 40% of the world imports of this commodity (FAO-STAT). However, little economic research has been done on the EU markets for peanuts. To the best of our knowledge, this is the first paper analyzing the EU import demand for peanuts.
A few studies have been conducted on the topic of the demand for U.S. peanuts in foreign markets. For example, Zhang, Fletcher, and Carley estimated Japan's import demand for peanuts to determine the impact of an increase in the Japanese raw peanut quota on peanut imports from the United States and its competitors. In the past, economic studies of peanut markets have focused on the analysis of government intervention in the U.S. domestic and/or international peanut market (e.g., Borges; Borges and Thurman; Miller and Mabbs-Zero). These topics have become less important due to the fact that the free-trade agreements signed by the United States have forced a cut in trade, distorting domestic support to the agricultural sector. Furthermore, the Untied States changed the peanut program in 2002 from a domestic supply management with government intervention to a competition market program. Therefore, a new focus of economic analysis on the U.S. competitive stand in the global peanut market is warranted.
The EU imports peanuts mainly from China, the United States, Latin America, and Africa. EU countries import two types of peanuts: 1) in-shell peanuts and 2) shelled peanuts. Both types are completely different because in-shell peanuts are consumed directly by consumers but shelled peanuts are imported by processors to produce peanut butter, candy, snacks, etc.
The principal objective of this study is to analyze the EU import demand for in-shell peanuts, with a special emphasis on the EU demand for U.S. in-shell peanuts. Specific objectives include the estimation of price and expenditure elasticities and testing product aggregation to see whether U.S. in-shell peanuts are differentiated from other countries. The results are used to analyze the competitiveness of U.S. in-shell peanuts in the EU market and suggest some implications of our findings for marketing this type of U.S. peanut in the EU.
The study of EU markets for in-shell peanuts is useful for U.S. in-shell peanut exporters because in the last decade, the U.S. share of the total import quantity of in-shell peanuts in the EU has declined, whereas China, Africa, and Latin America have increased their export shares. The results from this study can be utilized to better understand the changes in the demand for in-shell peanuts in the world market and to provide useful information for policy makers in government and peanut industry organizations assisting the U.S. peanut producers and processors to compete in a highly competitive market environment. For example, policy makers and researchers could use these results to analyze and quantify the potential impact of federal promotion programs on the EU demand for U.S. in-shell peanuts.
Background Information on World Peanut Trade
According to the Production, Supply, and Distribution (PSD online) database of the Foreign Agricultural Service of the U.S. Department of Agriculture (USDA), the main suppliers of peanuts to the world peanut market are Argentina, China, and the United States, which account for 70% of total export quantity. In 2005, the total quantity of world peanut exports was 2,005 million kg. Out of this total amount, Argentina exported 400 million kg, China exported 784 million kg, and the United States exported 223 million kg. Out of the total U.S. peanut exports, the U.S. export quantities of in-shell peanuts make up only 7.35%, but it is very important for some U.S. states, such as Virginia, North Carolina, South Carolina, Texas, and New Mexico, which are the main source of in-shell production.
The main importers of peanuts in the world are Canada, the EU, Japan, and Mexico. These four importers account for more than 60% of total imports. In 2005, Canada, Japan, and Mexico each accounted for about 7% of the total world import quantities, and the EU accounted for around 38-39%. Hence, the EU is the largest importer of peanuts by quantity and value. The total quantity of peanuts imported by the EU was 778 million kg. Out of this total amount, the EU total quantity import of in-shell peanuts was around 109.57 million kg or 14% of their total import quantities of peanuts.
China, the United States, Latin America, and Africa are the major exporters of in-shell peanuts to the EU. They accounted for more than 96% of the total value and quantity of EU in-shell peanut imports between 1991 and 2005 (EUROSTAT). Whereas the demand for in-shell peanuts in Europe has been steady, the competition among exporters has changed. China is now the dominant exporter of in-shell peanuts. Both African and Latin American countries have increased their export share, while the U.S. export share for in-shell peanuts has decreased over time.
