Recipient Organization
VIRGINIA POLYTECHNIC INSTITUTE
(N/A)
BLACKSBURG,VA 24061
Performing Department
Agricultural & Applied Economics
Non Technical Summary
International trade in agricultural products is vital to the prosperity of agriculture and food industries and to the well-being of food consumers. Producers gain access to foreign markets, expand their sales, and receive higher prices and incomes. Consumers benefit from lower prices and a greater variety and more consistent supplies throughout the year. In the U.S, the value of agricultural exports now represents 20% of gross farm receipts and every dollar of exports creates an additional $1.17 in supporting activities to process, package, finance and ship agricultural products (USDA ERS 2018).The World Trade Organization and its predecessor, the General Agreement on Tariffs and Trade, were set up to promote the movement of goods across international borders and discourage the use of protectionist policies set unilaterally by individual countries. Trade negotiations to improve the terms for market access at the bilateral and regional levels have also flourished over the past 30 years.Meanwhile, the U.S. trade policy orientation radically changed since 2017 under the Trump administration. The United States withdrew from the Preferential Trade Agreements (PTA) negotiations in which it was involved: The Trans-Pacific Partnership (TPP, that became without the U.S. the CPTPP) and the Transatlantic Trade and Investment Partnership (TTIP) with the European Union. Instead, the U.S. strengthened ties through the establishment or consolidation of bilateral and smaller regional agreements. For example, the U.S.-Japan Trade Agreement signed on October 2019 will eliminate or reduce tariffs on certain agricultural and industrial products between the U.S. and Japan. The new Agreement between the U.S., Mexico and Canada (USMCA) signed on December 2019 is also part of the U.S. policy promoting regional PTAs. This agreement builds on the former North American Free Trade Agreement (NAFTA) but goes further in terms of intellectual property, digital trade, financial services and the environment.The landscape of international trade has also changed significantly in recent years as a result of the trade "war" of unprecedented magnitude between the U.S. and some of its largest and most significant trading partners. China, Canada, Mexico, and the EU have responded to increases in U.S. manufacturing tariffs with retaliatory tariffs against a multitude of U.S. agricultural products. These tariffs lead to higher food costs for the retaliating nation's consumers and lower prices and lost export markets for U.S. producers (Grant et al. 2019). The effects of the particularly fierce trade war between China and the U.S. may be mitigated by the signature of the U.S.-China Phase One economic and trade agreement on February 14, 2020 (Bown 2020).The emergence of Covid-19 in early 2020 and the containment policies and economic crisis that followed, however, makes volumes of trade and the implementation of trade agreements uncertain. Generally, the Covid-19 led to a major disruption of international trade, whose decline is difficult to assess at the present time. According to WTO press release of April 2020, world merchandise trade is set to plummet by between 13% and 32% in 2020 due to the Covid-19 pandemic. The future of international trade relationship is also uncertain, as the economic crisis may lead some countries to apply new unilateral trade restrictions, particularly in the strategic agricultural sector, as has recently been the case in Russia, or Vietnam who decided in March 2020 to restrict their exports of wheat or rice (Glauber et al. 2020).The agreements, dispute tariff actions and Covid-19 related trade disruptions receive much attention in the media and policy world. However, the varied nature of these crisscrossing policy changes and disruptions makes it difficult to assess how they are redefining trade patterns. While the various agreements undoubtedly create commercial opportunities for the participating countries, they may also disadvantage the other exporters that do not benefit from tariff's reductions. Trade disputes increase costs and uncertainties for exporters, but may also favor world competitors who are not involved in such disputes. The Covid-19 crisis has disrupted trade and potential trade restriction induced by the crisis may significantly impact trade and reduce the effect of trade agreements.The goal of this proposal is to analyze the impact of recent trade policy shifts on world agricultural and food trade, with a specific focus on U.S. exports. The project will perform an ex-post analysis of the different trade policy events, in order to provide insights on the recent changes in world agricultural trade. We will estimate the impact of trade agreements, trade disputes and trade restrictions both on trade values and prices, as trade policies may also affect export prices through changes in world excess demand or supply. While the effects of trade policies may be difficult to assess after the 2020 disruption related to Covid-19, we expect to observe trade creation, importer price reductions and exporter prices increases in the case of trade agreements, and trade reduction, importer price increases and exporter price reductions in the case of trade disputes or trade restrictions. Trade agreements without the U.S. may reduce U.S. exports and increase trade flows of U.S. competitors, what is called trade diversion in the