Printer Friendly

Did the Uruguay Round Agreement on Agriculture affect trade flows? An empirical investigation for meat commodities.

Abstract The Uruguay Round Agreement on Agriculture (URAA) was designed to reduce trade barriers and temper domestic and export policies that affect trade flows in agricultural commodities. This paper employs a dynamic framework to estimate the effects of the URAA on trade flows, as measured by export volumes, of three meat commodities. The model controls for several important factors, including regional trade agreements, and finds that the URAA had mixed effects on meat exports. Several complicating factors and policy responses, including increasing use of non-negotiated trade barriers, smaller export subsidies and domestic support, and limited effectiveness of market access provisions, explain the mixed empirical results.

Keywords Agricultural trade * Dynamic panel estimation * Market access * Meat commodities * Uruguay round agreement on agriculture

JEL F14 * F15


The primary motivation for countries to enter into multilateral trade agreements is to benefit from the gains from trade. There are two ways that multilateral trade agreements can facilitate member countries in accruing the gains from trade. The first is to establish predictable trading rules by which member countries agree to abide. Such rules reduce the uncertainty associated with international trade and can provide incentives to invest in trading relationships. The second is to bind and reduce border measures and domestic trade-distorting policies in member countries that prevent nations from specializing in and exporting those products in which they may have comparative advantages. Neoclassical trade theory tells us that as trade-distorting policies are removed and countries are free(r) to trade, production patterns will become specialized across countries. Global output will expand and trade flows between countries will grow, leading to increases in global welfare.

The first step towards attaining the gains from trade is the reduction of trade barriers. The Genera! Agreement on Tariffs and Trade (GATT) and its successor, the World Trade Organization (WTO), have gone a long way in reducing border measures and other trade-distorting policies. However, it is not clear that the implementation of multilateral trade agreements have led to increased trade flows between member countries. Studies that have evaluated the impacts of multilateral agreements on trade at the aggregate and industrial levels have had mixed results. Rose (2004a) employed an empirical gravity equation to evaluate whether membership in the GATT/WTO has increased trade between member countries. The GATT/WTO effect is estimated by a binary variable that takes values of one after GATT/WTO accession and zero before (and for countries that are not members of the GATT/WTO). Rose (2004a) found little evidence that membership in the GATT/WTO has increased trade, despite the success of the GATT/WTO in reducing trade barriers (especially ad valorem tariffs) and converting Non-Tariff Barriers (NTBs) to tariff barriers (particularly in non-agricultural industries). These results were contested by Subramanian and Wei (2007) who showed that the GATT/WTO has generally promoted trade flows when modeled in a gravity framework that accounts for missing multilateral resistance terms (Anderson and van Wincoop 2003) and considers asymmetric effects between developed and developing countries and between sectors. (1) Rose (2004b) investigated whether the GATT/ WTO resulted in less volatile and more predictable trade patterns. Rose (2004b) found little evidence that the GATT/WTO led to more predictable trade flows, despite considering the issues raised by Subramanian and Wei (2007).

The objective of this paper is to examine the aggregate effects of the multilateral Uruguay Round Agreement on Agriculture (URAA) on trade in meat commodities using a dynamic empirical framework. The wide variation in trade and domestic support policies for meat commodities across WTO member countries make the analysis of the URAA's effects on meat trade an interesting case. For example, the Most Favored Nation (MFN) tariff rates applied by the European Union (EU) on bovine meat imports remain above 70% after URAA implementation, whereas some countries followed much more liberal trade policies (e.g., Australia). Furthermore, domestic support and export subsidies are prevalent in some countries' meat sectors (e.g., Norway and the EU). This paper unravels the effects of the URAA through a structural shift following the implementation of the URAA (distinction is made at regional levels and between countries at different stages of development). The remainder of the paper is organized as follows. The next section provides an overview of the disciplines that arose from the URAA and outlines current Doha Development Agenda (DDA) modalities for agriculture. We then discuss the expected effects of the URAA on trade flows, The empirical specification, estimation procedure, and data sources are then described, followed by a discussion of the empirical results. The final section provides concluding remarks.

Agriculture in the WTO

Agricultural trade was treated differently from non-agricultural industries in GATT negotiations prior to the Uruguay Round negotiations in 1986. There were no disciplines on quantitative import barriers or export subsidies, and many other distortions related to trade in agricultural products (domestic support and state-trading enterprises) were not disciplined (Warley 1987). Parallel negotiations over a multilateral scheme to reduce trade barriers and trade-distorting subsidies in agricultural sectors were conducted during the Uruguay Round negotiations from 1986 to 1995. These negotiations resulted in the URAA, which came into effect in 1995. The URAA is a binding agreement over three policy categories: market access (Article IV), domestic support (Article VI), and export subsidies (Article IX). The URAA dictated the implementation of disciplines in each of these pillars over 6 years for developed countries (1995-2000) and 10 years for developing countries (1995-2004). The implementation period of the URAA has passed and the DDA negotiations continue (though at a glacial pace); an empirical analysis of the ex post trade effects of the URAA provides insights into the success of the URAA and prospects for increased agricultural trade flows from the DDA.

