UNDERSTANDING THE ROLE OF MARKET STRUCTURE IN ACHIEVING GLOBAL AGRICULTURAL SUSTAINABILITY: THE CASE OF SOYBEANS

The overall goal of our research program is to understand how market structure shapes the effects of policies and regulations concerned with environmental sustainability. We focus on soybean due to its importance in environmental policies in both the US and Brazil, its centrality to China's food security, its role as a driver of deforestation, and its highly concentrated structure downstream import destinations. Importantly, these destinations differ in their environmental preferences, with the EU considering banning imports of commodities produced in recently cleared lands in Brazil. Similar legislation is being considered in the US and in the United Kingdom. To achieve our goal we have the following supporting objectives:Supporting Objective 1: Upstream Brazilian Market. Estimate a Structural Model of the Soybeans Procurement Market in Brazil, with a direct link to spatially explicit land use patterns. We will build and estimate a model of spatial competition for Brazil soybeans to understand the demand for soybeans in a highly concentrated and spatially fragmented (due to transportation and other transaction costs) markets, and its effects on market outcomes, including the spatial pattern of land use. A key aspect of our structural model is that it can map from equilibrium prices and exported quantities to spatially explicit patterns of land allocation by soybean farmers. This is very important in Brazil because additional quantities of soybeans can push the agricultural frontier to areas that are currently covered with native forest, either in the Amazon or in the Cerrado, thereby raising deforestation, and unleashing unwanted environmental damage. This feature of soybean procurement markets in Brazil, motivates all the policy initiatives, from voluntary efforts to government-led efforts to support deforestation-free soybeans. We will consider two prominent counterfactuals directly addressing deforestation: a high-quality certification policy, and a minimum quality standard.Supporting Objective 2: Upstream US Market. Estimate a Structural Model of the Soybean Procurement Market in the US Midwest, with a direct link to spatially explicit land use patterns. To achieve this objective, we will use similar methods to those in the previous objective, using the data and methods described below. Key insights of this objective on its own are the characterization of spatial competition in US soybeans, which is hitherto unexplored and the ability to simulate changes in land use from relevant policies, chiefly the expansion of biodiesel.Supporting Objective 3: Downstream (international) market for soybean exports. Estimate a Structural Model of the Soybeans Procurement and Export Markets considering an oligopsony/oligopoly structure: The structural models of soybeans procurement and land use developed in supporting objectives 1 and 2 are useful to understand first-order effects of shocks to the US soybeans-for-export market (supporting objective 2) and shocks to the Brazilian soybeans-for-export market (supporting objective 1). However, since US exporters and Brazilian exporters sell to some of the same destination markets (e.g., China and the EU) and have a large share of those markets, the US and Brazilian soybeans markets are closely interlinked. As a result, shocks to one market may spillover into the other and cause significant shifts in equilibrium, affecting land use, farm surplus, and firm profits.Supporting Objective 4: Analysis of the effects of changes in policies on land use and other market outcomes. The main focus of this objective is to study the effect of likely changes in the regulatory framework on the performance of international soybean markets (measures of performance are also discussed in more detail in the Methods section). This objective captures the main comparative statics of interest in our study. To study these changes, we use the structural models developed in supporting objectives 1, 2, and 3. Our strategy is to start with a baseline equilibrium scenario, then introduce the shock and use our structural models to recompute equilibrium quantities and prices. We then recompute key indicators of market performance as a function of the new vector of equilibrium quantities and prices. A comparison of market outcomes after the shock vis-à-vis market outcomes before the shock (the baseline) will allow us to characterize changes in land use associated with the shocks, as well as how shocks shift rents along the vertical supply chain and across space.