literature (Lipsey 1957). The effects on price in this latter case are unclear.The methodology employs a structural gravity model (Head and Mayer 2013; Yotov et al. 2016), which is an econometric model of trade flows that controls for all supply and demand determinants and allows for assessment of the impact of trade agreements, disputes or other measures on trade flows, at the aggregated and product level. The main difficulty, but also the interest, of this study will be to measure the overall impact of simultaneous trade policies, which may have opposite effects on trade. Another difficulty will be to deal with the major trade disruption entailed by the Covid-19 crisis. We will thus distinguish the effect of trade policies before and after the crisis, as this latter is likely to lead to a deep change in the patterns of world trade.At the end, the project will provide an evaluation of gains and losses for American farmers due to U.S. trade agreements, to the signature of trade agreements between U.S. competitors (as Australia, Brazil) and U.S. destination markets (as Asian countries or Europe), to U.S. trade disputes, and to unilateral trade restrictions. It will permit to rank the U.S. agricultural and food products whose export values and export prices are most affected in the current trade policy landscape, either positively or negatively. The analyze at the sector level will also allow to discuss the potential impacts of trade policy shifts for the Virginian agriculture and food industry more specifically.Agriculture and agri-food industries are among Virginia's largest sectors. According to the Virginia Department of Agriculture and Consumer Services (VDACS), their economic impact corresponds to $70 billion annually and the sector provides more than 334,000 jobs in the Commonwealth.International trade is growing in importance at the state level. Although Virginia ranks 33rd in agricultural export sales among the 50 States, Virginia agriculture and food industry are dependent on foreign markets for its output (VDACS 2020). In 2018 agricultural and forestry exports were valued at $2.97 billion. By volume, they accounted for more than half of containerized exports moving through the Port of Virginia (VDACS 2020). Soybean, livestock and tobacco make up the top three exporting categories. Trade agreements and trade disputes are thus expected to have a direct impact on Virginia agriculture, in particular those with Canada and China which are the main destinations of Virginia's agricultural exports (VDACS 2020).
Animal Health Component
70%
Research Effort Categories
Basic
30%
Applied
70%
Developmental
0%
Goals / Objectives
The main goal of the proposal is to assess the extent to which the trade policy changes between 2017 and 2021 impacted the values and prices of international trade and U.S. exports. The considered trade policies are the following:The U.S.-Japan Trade Agreement October 2019The U.S.-China Phase One economic and trade agreement February 2020The new Agreement between the U.S., Mexico, and Canada (USMCA) December 2019The Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) December 2018The U.S. trade dispute with China, the EU, Mexico, Canada and Turkey that began in 2018-2019The potential unilateral trade restrictions (both at the export and import level) the countries may apply during various periods following the Covid-19 crisisThese various events are expected to have diverse effects on world trade and U.S. exports. While the trade agreements signed by the U.S. should increase U.S. export by reducing the tariffs faced by U.S. agricultural exports on the Japanese, Canadian, Mexican and Chinese markets, the trade disputes occurring from 2018 to 2020 have decreased U.S. exports to China, Canada, Mexico, Turkey and the EU for products targeted by retaliatory measures. In addition to these trade policies directly involving the U.S., the CPTPP agreement between the countries of the Pacific area may lead to a diversion of trade, as some agricultural exporters such as Australia and New Zealand may replace the United States in its traditional destination markets. The different kind of trade restrictions some countries may apply following the Covid-19 crisis, both at the export and at the import level, may also have a significant impact on world trade values and prices, even if it is now difficult to predict their extent.The proposal has four objectives:1. To monitor agricultural trade policies implemented after the Covid-19 crisisThe Covid-19 pandemic caused a major economic crisis and disruption in world trade, mainly due to the containment policies put in place by countries around the globe. It has also led some countries to apply restrictive trade policies for medical products and for certain agricultural products such as wheat (Russia, Tajikistan) or rice (Vietnam). It is possible that this crisis will lead countries to reassess the risks of being too dependent on the international market to feed their population and generate new restrictions on agricultural trade, both in imports and exports. The first objective of this project is to monitor the different trade policies implemented in 2020 and 2021 in agriculture following the Covid-19 crisis, including implementation of the U.S.