The URAA called for the tariffication of import-restricting policies by converting NTBs (e.g., variable levies and quotas) into ad valorem tariff equivalents, and imposed ceilings on previously unbound tariff rates. Also, as tariffication resulted in prohibitive tariff levels in some cases, a two-tier Tariff-Rate Quota (TRQ) system was introduced to ensure that historic import quantities were accessible after implementation of the agreement. (2) Developed countries agreed to enhance market access by an unweighted average tariff cut of 36% with a minimum cut per product of 15%. These figures are 15% and 10% for developing countries. Least developed countries were not committed to any reductions. (3) The base tariff rate is generally specified to the custom duty applied at the beginning of September 1986. The commitments were implemented in equal annual instalments throughout the implementation period.

Domestic support and export subsidy disciplines primarily targeted developed countries, where these policies arc most prevalent. The URAA quantified domestic support by an Aggregate Measure of Support (AMS) over policies that art-designated as distorting (e.g., market price supports) at the baseline period 1986-1988. (4) The URAA specified cuts in AMS of 20% and 13% for developed and developing countries, respectively. These cuts applied to programs once the AMS exceeded the de minimis provision of 5% and 10% of the total value of production for developed and developing countries, respectively.

Export subsidies are measured in terms of the value of subsidy expenditure and the volume of subsidized exports. Developed countries were committed to cut the value of subsidy expenditures and the volume of subsidized exports by 36% and 21%, respectively. Developing countries committed to cuts of 24% and 14%, respectively. The baseline period was generally 1986-1990. The cut in the value of subsidy expenditures and the volume of subsidized exports were to be at least 6% and 3.5% in the first year, respectively, with the remaining reductions conducted on equal annual instalments thereafter.

Analyzing the effects of the URAA is pertinent after the scheduled implementation of the URAA commitments by member countries and in light of ongoing DDA negotiations. The success of the URAA in increasing trade flows between member countries can inform expectations about the effects of a new DDA deal on agricultural trade. (5)

Expected Effects of the URAA on Trade

Given the complexity of the URAA and DDA liberalization schemes and unforeseen policy adjustments by countries, it is difficult to develop hypotheses about the direction in which multilateral trade liberalization agreements will affect agricultural trade flows. As trade barriers that would have constrained meat exports are being reduced during the implementation period of the URAA, it is expected that better market access would increase meat exports, ceteris paribus, particularly for countries that are characterized by comparative advantages in meat production. However, better market access could reduce exports from other countries where comparative advantages in meat production are not present. Countries that imported meats from trading partners because of low regional or bilateral trade barriers and, therefore, low landed prices, may now turn to exporting countries with relatively lower costs of production.

Tariff reduction schemes target bound rates and not applied rates; applied rates are below bound rates in many cases. In such cases, market access provisions that do not reduce bound rates to below applied rates would have limited effects on trade flows. For example, Japan's average bound tariff rate on bovine meat imports was 46.9% in 1997 and its average applied tariff rate was just 18.9% (Walkenhorst and Dihel 2003). Japan's bound rate would have to have been cut by more than half before URAA implementation would have any real effects on market access. Similarly, the United States (US) average applied tariff rate on pig meat and poultry meat was 0.6% in 1997; this rate is just one-quarter of the bound rate of 2.4% (Walkenhorst and Dihel 2003). (6) Furthermore, it is not clear that the conversion of NTBs into ad valorem tariff barriers generates equivalent barriers. The conversion follows complex methods and historic baselines that make it difficult to assess the equivalence of tariffs and NTBs.

Disciplines on domestic support and export subsidies/credits are expected to decrease exports, particularly from countries with generous programs such as Norway and the EU. For example, EU export subsidies for bovine meat fell from above US$2 billion in 1995 to approximately US$300 million in 2005, and subsidies for pig meat fell from approximately US$130 million to near US$23 million over the same time period (USDA 2009).

There are additional reasons to believe that reductions in negotiated trade barriers may have smaller than expected effects on trade flows, even after considering the confounding effects of export subsidies. Copeland (1990) argued that reductions in negotiated tariffs can induce policy makers to tighten non-negotiated trade barriers. Anderson and Schmitt (2003) showed that governments have no incentives to introduce NTBs when they are free to set tariffs, but such incentives do exist when tariff's are determined cooperatively (as they are in WTO negotiations). Also, Crowley (2006) argued that the trade effects of negotiated tariff reductions could be undermined by production subsidies to import-competing producers. These measures act as implicit trade barriers and could offset the effects of liberalizing market access provisions. An empirical study by Feinberg and Reynolds (2007) found that Uruguay Round tariff reductions increased the incidence of anti-dumping measures and the total number of anti-dumping petitions.

Trade in agricultural products is particularly prone to disruptions from NTBs. Despite the presence of the WTO's Agreement on the Application of Sanitary and Phytosanitary (SPS) Measures and the Agreement on Technical Barriers to Trade (TBT), agricultural trade has been vulnerable to the erection of non-tariff border measures that impede trade flows. The most high-profile example of such measures is the recent implementation of Mandatory Country of Origin Labelling (MCOOL) in the US. Many observers believe that this policy will impede the flow of meat products into the US (see, for example, Rude et al. 2006; Ward et al. 2009). The negative effects of NTBs, such as MCOOL, on trade in meat commodities may confound the positive effects of lower tariff barriers that resulted from the URAA.

