Trade and Agriculture

A conference on Trade and Agriculture took place on May 17-18 in Cambridge. Research Associate Dave Donaldson of MIT organized the meeting, which was sponsored by the Economic Research Service at the U.S. Department of Agriculture and the Giannini Foundation at the University of California. These researchers' papers were presented and discussed:

Jayson Beckman, Department of Agriculture; Carmen Estrades and Manuel Flores, Universidad de la República (Uruguay); and Angel H. Aguiar, Purdue University

The Impacts of Export Taxes on Agricultural Trade (NBER Working Paper No. 24894)

Most major types of trade barriers on agricultural products have been reduced since the formation of the World Trade Organization. Globally, tariffs have decreased significantly, export subsidies are scheduled to be eliminated, and the Trade Facilitation Agreement holds some promise on addressing non-tariff measures. However, export restrictions -- mainly export taxes -- have remained prolific and in fact, their number has increased over the last ten years. Perhaps because they are only used by a subset of countries or on some commodities, export taxes have not received the same scrutiny in multilateral trade negotiations as other trade barriers. This is despite the fact that export taxes often occur when food prices are high and/or volatile. Where export taxes have attention in the literature tends to be in studies that examine only a single commodity or country. Beckman, Estrades, Flores, and Aguiar seek to provide more detail to investigating the linkages between export taxes, trade, and food prices. To do so, they utilize both a partial equilibrium model using a econometric-based gravity framework and a global general equilibrium model, which, in tandem, capture different aspects of these linkages. The results show that export taxes do not have a widespread impact on international prices, but rather that the impact is concentrated in few goods. With the researchers' partial equilibrium model, they observe a positive impact on prices of dairy products, live plants, vegetables, oilseeds and oils. In contrast, with their computable general equilibrium model, the positive impact on prices is verified on wheat, coarse grains, and cattle and beef. A removal of export taxes currently in place would not have a significant impact on global prices, and would benefit regions currently applying taxes by an increase in production and exports and a reduction in poverty. However, some regions, mainly those that currently export commodities taxed in other countries, could be harmed by the removal of export taxes due to the increased competition of exports in international markets, although they could benefit from a fall in domestic prices. These, global results highlight the need to consider the general equilibrium effects of the removal of export taxes.

Douglas Gollin, University of Oxford, and Casper W. Hansen and Asger Wingender, University of Copenhagen

Two Blades of Grass: The Impact of the Green Revolution (NBER Working Paper No. 24744)

Gollin, Hansen, and Wingender examine the economic impact of high-yielding crop varieties (HYVs) in developing countries 1960-2000. They use time variation in the development and diffusion of HYVs of 10 major crops, spatial variation in agro-climatically suitability for growing them, and a differences-in-differences strategy to identify the causal effects of adoption. In a sample of 84 counties, the researchers estimate that a 10 percentage points increase in HYV adoption increases GDP per capita by about 15 percent. This effect is fully accounted for by the direct effect on crop yields, factor adjustment, and structural transformation. They also find that HYV adoption reduced both fertility and mortality.

Christophe Gouel, INRA-AgroParisTech, and David Laborde, IFPRI

The Crucial Role of International Trade in Adaptation to Climate Change

Climate change effects on agricultural yields will be uneven over the world with a few countries, mostly in high latitudes, that may experience gains, while most will see average yield decrease. Gouel and Laborde aim to quantify the role of international trade in attenuating the effects of climate change by allowing the expression of the new climate-induced pattern of comparative advantages. To do this, they develop a new quantitative general equilibrium trade model where the representation of acreage and land use choices is inspired from modern Ricardian trade models but also consistent with theoretical and empirical literature on land use choices. The model is calibrated on spatially explicit information about potential yields before and after climate change coming from the agronomic literature. The results show that, because demand for food is quite inelastic, the climate-induced yield changes generate large price movements that incentivize adjustments in acreage and trade. The new trade pattern is very different from the current one showing the important role of trade flows in adapting to climate change. This is confirmed by large increased welfare losses from climate change when adjustments in trade flows are constrained.

Jonathan I. Dingel, University of Chicago and NBER; Solomon M. Hsiang, University of California, Berkeley and NBER; and Kyle C. Meng, University of California, Santa Barbara and NBER

The Spatial Structure of Productivity, Trade, and Inequality: Evidence from the Global Climate

Dingel, Hsiang, and Meng show that welfare inequality in a trading network is greater when productivities are rearranged such that neighboring locations are more similar. An increase in the spatial correlation of productivities amplifies cross-country welfare dispersion by increasing the correlation between productivity and the gains from trade. To empirically examine this prediction, they study how global agricultural trade responds to exogenous changes in the spatial correlation of agricultural productivity driven by a naturally occurring global climatic phenomenon. As predicted, higher spatial correlation in cereal yields increases the correlation between productivity and the gains from trade. In a forecasting application, climate-change projections for 2099 that incorporate this general-equilibrium effect exhibit substantially greater global welfare inequality, with higher welfare losses in most African, South American, and many Asian countries.