In the early 1990s, the United States and China were the main exporters of in-shell peanuts to the EU. For example, in 1991, the United States and China exported to the EU 35.62 and 40.90 million kg of in-shell peanuts, respectively. The U.S. exports of in-shell peanuts peaked in 1995, when it reached a level of exports of 70.53 million kg, followed by China (29.75 million kg), Latin America (2.54 million kg), and Africa (2.52 million kg). In 1996, China became the dominant exporter of in-shell peanuts to the EU markets, exporting 36.51 million kg, followed by the United States (21.77 million kg), Africa (2.73 million kg), and Latin America (1.96 million kg). The U.S. exports of inshell peanuts have been declining over time. In 2005, the United States only exported around 10.90 million kg of in-shell peanuts to the EU.
Source-Differentiated AIDS Model
A Source-Differentiated AIDS (Almost Ideal Demand System) Model (SAIDS) was first specified by Yang and Koo. The SAIDS model allows for source differentiation, and it closely follows the derivation of the AIDS model proposed by Deaton and Muellbauer. This model has been previously used for import demand for meat (Henneberry and Hwang; Yang and Koo), import demand for fresh fruits (Andayani and Tilley; Schmitz and Seale), and import demand for wine (Carew, Florkowski, and He). The SAIDS model also allows the user to disaggregate and differentiate products by origin.
The SAIDS model is derived from a price-independent generalized logarithmic expenditure function that accounts for the importer's behavior and differentiates goods from different origins. Even though tariffs and quotas on imported peanuts and peanut products have become negligible in the EU, there are several reasons why EU consumers may still differentiate peanuts from different origins. One of these reasons is food safety concerns, especially aflatoxin level. Since peanuts from different sources have different quality attributes (Bliss), it is important to recognize quality differences among peanut exporters when analyzing the EU peanut import demand.
Previous studies using the SAIDS have used the Stone price index as a proxy for the price index derived analytically from the AIDS cost function; however, several studies on consumer demand have shown that the Linear Almost Ideal Demand System could produce biased estimates of demand elasticities (Buse; Chen; Moschini). Therefore, the SAIDS model should be estimated as a nonlinear model (NLSAIDS) instead of using its linear approximation, because policy evaluations and simulations require reliable estimates of demand responsiveness to prices and expenditure.
These restrictions imply that the price and expenditure coefficients from the different sources are equal.
Data and Procedures
The data used to estimate the model are quarterly data from 1991 to 2005, representing a total of 60 observations. The sources of origin of the EU imports of in-shell peanuts considered in this study are China, the United States, and rest of the world (ROW)2. The data were obtained from the EUROSTAT database. The quantity of imports from each source is measured in 100,000 kg, and the value of imports is measured in 1,000 Euros. Since import price data are not available, unit prices3 are used as import prices.
The EU countries grow a trivial amount of peanuts because the climate condition in Europe is not suitable for growing peanuts. Their peanut production is infinitesimal relative to the amount of their peanut imports4; therefore, domestic production can be ignored.
Diagnostic Tests
Model misspecification may lead to biased and inconsistent estimators and/or inappropriate statistical inferences. Therefore, it is important to perform diagnostic tests on the models. In the context of demand systems of equations, practitioners usually perform misspecification tests separately for each of the equations in the system and then combine the results in an ad hoc manner. A more appropriate approach is to conduct the misspecification tests for the system as a whole (McGuirk et al.; Shukur).
Following Shukur, the Breusch-Godfrey (BG) systemwise test was used to test for autocorrelation, the Breusch-Pagan (BP) test was used to test for heteroskedasticity, and the systemwise RESET test was utilized for functional misspecification. All the tests were performed using a multivariate F-test (Shukur, p. 710).
To test for the presence of expenditure endogeneity, we utilized the approach suggested by Blundell and Robin (p. 76). In this approach, each share equation is augmented with the error v from a reduced form for the log expenditure (ln x^sub g^). The error u^sub i^ in Equation (1) is written as the orthogonal decomposition u^sub 1^ = ω^sub i^v + ξ^sub i^, where E(ξ^sub i^ | ln x^sub g^, ln p^sub 1^, . . . , ln p^sub g^) = 0. Similar to Blundell and Robin, we define the reduced form equation for log expenditure (ln x^sub g^) as a function of a linear trend, seasonal dummies, and prices of in-shell peanuts. The hypothesis that the parameters ω^sub i^ are significantly different from zero can be used to test the exogeneity of log expenditures.