Trade flows can also be impeded through the administration of TRQs. Allocating quota shares according to historical market shares can result in non-neutral application of trade barriers (Monnich 2003), and other TRQ administrative practices (e.g., granting quota on a first-come-first-served or license-on-demand basis) can result in under-filled quotas (Jales et al. 2005). Such practices can undermine market access provisions and can impede growth in trade flows.

Despite these policy responses from governments, it can be argued that the WTO's tarifflcation of some NTBs and the provisions of the URAA have gone some distance in mitigating supply rigidities and preventing the abuse of technical regulations. The empirical investigation in this paper analyzes whether the net effects of the URAA on trade in meat commodities have been significant.

Empirical Specification

We introduce dynamics into a panel data model to characterize the patterns of meat (bovine, pig, and poultry) exports. Theoretical studies indicate that current exports are dependent on historical exports in the presence of entry and adjustment costs (Baldwin 1988; Dixit 1989a, 1989b; Roberts and Tybout 1997). A recent paper by Kandilov and Zheng (2007) found that these costs are very important in agricultural exports; specifically, entry and adjustment costs arc higher for meat and dairy exports than for fruit and vegetable exports. Furthermore, equilibrium trade levels may not be achieved at a given time t. The response of international trade to growth of business and social networks and to improvements in information, communication, and transportation technology is expected to generate lagged adjustment patterns. Introduction of a lagged dependent variable can identify these dynamic patterns in the empirical model.

The effects of the URAA on trade arc detected by testing for the occurrence of structural breaks through evolving patterns of aggregate exports. Disentangling the effects of the URAA requires controlling for a range of macroeconomic and commodity variables, and for the effects of Regional Trade Agreements (RTAs). Let [] represent the total volume of exports of commodity k from country i at time t; the basic autoregressive model is specified as:

In [] = [[alpha].sub.0] + [[alpha].sub.1] In [EXP.sub.[it-l].sup.k] + [[alpha].sub.2] In [PROD.sub.[it-l].sup.k] + [[alpha].sub.3] In [] + [[alpha].sub.4][] + [[alpha].sub.5][] + [[alpha].sub.6][] + [[alpha].sub.7][] + [[alpha].sub.8][TREND.sub.t] + [[mu].sub.i] + [[epsilon]] (1)

where: [PROD.sub.[it-1].sup.k] represents the production size of commodity k, is lagged one year and is treated as a predetermined variable to lessen concerns of contemporaneous correlation. This variable is expected to have a positive effect on exports. The variable [] represents real Gross Domestic Product (GDP) per capita and serves as a proxy for the level of technology, infrastructure development, and capital-to-labor ratio in the exporting countries (Bergstrand 1989). This variable is expected to positively impact export volumes. The primary independent variable of interest, [] is a break binary variable taking the value of one from 1995 (the first implementation year of the URAA) onward for countries that joined the WTO in 1995 or prior to 1995, and from the WTO accession year onward for countries that became WTO members after 1995 and are bound by the URAA. Improved market access resulting from reduced trade barriers is expected to increase meat exports, ceteris paribus. However, as discussed in the previous section, disciplines on export subsidies and domestic support, as well as limitations in market access provisions and higher non-negotiated trade barriers, could confound these effects. The estimate of the coefficient on [] reveals the aggregate net effects of the URAA on meat exports. The effects of the URAA are likely to occur gradually over time due to lagged adjustments and the phased nature of the agreement. Hence, the estimated coefficients on the URAA binary variables capture the average effect over the post-URAA period.

The effects of RTAs are captured by the set of binary variables [] which take the value of one when country i is a member of an RTA at time t and zero otherwise. The set of RTAs that are considered in this specification includes the EU, Mercado Comun del Sur (MERCOSUR), North American Free Trade Agreement (NAFTA), Association of Southeast Asian Nations (ASEAN), and the Andean Community of Nations (ANDEAN). The implications of Bovine Spongiform Encephalopathy (BSE) and Avian Influenza (AI) outbreaks are controlled by adding the binary variables [] and [] to the bovine meat equation and to the poultry meat equation, respectively. These binary variables take the value of one for exporting countries that are designated as outbreak areas by the World Organization for Animal Health and zero otherwise. Outbreaks of BSE and Al are expected to reduce export volumes. The variable Trend; is a time trend, and [[mu].sub.i] ~ IID (0, [[sigma].sub.[mu].sup.2]) and [[epsilon]] ~ IID(0, [[sigma].sub.[epsilon].sup.2]) are assumed to be independent of each other and among themselves. Finally, the well-documented substitutability between meat commodities (Chen 1998; Wachenhcim et al. 2004) suggests that trade in one meat commodity can replace trade flows of other meat commodities. Conversely, it can be argued that trade in one meat commodity can establish or strengthen business networks in the meat sector, resulting in the enhancement of exports of other meat commodities. We include exports of other meat commodities as lagged and predetermined variables to capture the net influence of these opposing effects.

