Rice, State, and Income

What fundamentally causes the various levels of per capita income across regions within the same country that share similar political institutions? We present a “rice hypothesis”: that the mix of crops grown in different regions explains the variation in economic development. Using sub-national data from China, we demonstrate that rice cultivating regions are significantly wealthier than wheat cultivating regions, controlling for latitude, geography type, natural resources, distance to ports, colonial legacies, and policies. For every 10% increase in cultivated land devoted to rice paddies in the early 1990s, per capita gross domestic product increased by an average of 1.1 yuan/person in the 2000s. Exploiting a geographic regression discontinuity design based on China’s Qin Mountains-Huai River line, we show that rice cultivation has a causal effect on per capita income. Causal mediation analysis shows that rice cultivation contributes to per capita income through strong state fiscal capacity rather than social capital. Caution should be exercised when applying the theory outside Asia. ∗Assistant professor of Government at Harvard University. Authors are in reverse alphabetical order, and both authors contributed equally to the research and writing of the article. †Assistant professor in the Division of Social Sciences at Hong Kong University of Science and Technology.