Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Agricultural Economics

Committee Chair

Thomas W. Hertel

Committee Co-Chair

Frazad Taheripour

Committee Member 1

Russel H. Hillberry

Committee Member 2

Wallace E. Tyner

Committee Member 3

Jianguo Liu


China’s soybean demand boom in the past two decades has been very dramatic. It involves socioeconomic and environmental interactions of multi-coupled systems. Over this period, China doubled its GDP, and the ensuing income growth generated strong growth in the demand for livestock products -- a major consumer of soybean meals. In addition, the goal of producing more meat and milk boosted protein content requirements in feed formulations and intensified China’s soybean meal demands. Brazil and Argentina stepped in to satisfy this increased demand. In the case of Brazil, rapid technical change, coupled with the expansion of cultivated area, played a key role in meeting the increased soybean demand in China’s global soybean boom. In 2011, Brazil became the largest soybean supplier for China, and soon in 2013, it overtook the US as the leading global soybean exporter. Soybean trade offers a notable instance of the emerging “telecoupling” concept – China, Brazil and the US closely interact with each other across distances. Chapter 2 aims to bridge agricultural trade with this telecoupling concept. The goal of Chapter 2 is to understand the historical soybean boom by focusing on the supply-demand-trade nexus of these three countries with a modified version of the GTAP-BIO model. We decompose historical changes into five groups of socio-economic drivers – macroeconomic growth, soybean productivity, other crop productivity, government policies, and pasture and forestry changes – quantifying each driver’s contributions to soybean trade, production, and land use changes over 2004-2011.We find that China’s macroeconomic growth boosted soybean production and exports from Brazil and the US, whereas macroeconomic growth in the latter two regions actually dampened soybean exports over the 2004-2011 period under examination. Brazil’s strong soybean productivity growth over this period, allowed that country to become dominant in the global soybean market. It also had strong spillover effects, displacing the US in the Chinese market and reducing overall growth in soybean output in the US. This strong soybean productivity growth also contributed to cropland expansion in Brazil. We introduce Genetically Modified (GM) and non-GM soybeans into our modified version of the GTAP-BIO model, which requires new trade elasticity estimates, especially the elasticity between GM and non-GM soybeans. However, such estimates are missing from the existing literature, and current trade data does not distinguish GM and non-GM varieties. In this dissertation, we treat soybeans from countries that predominantly export GM and non-GM varieties as GM and non-GM soybean bundles. In Chapter 3, we apply a structural gravity model to estimate three parameters: elasticities of substitution across GM and non-GM soybean bundles, respectively, and substitution between nested constant elasticity of substitution (CES) bundles of GM and non-GM soybeans. Following the Armington assumption, we employ a single nest CES structure for the elasticities of substitutions among each soybean bundle and a nested CES structure for the elasticity of substitution between GM and non-GM soybean bundles by using Poisson Pseudo Maximum Likelihood (PPML) estimators. Our estimates show that the elasticity among GM soybean bundles is as high as 29, indicating GM soybeans are homogeneous productions. The elasticity among non-GM soybean bundles is lower at 12. Although varieties of non-GM soybean bundles are substitutable, their qualities are differentiated by its origins. Low substitutability between GM and non-GM soybeans at 0.4 implies that GM and non-GM soybean bundles are viewed as poor substitutes by countries. By applying the historically-validated and well-tuned GTAP-BIO model from Chapter 2 and the trade elasticities estimated from Chapter 3, we aim to understand soybean boom from the supply side and investigate how the US lost its lead in the global soybean trade. We decompose changes of two main indices – the US/Brazil soybean production ratio and the US/Brazil soybean exports to China ratio – into a more detailed specification of socio-economic drivers. By pinpointing negative and positive drivers, we shed light on the factors driving to the US “losses” and “gains” in soybean exporting to China over 2004-2011 and provide insights on future soybean trade patterns.