Standard Errors in the Nonlinear AIDS Model
Estimation of the standard errors for the elasticities in the nonlinear AIDS (and the nonlinear SAIDS) is complicated by the fact that these elasticities are nonlinear functions of several parameter estimates. For example, the cross price elasticity ε^sub ij^ (Equation [3)] is a function of γ^sub ij^, β^sub i^ and also of all the parameters that enter into the price index p^sub g^. The bootstrap method was utilized to calculate the standard error of elasticities in this study, specifically, a modified version of the moving blocks bootstrap (MBB) method for time series (Goncalves and White). Previous studies using the bootstrap method to estimate standard errors of elasticities (Green, Hahn, and Rocke; Li and Maddala) have assumed specific data-generating processes. Green, Hahn and Rocke assumed a first-order autocorrelation process, whereas Li and Maddala assumed an autoregressive distributed lag model. In contrast, the MBB does not assume any specific dynamic data-generating process (Hardie, Horowitz, and Kreiss).
Empirical Estimation
The estimated system of equations is conditional on the EU total expenditure on imported in-shell peanuts. To make the estimation manageable, we assume that the EU consumers allocate their total expenditure among groups of goods, in-shell peanuts being one of them. We also assume that preferences among these groups are weakly separable. For the allocation of expenditure for the in-shell peanut group, the EU consumers select imported in-shell peanuts from three different sources (Cliina, the United States, and ROW).
The conditional demand system contains three equations representing EU import demand equations for Chinese, the United States, and ROW in-shell peanuts. The additional explanatory variables included in the conditional demand system are time trend and seasonal dummy variables. The time trend is included to model any systematic effect that is not captured by income, price, or seasonal dummy variables (Burton and Young; Piggott et al.). The ROW equation for in-shell peanuts was dropped to avoid singularity problems since the expenditure shares in the conditional demand system sum to one. The parameters of both the unrestricted and restricted conditional demand system are estimated by the seemingly unrelated regression (SUR) method.
Enlarge this image.Table 1. Test Results for Demand Restrictions and Product Aggregation
Tests of homogeneity and symmetry were conducted on the unrestricted demand system, taking into account that cross price effects are source-differentiated within a product. The test of product aggregation was performed on the restricted demand system (Carew, Florkowski, and He). The MODEL procedure from SAS was used for estimation. The Lagrange multiplier test was utilized to test for serially correlated residuals in each equation.
Results
The null hypothesis of no autocorrelation in the Breusch-Godfrey test was rejected, indicating that the data are serially correlated. The system of demand equations was estimated and corrected for first-order autocorrelation. Since the model is a singular equation system, we follow the first-order autocorrelation correction procedure proposed by Berndt and Savin. This approach assumes a constant autocorrelation coefficient in all the equations of the system of equations and zero cross equation autocorrelation. The estimated value of the first autocorrelation coefficient is 0.399, which was significant at the 5% statistical level.
After correcting for the autocorrelation problem, the systemwise RESET test for functional misspecification (p value = 0.925), the Breusch-Pagan systemwise test for heteroskedasticity (p value = 0.745), and the Breusch-Godfrey systemwise test for autocorrelation (p value = 0.244) failed to reject the null hypothesis that there was no misspecification problem. The exogeneity test also failed to reject the null hypothesis that log expenditures were exogenous.
The results of the tests of the homogeneity and symmetry restrictions in the SAIDS model after correction for the first autocorrelation are presented in Table 1 . The likelihood ratio tests failed to reject the null hypothesis that symmetry or homogeneity, or symmetry and homogeneity restrictions are satisfied.
The result of the test of product aggregation is shown in Table 1. A Wald F-test was used for this purpose. The null hypothesis of product aggregation, which maintains that inshell peanuts from different production sources are perfect substitutes, is rejected. These results suggest that in-shell peanuts from different sources are differentiated by EU consumers, which can be attributed to their different quality characteristics. In fact, inshell peanuts are not a homogeneous commodity. They are distinguished by quality characteristics such as freshness and flavor (American Peanut Council).