Arellano and Bond (1991) implemented a difference Generalized Method of Moments (GMM) estimator to estimate dynamic panel equations where differencing wipes out unobserved country-specific effects. The moment conditions exploit the orthogonality conditions between the differenced errors and lagged values of the dependent variable. Blundell and Bond (1998) argued that lagged levels are weak instruments to the first differenced equations when individual series are persistent over time and when there is a relatively small number of time-series observations. In a follow-up paper, Blundell and Bond (2000) demonstrated that the first-differenced GMM estimator has a large finite sample bias and limited precision. They showed that these potential problems are attenuated by considering the adjustments suggested by Arellano and Bover (1995) and by Blundell and Bond (1998). These adjustments consist of complementing the differenced equations with the original set of equations in levels where lagged differences are employed as instruments. The system is then estimated as a whole.

The GMM-system estimator with heteroskedasti city-robust standard errors is considered in the empirical specification. (7) The GMM-system estimator is associated with two essential tests proposed by Arellano and Bond (1991), Arellano and Bover (1995), and Blundell and Bond (1998). The first test is the Sargan/Hansen test of over-identifying restrictions with the null hypothesis being the inability to reject the over-identification restrictions. The second test is for the presence of serial correlation. While first-order autocorrelation in differenced residuals maintains the consistency of estimates, second-order autocorrelation does not. Therefore, the second test diagnoses second-order serial autocorrelation. Failure to reject the null hypothesis implies that the original error term is serially uncorrelated.

Export and production data are compiled from Food and Agricultural Organization Statistics (FAOSTAT). The definitions of bovine meat, pig meat, and poultry meat correspond to the FAOSTAT codes of 867, 1035, and 1058, respectively. (8) Real GDP per capita in current US dollars are derived from the Penn World Table 6.2 database. Summary statistics are provided in Table 1. The final dataset includes 106 WTO member countries and covers 25 years (from 1981 to 2005). The list of countries and years of accession into the GATT/WTO are presented in Table 4 of the Appendix.
Table 1 Summary statistics

           Variable              Mean    Standard  Minimum    Maximum

Exports (bovine meat, MT)       14,569     49,604       0      749,641

Exports (pig meat, MT)          25,193    100,485       0    1,173,346

Exports (poultry meat, MT)      35,398    192,860       0    2,794,685

Production (bovine meat, MT)   416,990  1,289,150     120   12,426,974

Production (pig meat, MT)      648,287  3,240,385       0   51,200,507

Production (poultry meat, MT)  375,149  1,286,111     122   15,869,030

Real GDP per capita (current     7,570      7,894     271       41,188


Table 4 List of countries and GATT/WTO accession year

Country               GATT/WTO Accession


Albania                      2000

Antigua and Barbuda          1987

Argentina                    1967

Australia                    1948

Austria                      1951

Bangladesh                   1972

Barbados                     1967

Belgium and Lux.             1948

Belize                       1983

Benin                        1963

Bolivia                      1990

Brazil                       1948

Brunei Darussalam            1993

Bulgaria                     1996

Burkina Faso                 1963

Burundi                      1965

Cambodia                     2004

Cameroon                     1963

Canada                       1948

Central African Rep.         1963

Chad                         1996

Chile                        1949

China                        2001

Colombia                     1981

Congo                        1963

Congo, Dem. Rep.             1997

Costa Rica                   1990

Cote d'Ivoire                1963

Cuba                         1995

Cyprus                       1963

Denmark                      1950

Dominica                     1993

Dominican Rep.               1950

Ecuador                      1996

Egypt                        1970

El Salvador                  1991

Fiji                         1996

Finland                      1950

France                       1948

Gabon                        1963

Germany                      1951

Ghana                        1957

Greece                       1950

Grenada                      1994

Guatemala                    1991

Guinea                       1994

Guinea Bissau                1995

Guyana                       1966

Haiti                        1950

Honduras                     1994

Hungary                      1973

India                        1948

Indonesia                    1950

Ireland                      1967

Israel                       1962

Italy                        1950

Jamaica                      1963

Japan                        1955

Jordan                       2000

Kenya                        1964

Madagascar                   1963

Malawi                       1964

Malaysia                     1957

Mali                         1993

Malta                        1964

Mauritania                   1995

Mauritius                    1970

Mexico                       1986

Morocco                      1987

Mozambique                   1992

Netherlands                  1948

New Zealand                  1948

Nicaragua                    1950

Niger                        1963

Nigeria                      1960

Norway                       1948

Pakistan                     1948

Panama                       1997

Peru                         1951

Philippines                  1979

Poland                       1967

Portugal                     1962

Rep. of Korea                1967

Romania                      1971

Rwanda                       1966

Saudi Arabia                 2005

Senegal                      1963

Solomon Islands              1996

South Africa                 1948

Spain                        1963

Sri Lanka                    1948

Suriname                     1978

Sweden                       1950

Switzerland                  1966

Thailand                     1982

Togo                         1964

Trinidad & Tobago            1962

Tunisia                      1990

Turkey                       1951

Uganda                       1962

United Arab Emirates         1994

United Kingdom               1948

Tanzania                     1961

United States                1948

Venezuela                    1990

Zimbabwe                     1948

Estimation Results

Table 2 displays the results from GMM-system estimation for all three commodities where the effects of the URAA on total exports are disaggregated for developed and developing countries. (9), (10) Table 2 also reports the Arellano-Bond test of AR(1) and AR(2), and the Sargan/Hansen test of over-identifying restrictions. The outcomes from both tests are satisfactory. The former test indicates the consistency of the GMM estimator and the latter test implies failure to reject the null hypothesis of over-identification restrictions.
Table 2 G MM-system estimation of the trade equations (effects of the
URAA on exports from developed and developing countries)