-China Phase One agreement. This data will then be added to the information we already have on trade agreements or trade disputes.2. To assess the overall effect of trade policy shifts between 2017 and 2021 on aggregate trade values The second objective of the project is to assess the overall effect of concomitant tariff reductions or increases on world trade values, with a specific focus on U.S. exports. The estimation of a gravity model at the aggregated level will allow us to see in particular whether the effect of U.S. trade disputes on total American agricultural exports is offset by the signature of trade agreements by the U.S. It will also provide an evaluation of the trade diversion occurring in the Pacific Area due to the CPTPP and to trade restrictions implemented following the Covid-19 crisis.This step will allow us to discuss the effects of the policies not only on the total amounts of U.S. exports but also on those of its main competitors on the world agricultural market, depending on the destination markets. We will also be able to see in which destination markets the effect of trade policies is most pronounced and where the United States has gained or lost the most market share and against which competitors. We will distinguish the pre-Covid-19 crisis from the post-crisis period to investigate whether the effect of trade policy changes after the Covid-19.3. To determine which products have been most impacted by these variations in market accessThe third objective of the project is to determine which products are the most impacted by the recent changes in trade policies. The effect of trade policies is expected to be heterogeneous across products. First, tariff reductions defined in trade agreements or the retaliations are different from one product to another. Second, the products are not sensitive to customs duties and trade costs in the same way. The existence of non-tariff measures, perishability of products, and different demand or supply elasticities, make markets for products react differently to a reduction or increase in tariffs. The aim will be to establish a ranking of the most impacted sectors or products. For which products was the trade agreements the most beneficial? Which products/sectors are the most impacted by diversion? In which sectors the trade dispute with China or the EU had the higher effects?To achieve this third objective, we will estimate structural gravity equations at the sectoral level, which allow us to have estimates of the effect of trade agreements or trade disputes for each sector (for example, corn, soybean, tobacco, beef...). Using the state's share in U.S. production, we will then be able to estimate the impact of trade policy shifts on Virginia's agricultural and food production and to discuss the effect of the world trade situation on Virginian farmers.4. To evaluate the effect of trade policies on trade pricesTrade policies not only affect the traded values, but can also impact prices. Depending on the structure of markets, the pass-through of customs duties may be complete (the variation in tariffs is directly reflected in the price of products) or incomplete (part of the tariff is absorbed by exporting firms). Moreover, trade policies, by changing trade quantities on the markets, also impact the level of competition on the destination markets, and thus affect prices.The last part of the project will investigate the impact of trade policies on trade prices. To be relevant, this analysis should rely on data at a very disaggregated level of the product nomenclature to avoid aggregation bias. As for the objective iii, we will proceed to an econometric estimation at the sector or product level to estimate the impact of tariff variations on prices level. We will be able to estimate these effects by exporter or by importer.
Project Methods
We will estimate the impact of trade policy shifts, both on trade values and prices, using a gravity model of trade. This is an econometric model widely used in international economics. Generally, a gravity model assumes that the values of trade between any two economies is directly proportional to the product of their economic masses and inversely proportional to the distance between them, controlling also for other important factors.Hundreds of papers have used the gravity equation to study and quantify the effects of various determinants of international trade, such as common currencies, trade agreements or geographical features. In our case, we will use this model to isolate and econometrically identify the effects of preferential agreement, trade dispute, and unilateral trade restrictions on U.S. agricultural exports by exploiting bilateral variation in trade patterns across exporting and importing countries before and after the imposition or the removal of trade restrictions.The success of the gravity model is due both to the simplicity of its application and solid theoretical foundations. Anderson (1979) was the first to provide an economic theory of gravity under the assumptions of constant elasticity of substitution (CES) and product differentiation by place of origin (Armington, 1969). The gravity model has been derived under many alternative microeconomic foundations including the Ricardian model (Eaton and Kortum 2002), the Heckscher-Ohlin model (Bergstrand 1989 and Deardorff 1998), monopolistic competition (Krugman 1980 and Bergstrand 1985) or entry of heterogeneous firms and selection into markets (Chaney 2008 and Helpmanet al. 2008). More recently, Anderson and Yotov (2012), Arkolakis et al. (2012), and Costinot and Rodríguez- Clare (2014) provide reviews of the theoretical foundations of gravity using a wider framework. Larch and Yotov (2016) offer derivations and comparison between the leading gravity frameworks on the demand and supply