Parameter Estimates
Estimation results of the nonlinear SAIDS model after the correction of the first-order autocorrelation are shown in Table 2. The time trend variable has a statistically significant negative effect for import demand for U.S. in-shell peanuts, which reflects U.S. export of in-shell peanuts decreasing over time. However, this variable has a positive effect but it is insignificant for EU imports from China and the rest of the world.
Enlarge this image.Table 2. Parameter Estimates for the Restricted Conditional Nonlinear SAIDS Model of EU Import Demand for In-Shell Peanuts (Homogeneity and Symmetry Imposed)
Dummy variables measuring the effects of seasonality show that import demand for U.S. in-shell peanuts is high during the October-December season, which coincides with the harvesting season of peanuts in the United States from September to November and also with the higher consumption demand of the product during the Christmas and New Year holiday season. The imported share of U.S. inshell peanuts declines around 6.7-8.7% in the first and second quarter and rises almost 13.7% in the third quarter. The seasonal dummy variables also show that import demand for Chinese in-shell peanuts does not coincide with the Chinese harvesting season from October to December. This indicates that in China in-shell peanuts are stored and sold during off season. Imported share of Chinese in-shell peanuts increased around 10.7% in the first quarter, rising approximately to 13.7% in the second quarter and 22.2% in the third quarter. All of the seasonal dummy variables are significant for China, and most of the seasonal dummy variables are significant for the United States. The seasonal dummy variables for rest of the world are negative, but they are insignificant.
Elasticities
Conditional expenditure elasticities are reported in Table 3. Conditional expenditures of imported in-shell peanuts from China (0.97) and ROW (0.19) are inelastic, while conditional expenditure of imported in-shell peanuts from the United States (1.42) is elastic. These results suggest that, as the EU's expenditures on in-shell ipeanut imports increase, the EU imports more in-shell peanuts from the United States than from other sources. Expenditure elasticities are usually higher for better products or preferred grades (Tomek and Robinson). These results suggest that U.S. in-shell peanuts have better quality or are perceived as being of better quality by European consumers.
Demand for U.S. in-shell peanuts is more sensitive to income changes than demand for Chinese and ROW in-shell peanuts. These results imply that when the economy or expenditure growth slows down in the EU, this occurrence will have a larger negative impact on the demand for U.S. in-shell peanuts than the demand for in-shell peanuts from other origins. On the other hand, when the economy is growing, this incidence will have a larger positive impact on the demand for U.S. in-shell peanuts than the demand for in-shell peanuts from other origins.
Enlarge this image.Table 3. Mean Elasticities for the Nonlinear SAIDS model of EU Import Demand for InShell Peanuts
Corresponding to the law of demand, all the Mashallian price elasticities (conditional own price elasticities) of demand for in-shell peanuts from different sources are negative (Table 3). The Marshallian own price elasticities of demand for U.S. and Chinese in-shell peanuts are - 1.87 and -1.74, respectively. These high elasticity values suggest that EU consumers are highly sensitive to price changes of in-shell peanuts imported from China and the United States. Hence, if Chinese and/or U.S. exporters can work as a cartel, they may have an incentive to decrease price to raise total sales.
The Marshallian own price elasticity of demand for the ROW in-shell peanuts is less own price elastic (-0.28), but it is insignificant. This indicates that EU consumers' demand for ROW in-shell peanuts is not very sensitive to its own price changes. Something that is important to point out is the fact that only the own and cross price elasticities corresponding to the United States and China were statistically significant because China and U.S. are the main exporters of in-shell peanuts (together they account for more than 85% of the EU imports), while other countries are considered to be small exporters of in-shell peanuts.
The Marshallian own price elasticities of demand for U.S. and Chinese in-shell peanuts are very elastic, probably because China and the United States are the main exporters of inshell peanuts to the EU. They have large volumes of export, quantities of in-shell peanuts, and their export volumes are more sensitive to change in prices, while ROW has a small volume of export quantities of in-shell peanuts, and its export volumes are less sensitive to change in the price level. These high elasticity values can also be the result of the high level of disaggregation being used and the fact that in-shell peanuts are mainly used by consumers as snacks. The elasticity of the aggregate commodity composed of in-shell peanuts from different sources was also calculated using the procedure outlined in Carpio, Wohlgenant, and Safley. This elasticity measures the effect of a proportional change in the price of all types of in-shell peanuts on the aggregate quantity demanded. The elasticity for the aggregate good is estimated to be - 1.00.