                                   Bovine Meat          Pig Meat

URAA (Developed countries)      0.246 (a) (0.087)     -0.045 (0.121)

URAA (Developing countries)        -0.042 (0.071)      0.069 (0.096)

[ln(Exp).sub.[t-1]] (Bovine     0.817 (a) (0.031)  0.065 (b) (0.028)

[ln(Exp).sub.[t-1]] (Pig meat)  0.072 (b) (0.035)  0.788 (a) (0.029)

[ln(Exp).sub.[t-1]] (Poultry    0.059 (b) (0.030)      0.022 (0.024)

[ln(Prod).sub.[t-1]]            0.153 (a) (0.042)  0.073 (a) (0.011)

ln(RGDPC)                       0.216 (a) (0.055)  0.102 (a) (0.023)

EU                              0.56l (a) (0.180)  0.613 (a) (0.141)

MERCOSUR                        0.110 (c) (0.066)      0.121 (0.090)

NAFTA                           0.272 (b) (0.135)  0.293 (c) (0.151)

ANDEAN                              0.098 (0.178)     -0.033 (0.104)

ASEAN                              -0.110 (0.156)      0.062 (0.134)

BSE                             0.136 (a) (0.051)


Trend                           0.009 (c) (0.005)  0.012 (b) (0.007)

Observations                                 2544               2544

Sargan/Hansen test (p-value)                 0.99               0.99

Arellano-Bond test AR(1)                     0.00               0.00

Arellano-Bond test AR(2)                     0.22               0.30

                                       Poultry Meat

URAA (Developed countries)          -0.197 (b) (0.083)

URAA (Developing countries)         -0.153 (b) (0.061)

[ln(Exp).sub.[t-1]] (Bovine meat)    0.037 (c) (0.022)

[ln(Exp).sub.[t-1]] (Pig meat)       0.048 (c) (0.026)

[ln(Exp).sub.[t-1]] (Poultry meat)   0.769 (a) (0.035)

[ln(Prod).sub.[t-1]]                 0.234 (a) (0.069)

ln(RGDPC)                            0.188 (a) (0.034)

EU                                   0.389 (a) (0.132)

MERCOSUR                             0.376 (a) (0.120)

NAFTA                                    0.088 (0.175)

ANDEAN                                   0.106 (0.132)

ASEAN                                    0.050 (0.144)


AI                                  -0.194 (a) (0.072)

Trend                                0.019 (a) (0.004)

Observations                                      2544

Sargan/Hansen test (p-value)                      0.99

Arellano-Bond test AR(1) (p-value)                0.00

Arellano-Bond test AR(2) (p-value)                0.23

Standard errors are in parentheses. The superscripts "a", "b", and "c"
denote significance at 1%, 5%, and 10% level, respectively

In the case of bovine meat and poultry meat, the coefficients on the URAA variable for developed countries are negative and significant at the 1% and 5% level, respectively. The coefficient on the URAA for developed countries is insignificant in the case of pig meat. For developing countries, the coefficient on the URAA variable is negative and significant at the 5% level in the case of poultry meat but is insignificant in the case of bovine meat and pig meat.

The provisions of the URAA were designed to enhance market access, with the objective of increasing trade flows. However, the estimated coefficients capture the net effects of implementing the URAA market access provisions, as well as many other factors that could have confounded the effects of more liberal market access. It is therefore not surprising that some estimated URAA coefficients are negative and some are insignificant. The reduction in export subsidies during the URAA implementation period goes some distance in explaining the net outcome of negative coefficients on the URAA variable. Many developed countries (e.g., the EU and Norway) provided generous export subsidy programs prior to the URAA. The URAA imposed reduction schedules on member countries, which would have resulted in less subsidized exports and fewer exports of once-subsidized products. This effect could offset market access enhancements achieved through the URAA. Consequently, the net effects of overall liberalization on exports could be negligible or even negative because the implementation of the URAA commitments on export subsidy programs would result in higher export prices and fewer exports. These results are reminiscent of Subramanian and Wei's (2007) finding that accession into the GATT/WTO had a negative effect on aggregate agricultural trade.

Furthermore, developed countries may have tightened existing, or initiated new forms of trade barriers, following the implementation of the URAA, as described in Copeland (1990), Anderson and Schmitt (2003), Crowley (2006), and Feinberg and Reynolds (2007). Such barriers could compensate for reductions in negotiated trade barriers and, therefore, could offset the positive effects of lower tariffs on trade flows. Also, market access provisions might not have been effective, given that reduction commitments targeted bound tariff rates and not applied tariff rates. Barriers were converted to ad valorem tariffs using the base period of 1986, so reductions in bound rates had no impact on market access for countries whose applied tariff rates were below their bound rates during the implementation period.

