Increasing trade between countries and gains in income have given consumers around the world access to a richer and more diverse set of commercial plant products (i.e., foods and fibers produced by farmers). According to the economic theory of comparative advantage, countries open to trade will be able to consume more-in termsof volume and diversity-if they concentrate production on commodities that they can most cost-effectively produce, while importing goods that are expensive to produce, relative to other countries. Here, we performa global analysis of traded commercial plant products and find little evidence that increasing globalization has incentivized agricultural specialization. Instead, a country's plant production and consumption patterns are still largely determined by local evolutionary legacies of plant diversification. Because tropical countries harbor a greater diversity of lineages across the tree of life than temperate countries, tropical countries produce and consume a greater diversity of plant products than do temperate countries. In contrast, the richer and more economically advanced temperate countries have the capacity to produce and consume more plant species than the generally poorer tropical countries, yet this collection of plant species is drawn from fewer branches on the tree of life. Why have countries not increasingly specialized in plant production despite the theoretical financial incentive to do so? Potential explanations include the persistence of domestic agricultural subsidies that distort production decisions, cultural preferences for diverse local food production, and that diverse food production protects rural households in developing countries from food price shocks. Less specialized production patterns will make crop systems more resilient to zonal climatic and social perturbations,but this may come at the expense of global crop production efficiency, an important step in making the transition to a hotter and more crowded world.
Bibliographical noteFunding Information:
This work was supported by the National Socio-Environmental Synthesis Center (SESYNC) under funding received from the National Science Foundation DBI-1052875. Matthew R. Helmus was supported by the Netherlands Organisation for Scientific Research (858.14.040). Erik Nelson wishes to acknowledge the salary support he received from Lund University's Biodiversity and Ecosystem Services in a Changing Climate strategic research area during the writing of this manuscript. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.