The Morishima elasticities of substitution were utilized to identify whether goods are substitutes or complements and to provide information on the degree of substitutability among in-shell peanuts from different sources (Table 3). The Morishima elasticities indicate that in-shell peanuts from China are substitutes for U.S. in-shell peanuts. The Morishima elasticities also indicate that in the EU market, the degree of substitutability between peanuts from China and the U.S. is higher than their degree of substitutability with in-shell peanuts from other countries. This is probably due to the fact that China and the United States are the major exporters of in-shell peanuts to the EU markets.
Enlarge this image.Table 4. Bootstrap Estimates for the Elasticity Standard Errors of the EU Import Demand for In-Shell Peanuts
Standard Errors of Elasticities
MBB estimates for the elasticity standard errors are also shown in Table 3. The standard errors were calculated using the MBB with blocks of size two and the SUR estimation method. The income elasticities based on the bootstrap standard errors were significant for the import demand for Chinese and U.S. inshell peanuts. The Marshallian own price elasticities were significant for Chinese and U.S. in-shell peanuts.
For sensitivity purposes, Table 4 shows the MBB (euro = 2) standard errors using NLS, as well as MBB standard errors using SUR, which were already presented in Table 3. Three block sizes (euro = 2, euro = 4, and t = 6) were included to assess the sensitivity of the results to the choice of the bootstrapping block size. MBB estimates obtained using SUR were in most cases smaller than the estimates obtained using MBB and NLS. Since the NLS and SUR can be seen as Generalized Method of Moments (GMM) estimators, they both result in consistent estimates of the parameters (Greene).
The smaller values of the MBB standard errors obtained with SUR might reflect additional efficiency gains obtained by taking into account the cross equation correlations. With regard to the size of the blocks, the results were robust to the block size choice.
In spite of the fact that the MBB method does not assume any specific dynamic data-generating process, our empirical results show good performance of the MBB method for the calculation of the standard errors of elasticities. Some efficiency gains seem to be achievby using SUR rather than NLS.
Summary and Conclusions
The EU is the largest importer of peanuts in the world; however, little economic research has been done on the EU markets for peanuts. The aim of this research was to estimate the import demand elasticities for in-shell peanuts in the European Union from three different sources: China, the United States, and ROW. A nonlinear source-differentiated AIDS model was used to estimate demand parameters, using data from 1991 to 2005. In the process, theoretical restrictions were also tested.
The null hypothesis of aggregation over product sources was rejected at conventional levels of significance, suggesting that peanuts from different sources are differentiated by the EU market; this can be attributed to their different quality characteristics. The expenditure elasticity for U.S. inshell peanuts is highly elastic, which is an indicator of the superior quality of U.S. inshell peanuts. Own price elasticities indicate that EU consumers respond more to price changes for in-shell peanuts imported from China and the U.S. than from other countries. Finally, the Morishima elasticity results indicate that in-shell peanuts from China are substitutes for U.S. in-shell peanuts and that China and U.S. in-shell peanuts have a higher degree of substitutability than other countries.
Our implementation of the MBB method for the estimation of the standard errors of the elasticities suggests that this method is suitable for estimating the reliability of elasticity estimates in nonlinear demand models. This result is important for practitioners, since currently available software used to estimate nonlinear systems of demand equations (e.g., proc model in SAS) does not provide a multivariate version of an autocorrelation heteroskedastically consistent covariance matrix that could be used to calculate asymptotic standard errors of elasticities. Even if they become available, the calculation of these standard errors in the nonlinear AIDS model is complicated by the fact that the elasticities are nonlinear functions of the parameter estimates. Given the increase in computational resources, standard errors of elasticities obtained using the MBB can be used as an alternative method to assess the reliability of elasticity estimates.