Other results from Table 2 show that an increase in domestic production has a positive effect on exports of all meat commodities. Also, the results indicate that an increase in real income has positive and significant effects on meat exports, reflecting the effects of technology, infrastructure development, and capital-to-labor endowments. The accession of an exporter into an RTA is expected to increase total exports due to the removal or reduction of trade barriers with other member countries. Hence, the effects of the RTA control variables on aggregate exports are expected to be positive a priori. Our empirical results indicate significant positive effects of RTAs for countries that are members of the EU, MERCOSUR, and NAFTA. Hence, accession into RTAs has had significant effects on meat exports for these countries, on average. The effects of ANDEAN and ASEAN are found to be insignificant for all meat commodities. The estimated coefficients on the BSE and Al variables are both negative and significant at the 1% level. These events had clear effects on aggregate international movements of bovine meat and poultry meat from outbreak countries.

The results from the bovine meat equation also show that exports of pig meat and poultry meat have positive and significant effects on bovine meat exports at the 5% level. These results can be partly explained through spill-over effects from business networks created through trade in pig meat and poultry meat on trade in bovine meat. The effect of bovine meat exports on pig meat exports is also positive and significant at the 5% level whereas the effect of poultry meat exports on pig meat exports is not significant. Bovine meat and pig meat exports have positive effects on poultry meat exports at the 10% level.

Table 3 disaggregates the effects of the URAA on trade flows down to the regional level. The coefficients on the URAA binary variables for Europe arc negative and significant at the 1% and 10% level in the case of bovine meat and poultry meat, respectively, and insignificant in the case of pig meat. As described earlier, the URAA disciplines on export subsidies and domestic support programs of the EU and other European countries (e.g., Norway) can go a long way in explaining these results. For example, EU export subsidies for bovine meat fell from US$2 billion in 1995 to approximately US$300 million in 2005. The coefficients on the URAA binary variable for North America arc insignificant in the case of bovine meat and pig meat and negative and significant at the 10% level in the case of poultry meat. The coefficients on the URAA binary variables for other regions are insignificant in the case of bovine meat and pig meat. In the case of poultry meat, the coefficients for North Africa and the Middle East, and for Sub-Saharan Africa are negative and significant at the 10% and 5% level, respectively.
Table 3 GMM-system estimation of the trade equations (effects of the
URAA on exports from different geographic regions)

                                    Bovine Meat           Pig Meat

URAA (Europe)                   -0.212 (a) (0.081)      0.025 (0.113)

URAA (North America)                -0.142 (0.135)      0.089 (0.132)

URAA (Central and South             -0.045 (0.063)     -0.083 (0.104)

URAA (North Africa and the          -0.094 (0.132)     -0.087 (0.085)
Middle East)

URAA (Sub-Saharan Africa)            0.021 (0.070)     -0.039 (0.064)

URAA (Oceania)                       0.097 (0.180)     -0.046 (0.050)

URAA (Asia)                          0.078 (0.105)      0.104 (0.116)

[ln(Exp).sup.[t-1]] (Bovine      0.806 (a) (0.029)  0.078 (b) (0.032)

[ln(Exp).sub.[t-1]] (Pig meat)   0.065 (b) (0.032)  0.794 (a) (0.028)

[ln(Exp).sub.[t-1]] (Poultry     0.069 (b) (0.031)      0.024 (0.026)

[ln(Prod).sub.[t-1]]             0.164 (a) (0.046)  0.079 (a) (0.013)

ln(RGDPC)                        0.210 (a) (0.051)  0.118 (a) (0.027)

EU                               0.531 (a) (0.174)  0.586 (a) (0.140)

MERCOSUR                         0.096 (c) (0.054)      0.102 (0.105)

NAFTA                            0.281 (b) (0.138)  0.304 (c) (0.160)

ANDEAN                              -0.080 (0.191)     -0.053 (0.120)

ASEAN                               -0.113 (0.151)      0.087 (0.128)

BSE                             -0.151 (a) (0.053)


Trend                            0.008 (c) (0.004)  0.013 (b) (0.007)

Observations                                  2544               2544

Sargan/Hansen test (p-value)                  0.99               0.99

Arellano-Bond test AR(I)                      0.00               0.00

Arellano-Bond test AR(2)                      0.23               0.30

                                            Poultry Meat

URAA (Europe)                            -0.124 (c) (0.070)

URAA (North America)                     -0.186 (c) (0.103)

URAA (Central and South America)              0.122 (0.092)

URAA (North Africa and the Middle East)  -0.202 (c) (0.114)

URAA (Sub-Saharan Africa)                 0.185 (b) (0.087)

URAA (Oceania)                               -0.067 (0.078)

URAA (Asia)                              -0.195 (b) (0.088)

[ln(Exp).sup.[t-1]] (Bovine meat)         0.042 (c) (0.023)

[ln(Exp).sub.[t-1]] (Pig meat)            0.055 (c) (0.029)

[ln(Exp).sub.[t-1]] (Poultry meat)        0.745 (a) (0.038)

[ln(Prod).sub.[t-1]]                      0.248 (a) (0.065)

ln(RGDPC)                                 0.166 (a) (0.030)