Osea Giuntella, University of Pittsburgh; Matthias Rieger, Erasmus University; and Lorenzo Rotunno, Aix-Marseille University

Weight Gains from Trade in Foods: Evidence from Mexico (NBER Working Paper No. 24942)

Giuntella, Rieger, and Rotunno investigate the effects of international trade in food on obesity in Mexico. They classify Mexican food imports from the U.S. into healthy and unhealthy and match these with anthropometric and food expenditure survey data. The researchers find that exposure to imports of unhealthy foods significantly contributes to the rise of obesity in Mexico. The empirical evidence also suggests that unhealthy food imports may widen health disparities between education groups. By linking trade flows to food expenditure and obesity, the research sheds light on an important channel through which globalisation may affect health.

Marshall Burke, Stanford University and NBER; Lauren F. Bergquist, Becker Friedman Institute; and Edward Miguel, University of California, Berkeley and NBER

Sell Low and Buy High: Arbitrage and Local Price Effects in Kenyan Markets (NBER Working Paper No. 24476)

Large and regular seasonal price fluctuations in local grain markets appear to offer African farmers substantial inter-temporal arbitrage opportunities, but these opportunities remain largely unexploited: small-scale farmers are commonly observed to "sell low and buy high" rather than the reverse. In a field experiment in Kenya, Burke, Bergquist, and Miguel show that credit market imperfections limit farmers' abilities to move grain inter-temporally. Providing timely access to credit allows farmers to purchase at lower prices and sell at higher prices, increasing farm profits and generating a return on investment of 28%. To understand general equilibrium effects of these changes in behavior, the researchers vary the density of loan offers across locations. They document significant effects of the credit intervention on seasonal price fluctuations in local grain markets, and show that these GE effects greatly affect individual level profitability estimates. In contrast to existing experimental work, the researchers' results thus indicate a setting in which microcredit can improve firm profitability, and suggest that GE effects can substantially shape estimates of microcredit's effectiveness. Failure to consider these GE effects could lead to substantial misestimation of the social welfare benefits of microcredit interventions.

Thibault Fally, University of California, Berkeley and NBER, and James E. Sayre, University of California, Berkeley

Commodity Trade Matters (NBER Working Paper No. 24965)

Primary commodities account for approximately 16 percent of world trade, yet they are used extensively as intermediate inputs into many production processes. Fally and Sayre show that ignoring several key features of trade in commodities leads to a large understatement of aggregate gains from trade despite their relatively small share of world trade. They quantify the welfare gains from international trade when we account for specific characteristics of most primary commodities: i) a low price elasticity of demand as a result of difficulty in finding substitutes, ii) a low price elasticity of supply, and iii) a high concentration of natural resources and production among a few countries. For instance, copper is difficult to replace in the electronic equipment industry, the supply and demand for copper vary only slightly with changes in prices, a large share of its supply comes from Chile and copper accounts for half of Chilean total export revenues. The researchers explicitly account for these features in a general-equilibrium model of consumption, production, and input-output linkages. In their simulations, they confirm that ignoring these specific features of commodities leads to a wide understatement of the aggregate gains from trade.

Uris Baldos and Thomas Hertel, Purdue University, and Frances Moore, University of California, Davis

The Biophysical and Economic Geographies of Global Climate Impacts on Agriculture (NBER Working Paper No. 24779)