Results imply that the United States can benefit the most when there is an expansion of the EU market to import in-shell peanuts. However, demand for in-shell peanuts in Europe has been steady, while competition among exporters has changed. Therefore, maintaining strong export markets are an important priority for the U.S. in-shell peanut industry. Our findings show that the EU market differentiates in-shell peanuts from source of origin, and evidence from other research (Bliss) suggests that EU consumers perceive U.S. peanuts as being of better quality than peanuts from other sources. These findings have at least two implications. First, U.S. producers and exporters should direct efforts to ensure that in-shell peanuts exported to the EU are of the best possible quality. Second, promotion efforts should stress the quality of U. S peanuts as an advertising tool.
[Received November 2007; Accepted May 2008.]
1 This study only focuses on EU-15 countries because the data for all EU-27 are not available prior to 1995. Moreover, the EU-15 countries account for 99% of the total import quantities for in-shell peanuts in the European Union (EUROSTAT). The countries included in the EU-15 are Austria, Belgium, Germany, Denmark, Spain, Finland, France, United Kingdom, Greece, Ireland, Italy, Luxemburg, Netherlands, Portugal, and Sweden. Currently the EU has 27 country members.
2 The reason for aggregating Latin America and Africa with the rest of the world was empirical. Initially, we modeled the EU import demand for inshell peanuts from five different sources: China, U.S., Latin American, Africa, and rest of the world. The parameters of the EU import demand for Latin American and African in-shell peanuts were mostly not statistically significant.
The average conditional budget share of in-shell peanuts from China and the United States was 0.53 and 0.33, respectively. Therefore, these two countries together account for more than 85% of EU import demand for in-shell peanuts.
3 Unit prices of imported in-shell peanut from each country were computed by dividing total value by total quantity of imports.
4 The EU production is less than 0.0001% of total world production and is less than 0.01% of total EU import of peanuts. The data for EU and world peanut production are available at Production, Supply, and Distribution (PSD online) from the FAS, USDA.
References
American Peanut Council. Peanut Taste and Freshness Are Key. Internet site: http://www.peanutsusa.org.uk/Europe/index.cfm?fuseac-tion=newsrelease.home (Accessed February 28, 2008).
Andayani, S. R. M., and D.S. Tilley. "Demand and Competition among Supply Sources: The Indonesian Fruit Import Market." Journal of Agricultural and Applied Economics 29(December 1997):279-89.
Berndt, E.R., and E.N. Savin. "Estimation and Hypothesis Testing in Singular Equation Systems with Autoregressive Disturbances." Econometrica 43(September-November 1975):937-58.
Blackorby, C, and R.R. Russell. "Will the Real Elasticity of Substitution Please Stand Up - A Comparison of the Allen-Uzawa and Morishima Elasticities." The American Economic Review 79(September 1989):882-88.
Bliss, R. M. "Taste Tests Prove it; U.S. Peanuts are Tops." USDA Agricultural Research Magazine August 2005.
Blundell, R., and J.M. Robin. "Latent Separability: Grouping Goods without Weak Separability." Econometrica 68(January 2000):53-84.
Borges, R. B. "Trade and the Political Economy of Agricultural Policy: The Case of the United States Peanut Program." Journal of Agricultural and Applied Economics 27(December 1995): 595-612.
Borges, R.B., and W.N. Thurman. "Marketing Quotas and Random Yields: Marginal Effects of Inframarginal Subsidies on Peanut Supply." American Journal of Agricultural Economics 76(November 1994):809-17.
Burton, M., and T. Young. "The Impact of BSE on the Demand for Beef and Other Meats in Great Britain." Applied Economics 28(June 1996):68793.
Buse, A. "Evaluating the Linearized Almost Ideal Demand System." American Journal of Agricultural Economics 76(November 1994):781-93.
Carew, R., WJ. Florkowski, and S. He. "Demand for Domestic and Imported Table Wine in British Columbia: A Source-Differentiated Almost Ideal Demand System Approach." Canadian Journal of Agricultural Economics 52(July 2004): 183-99.
Carpio, C.E., M.K. Wohlgenant, and CD. Safley. "A Structural Econometric Model of Joint Consumption of Goods and Recreational Time: An Application to Pick-Your-Own Fruit." American Journal of Agricultural Economics 90(August 2008):644-657.