EU                                        0.341 (a) (0.121)

MERCOSUR                                  0.372 (a) (0.115)

NAFTA                                         0.079 (0.186)

ANDEAN                                       -0.143 (0.125)

ASEAN                                         0.073 (0.124)


AI                                       -0.184 (a) (0.070)

Trend                                     0.017 (a) (0.005)

Observations                                           2544

Sargan/Hansen test (p-value)                           0.99

Arellano-Bond test AR(I) (p-value)                     0.00

Arellano-Bond test AR(2) (p-value)                     0.24

Standard errors are in parentheses. The superscripts "a", "b", and "c"
denote significance at 1%, 5%, and 10% level, respectively

Lagged meat production is treated as a predetermined variable in the benchmark regressions to alleviate concerns of contemporaneous correlation with the level of exports. An alternative specification of the empirical model that includes current production as an endogenous variable is also estimated and generates results that are equivalent to the benchmark results. The model is also estimated with an alternative specification that allows for different coefficients on the lagged exports and production variables for the pre-1995 period and for the 1995-onward period." The coefficients on the export and production variables are similar across sub-periods. The coefficients on the URRA binary variables are slightly lower than, but qualitatively equivalent to those obtained from the benchmark equation.

The empirical results do not provide strong evidence that the URAA increased trade flows of meat commodities. Estimated coefficients on the URAA variable are not uniformly significant and are frequently negative. The numerous confounding effects discussed above clearly affected the impact of the URAA's market access provisions on meat exports.

Concluding Remarks

The implementation period of the URAA has passed and the DDA negotiations continue; an empirical analysis of the implications of the URAA is a necessary step in the assessment of past and future trade liberalization. Given the complexity of the URAA schemes and subsequent policy adjustments in member countries, it is not clear that the URAA has significantly affected agricultural trade flows. This paper employs a dynamic empirical framework to estimate the effects of the URAA on trade of three meat commodities through the occurrence of structural breaks in the evolving patterns of trade flows. The effects of the URAA on meat exports are disentangled for developed and developing countries and for a range of geographic regions.

The empirical results do not lend support to the hypothesis that the URAA significantly affected trade flows in meat commodities. Estimated coefficients on the URAA variable arc frequently insignificant and are often negative. Many confounding factors that arose during the URAA implementation period (e.g., lower export subsidies, higher non-negotiated trade barriers, commitments that targeted bound rates and not applied rates) have offset market access provisions at the aggregate level. Considerable increases in trade flows may await further market access provisions to be realized upon completion of the DDA negotiations. The DDA draft modalities for agriculture include ambitious proposals to cut tariff barriers and to expand tariff quotas. Market access for meat commodities will increase if a deal is done. The draft modalities also outline disciplines that will eliminate export subsidies, thereby confounding the effects of enhanced market access. Also, it remains to be seen how successful a DDA agreement could be in constraining other NTBs, such as SPS and labelling measures.

It is important to note that the empirical results are aggregate and do not imply that some countries did not experience increased meat exports after the implementation of the URAA. Also, caution is warranted when generalizing the results of this analysis beyond meat commodities. Though meat commodities account for approximately 10% of global agricultural exports (FAOSTAT), the structure of trade and domestic support policies vary widely across agricultural commodities. Hence, the effects of the URAA on trade flows of other agricultural commodities may be markedly different.


Anderson, S. P., & Schmitt, N. (2003). Non-tariff barriers and trade liberalization. Economic Inquiry, 41 (1), 80-97.

Anderson, J. H., & van Wincoop. H. (2003). Gravity with gravitas: a solution to the border puzzle. The American Economic Review, 93(1), 170-192.

Arellano, M., & Bond, S. (1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies, 58(2), 277-297.

Arellano, M., & Bover, O. (1995). Another look at the instrumental variable estimation of error component models. Journal of Econometrics, 68(1), 29-51.

Baldwin, R. (1988). Hysteresis in import prices: the beachhead effect. The American Economic Review, 78 (4), 773-785.

Bergstrand, J. H. (1989). The generalized gravity equation, monopolistic competition, and the factor-proportions theory in international trade. The Review of Economics and Statistics, 7/(1), 143-153.

Blundell, R. W., & Bond, S. R. (1998). Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics, 87(1), 115-143.

Blundell, R. W., & Bond, S. R. (2000). GMM estimation with persistent panel data: an application to production functions. Econometric Reviews. 19(3), 321 340.

Chen, K. Z. (1998). The symmetric problem in the linear almost ideal demand system. Economics Letters. 59(3), 309-315.

Copeland, B. R. (1990). Strategic interaction among nations: negotiable and non-negotiable trade barriers. Canadian Journal of Economics. 23(1), 84-108.

Crowley, M, (2006). The agreement on subsidies and countervailing measures: Tying one's hands through the WTO. Federal Reserve Bank of Chicago. Working Paper, WP-06-22.

Dixit, A. K. (1989a). Hysteresis, import penetration, and exchange rate pass-through. Quarterly Journal of Economics. 104(2), 205-228.

Dixit, A. K. (1989b). Entry and exit decisions under uncertainty. Journal of Political Economy, 97(3), 620-638.