Baldos, Hertel, and Moore explore the interplay between the biophysical and economic geographies of climate change impacts on agriculture. It does so by bridging the extensive literature on climate impacts on yields and physical productivity in global crop production, with the literature on the economic geography of climate change impacts. Unlike previous work in this area, instead of using a specific crop model or set of models, the researchers instead employ a statistical meta-analysis which encompasses all studies available to the IPCC-AR5 report. This comprehensive approach to the assessment of the biophysical impacts of climate change has the added advantage of permitting the isolation of specific elements of the biophysical geography of climate impacts, such as the role of initial temperature, and differential patterns of warming across the globe. The researchers combine these climate impact estimates with the GTAP model of global trade in order to estimate the national welfare changes which are decomposed into three components: the direct (biophysical impact) contribution to welfare, the terms of trade effect, and the allocative efficiency effect. They find that the terms of trade interact in a significant way with the biophysical geography of climate impacts. Specifically, when they remove the biophysical geography, the terms of trade impacts are greatly diminished. And when the researchers allow the biophysical impacts to vary across the empirically-estimated uncertainty range, taken from the meta-analysis, they find that the welfare consequences are highly asymmetric, with much larger losses at the low end of the yield distribution than gains at the high end. Furthermore, by drawing on the estimated statistical distribution of trade elasticities, they are also able to explore the interplay between economic and biophysical uncertainties. Here, the researchers find that regional welfare is most sensitive to extremely adverse yield outcomes in the presence of uncertainty in trade elasticities.

Colin A. Carter, University of California, Davis, and Sandro Steinbach, ETH Zurich

Trade Diversion and the Initiation Effect: A Case Study of U.S. Trade Remedies in Agriculture (NBER Working Paper No. 24745)

Carter and Steinbach estimate the impact of U.S. trade remedy (TR) actions on agricultural trade from 1990 to 2014. Most previous studies of the effects of TR actions have left out agricultural products. The researchers use a four-country oligopolistic trade model to study the impact of TR duties on imports from nonnamed countries, and improve on methodological issues present in earlier studies. The researchers' empirical results show that TR investigations benefit non-named foreign exporters and U.S. imports from non-named countries increase even before the implementation of a TR duty. The extent of trade diversion is positively related to the size of the duty. Moreover, they find evidence for an initiation effect revealed by a significant increase in imports from non-named countries that did not previously trade the relevant product with the United States. The considerable extent of trade diversion in agriculture provides robust evidence for leakage effects of TR laws which have a detrimental impact on their enforceability.

Kari Heerman, Department of Agriculture, and Ian M. Sheldon, The Ohio State University

Gravity and Comparative Advantage: Estimation of Trade Elasticities for the Agricultural Sector (NBER Working Paper No. 24772)

Heerman and Sheldon present a structural gravity model which features intra-sector heterogeneity in agricultural productivity systematically linked to land and climate characteristics. The "systematic heterogeneity" (SH) gravity model predicts that countries with similar land and climate characteristics will tend to specialize in the same agricultural products. Agricultural trade flow elasticities then depend on comparative advantage, with larger-magnitude trade flow responses predicted among countries more likely to specialize in similar agricultural products and thus compete head-to-head in foreign markets. This is in contrast to standard log-linear gravity models, which impose a restrictive pattern of trade flow elasticities that depend only on absolute advantage in the agricultural sector. The researchers also show how the SH gravity model can accommodate product-specific trade costs. This allows the model to analyze changes in the dispersion of trade costs across products. Such analysis cannot be carried out in a standard gravity model, in which trade costs are assumed constant. The results confirm economically and statistically significant heterogeneity in the effects of the variables that typically proxy for trade costs in gravity models.

Shilpa Aggarwal, Indian School of Business; Brian J. Giera, Amazon Research; Dahyeon Jeong and Alan Spearot, University of California, Santa Cruz; and Jonathan Robinson, University of California at Santa Cruz and NBER

Market Access, Trade Costs, and Technology Adoption: Evidence from Northern Tanzania

While it is widely believed that poor access to markets limits agricultural productivity in rural Africa, quantifying the magnitude of this effect is complicated by a lack of granular data on input and output prices and on costs of accessing local markets. Aggarwal, Giera, Jeong, Robinson, and Spearot collect precise data on travel costs and do detailed surveys with farmers and agro-retailers in each of the 570 villages of the Kilimanjaro region of Northern Tanzania. They document substantial spatial price dispersion for fertilizer: going from the 10th to 90th percentile of the travel cost-adjusted price distribution increases prices by 40-50% of the mean. They find that an additional log point of remoteness (measured as travel time from the regional hub) is associated with 50% lower market participation and input adoption. The researchers develop a spatial model of input adoption and estimate that farmers behave as if they face travel costs of 3-4% ad-valorem per kilometer of travel (equivalent to a total of approximately 30% for the average purchase), which is approximately 3 times measured pecuniary travel costs and suggests the presence of meaningful search costs. Counterfactuals suggest sizable effects of reducing the costs of accessing both input and output markets on adoption levels, however, improving output access does not affect the remoteness gradient, while improving input access does. Holding exogenous local factors fixed, they estimate that reducing travel costs by 50% (approximately the effect of paying rural roads) would reduce the adoption-remoteness gradient by 45%.

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us