Chen, K..Z. "The Symmetric Problem in the Linear Almost Ideal Demand System." Economics Letters 59(June 1998):309-15.
Deaton, A., and J. Muellbauer. "An Almost Ideal System." The American Economic Review 70(June 1980):312-26.
European Commission. EUROSTAT. Internet site: http://epp.eurostat.ec.europa.eu/newxtweb/(Accessed December 2006).
Food and Agriculture Organization of the United Nations. FAOSTAT. Internet site: http://faostat.fao.org/site/535/default.aspx#ancor (Accessed April 2007).
Goncalves, S., and H. White. "Bootstrap Standard Error Estimates for Linear Regression." Journal of the American Statistical Association 100(September 2005):970-79.
Green, R., W. Hahn, and D. Rocke. "Standard Errors for Elasticities: A Comparison of Bootstrap and Asymptotic Standard Errors." Journal of Business & Economic Statistics 5(January 1987): 145-49.
Greene, W.H. Econometric Analysis. Upper Saddle River, NJ: Pearson Educ, Ine, 2003.
Hardie, W., J. Horowitz, and J.P. Kreiss. "Bootstrap Methods for Time Series." International Statistical Review 71(August 2003):435-59.
Hayes, D.J., T.I. Wahl, and G. W. Williams. "Testing Restrictions on a Model of Japanese Meat Demand." American Journal of Agricultural Economics 72(August 1990):556-66.
Henneberry, S.R., and S.H. Hwang. "Meat Demand in South Korea: An Application of the Restricted Source-Differendated Almost Ideal Demand System Model." Journal of Agricultural and Applied Economics 39(April 2007):47-60.
Li, H., and GS. Maddala. "Bootstrap Variance Estimation of Nonlinear Functions of Parameters: An Application to Long-Run Elasticities of Energy Demand." The Review of Economics and Statistics 81(November 1999):728-33.
McGuirk, A., P. Driscoll, J. Alwang, and H. Huang. "System Misspecification Testing and Structural Change in the Demand for Meats." Journal of Agricultural and Resource Economics 20(1995):1-21.
Miller, B. R., and C. Mabbs-Zero. "Estimates of Government Intervention Levels in U.S. Peanut Markets." Southern Journal of Agricultural Economics 24(JuIy 1992): 153-62.
Moschini, G. "Units of Measurement and the Stone Index in Demand System Estimation." American Journal of Agricultural Economics 77(February 1995):63-68.
Piggott, N.E., J.A. Chalfant, J.M. Alston, and G.R. Griffith. "Demand Response to Advertising in the Australian Meat Industry." American Journal of Agricultural Economics 78(May 1996): 268-79.
Schmitz, T.G., and J.L. Seale. "Import Demand for Disaggregated Fresh Fruits in Japan." Journal of Agricultural and Applied Economics 34(December 2002):585-602.
Shukur, G. "Dynamic Specification and Misspecification in Systems of Demand Equations: A Test Strategy for Model Selection." Applied Economics 34(April 2002):709-25.
Tomek, W.G., and K.L. Robinson. Agricultural Product Prices. Ithaca, NY : Cornell University Press, 1990.
U.S. Department of Agriculture, Foreign Agricultural Services (FAS). Production, Supply and Distribution (PSD online). Internet site: http://www.fas.usda.gov/psdonî.ine/psdQuery.aspx (Accessed April 2007).
Yang, S.R., and W.W. Koo. "Japanese Meat Import Demand Estimation with the Source-Differentiated AIDS Model." Journal of Agricultural and Resource Economics 19(December 1994):396-408.
Zhang, P., S.M. Fletcher, and D.H. Carley. "Japan's Peanut Import Demand: Implications for United States Exports." Agricultural Economics 11 (September 1994):51-59.
Tullaya Boonsaeng is a postdoctoral associate in the Department of Agricultural Applied Economics at University of Georgia, Griffin, GA. Stanley Fletcher is a professor in the Department of Agricultural Applied Economics at University of Georgia, Griffin, GA. Carlos Carpio is an assistant professor in the Department of Applied Economics and Statistics at Clemson University, Clemson, SC.
Copyright Southern Agricultural Economics Association Dec 2008