side and discuss methods to perform general equilibrium analysis with the structural gravity model.From the empirical side, Head and Mayer (2014) offer representative estimates and evidence for the empirical success of gravity with aggregate data. Anderson and Yotov (2010) present and discuss sectoral gravity estimate with goods trade. Anderson et al. (2015) demonstrate that gravity works very well with services sectoral data and Aichele et al. (2014) estimate sectoral gravity for agriculture, mining, manufacturing goods and services.We follow Yotov et al (2016) and use the most popular version of the empirical gravity model:lnXijkt =lnEjkt +lnYikt - lnYt + (1-σ) ln tijkt - (1-σ) Πikt - (1-σ) Pjkt +?ijktwhere Xijkt is the bilateral trade value between the exporting country i, the importing country j, the product k at time t, Ejkt is the expenditure for product k in country j at time t, Yikt the production of product k in country i at time t, lnYt is the world GDP at time t, tijkt the bilateral trade cost for product k at time t, Πikt the outward Multilateral Resistance for country i, product k at time t (which measure the average trade costs faced by the exporter i on all its markets), Pjkt the inward Multilateral Resistance for country i, product k at time t (which measure the average trade costs imposed by the importer j to all its partners) and σ the trade elasticity. We follow Head and Mayer (2014) and Fally (2015) by introducing importer-product-time fixed effect and exporter-product-time fixed effect in the equation to control for Ejkt, Pjkt, Πikt and Yikt. Note that this fixed effects also control for other markets characteristics such as Non-Tariff Measures or price levels. We also use bilateral fixed effect to control for time invariant bilateral trade costs. We follow Santos Silvia and Tenreyro (2010) and estimate our gravity model using Poisson regression by pseudo maximum likelihood.At the end, our estimated equation is the following:Xijkt=FEikt + FEjkt + FEij +α PTAijkt + β Disputeijkt + γ CPTPPijkt + µx Restrict-x-ikt + µm Restrict-m-ikt + ?ijktwhere FE are fixed effects, PTAijkt is a dummy variable equal to one whether there is a trade agreement between the U.S. and country j that defines tariff reduction or trade facilitation for the product k at time t and zero otherwise. Disputeijkt is a dummy variable equal to one if the product k from the U.S. is targeted by retaliatory measures by the importer j at time t and zero otherwise. CPTPPijkt is a dummy variable equal to one if the exporter is the U.S and the importing country j is signatory countries of the Trans-Pacific CPTPP agreement at time t. This variable will allow us to estimate how U.S. exports are affected by trade diversion due to the CPTPP. Restrict-x-ikt is a dummy variable equal to one if the exporter i applies an export restriction for product k at time t and zero otherwise. Restrict-m-ikt is a dummy variable equal to one if the importer j applies an import restriction for product k at time t and zero otherwise. These four dummies allow us to isolate and econometrically identify the effects of preferential agreement, trade dispute and trade restrictions on world agricultural and food exports. By exploiting bilateral variation in trade patterns across exporting and importing countries before and after the implementation of preferential agreements, imposition of retaliatory duties, or imposition of trade restriction, and including fixed effects to control for supply and demand shocks as well as seasonality and product level effects, we provide an assessment and ranking of the change in export values that can be attributed to trade policy shifts between 2017 and 2021. Restricting the sample to U.S. exports allows us to focus on the impact of trade policies on U.S. exports while the baseline estimation provides an assessment of the overall effect of trade policies on world trade flows. In a second step, we also interact the different variables of interest with a dummy Covid-19 equal to one after the appearance of Covid-19 in order to test whether the effect of trade agreements, disputes and restrictions on trade change after the crisis.Our analysis relies on trade data from the Global Trade Atlas (GTA) database (https://www.gtis.com/gta/) which provides monthly bilateral traded quantities and values at the 6-digit level of the Harmonized System (HS) of product codes for the recent period. The database will have to be updated during the project in order to have the latest data on world trade and assess the effect of trade agreements and trade dispute in 2020 and 2021. Trade policy shifts from 2017 to 2018 will come from the MacMaps Dataset (http://www.cepii.fr/CEPII/fr/bdd_modele/presentation.asp?id=12) while we will build a dataset of recent trade policy changes using different sources of information such as WTO website or UNCTAD Trade Analysis Information System (TRAINS) (objective 1).We will first perform the estimations by pooling all the sectors together in order to have an aggregated estimate of the effects of trade agreements, trade disputes and trade restrictions on U.S. exports (objective 2). In a second step, we will perform the same estimations separately for each sector in order to estimate the effect of trade policy at the product level (objective 3). Finally, in a third step, we will estimate the same equations with prices (trade unit value) as explained variable rather than trade value, in order to discuss the impact of trade policies on the prices of traded products (objective 4).