Feinberg, R. M., & Reynolds. K. M. (2007). Tariff liberalisation and increased administrative protection: is there a quid pro quo? The World Economy. 30(b). 948-961.

Jales, M., Josling, T, Nassar, A., & Turwilcr, A. (2005). Options for agriculture: From framework to modalities in market access/Domestic support/export competition. International Agricultural Trade Research Consortium, Trade Policy Issues Paper, 2.

Kandilov, I., & Zheng, X. (2007). Are entry and adjustment costs in international agricultural markets important? North Carolina State University, Department of Agricultural and Resource Economics. Working Paper.

Monnich, C. (2003). Tariff rate quotas: Docs administration matter? Center for International Development and Environmental Research, Discussion Paper, 26465.

Roberts. M. J., & Tybout, J. R. (1997). The decision to export in Colombia: an empirical model of entry with sunk costs. The American Economic Review, 87(4), 545-564.

Rose, A. (2004a). Do we really know that the WTO increases trade? The American Economic Review. 94 (I), 98-114.

Rose, A. (2004b). Docs the WTO make trade more stable? National Bureau of Economic Analysis (NBER). Working Paper. 10207.

Rude, J., Iqbal, J., & Brcwin, D. (2006). This little piggy went to market with a passport: the impacts of U. S. country of origin labeling on the Canadian pork sector. Canadian Journal of Agricultural Economics. 54(3), 401-420.

Subramanian, A.. & Wei, S.-J. (2007). The WTO promotes trade, strongly but unevenly. Journal of International Economics. 72(1), 151-175.

United States Department of Agriculture (USDA). (2009). WTO agricultural trade policy commitments database: WTO export subsidy notifications. Available online at: database/. Accessed March 19, 2010.

Wachcnheim. C. J., Mattson, J. W., & Koo. W. W. (2004). Canadian exports of livestock and meat to the United States. Canadian Journal of Agricultural Economics. 52(1), 55-71.

Walkenhorst, P., & Dihcl, N. (2003). Tariff bindings, unused protection, and agricultural trade liberalisation. OECD Economic Studies. 36(1), 231-249.

Ward, E. C, Schrocder, T. C, & Schulz, L. (2009). Impacts from government regulations on the Canadian-U.S. basis for fed cattle. Agricultural and Applied Economics Association (AAEA), 2009 Annual Meeting, Selected Paper.

Warley, T. K. (1987). Issues facing agriculture in the GATT negotiations. Canadian Journal of Agricultural Economics. 35(3), 515-534.

World Trade Organization (WTO). (2008). Revised draft modalities for Agriculture. TN/AG/W/4/Rev.4.

(1) Subramanian and Wei (2007) found that membership in the GATT/WTO had a negative effect on aggregate trade flows in agricultural products. They associated these results with protection permissiveness (i.e., tolerance to protectionist measures) in the agricultural sector.

(2) In many cases, TRQs act de facto as import quotas because over-quota tariff rates arc prohibitively high.

(3) The URAA allows for exemptions under the special treatment and safeguard provisions. Countries arc allowed to charge higher tariffs when prices are below, and imports are above, a trigger threshold.

(4) Support policies that are kept outside this measure are those categorized as green box (e.g., retirement schemes, training, R&D) and blue box (e.g., payments associated with limiting acreage, number of livestock head, and yield).

(5) See Revised Draft Modalities for Agriculture (WTO 2008) fur more details on DDA negotiations.

(6) Many countries reduced the applied MFN rates following implementation of the URAA. For example, estimated figures from the Trade Analysis and Information System (TRAINS) database indicate that the EU applied MFN tariff rates on bovine meat have decreased from 110% in 1995 to 80% in 2005.

(7) We report the results from the one-step robust estimation. The two-step robust model is also estimated and generates quantitatively and qualitatively analogous results.

(8) It is worth noting that the numerous countries in the sample that arc major exporters of meat commodities also imported meat commodities over the same sample period; this phenomenon is known as cross-hauling. For example, Canada exported 110.266 MT and imported 6,315 MT of bovine meat in 1995. The US exported 67,620 MT and imported 108,111 MT of bovine meat in 1995, mostly from Canada.

(9) We follow the International Monetary Fund (IMF) classification of developed and developing countries.

(10) The first URAA binary variable is assigned a value of one for developed countries at the year of their implementation and zero otherwise. The second URAA binary variable is constructed in the same manner for developing countries.

(11) These variables are interacted with the URAA binary variable to identity pre- and post-URAA effects.

P. L. Ghazalian ([??])

Department of Economics. University of Lethbridge, Lethbridge, AB, Canada


R. Cardwell

Department of Agribusiness and Agricultural Economics, University of Manitoba, Winnipeg, MB, Canada

DOI 10.1007/s11294-010-9278-8
COPYRIGHT 2010 Atlantic Economic Society
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Comment:Did the Uruguay Round Agreement on Agriculture affect trade flows?
Author:Ghazalian, Pascal L.; Cardwell, Ryan
Publication:International Advances in Economic Research
Geographic Code:3URUG
Date:Nov 1, 2010
Previous Article:Equity risk factors and macroeconomic growth in Greece.
Next Article:Exchange rate determination: an application of a monetary model for Brazil.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters