Science Briefs

Study Assesses Fragility of Global Food System

If you were in the New York metropolitan area (the home of the NASA Goddard Institute for Space Studies) in the days after Hurricane Sandy, you might have experienced something seemingly unthinkable in 21st century America: empty food shelves at the supermarkets for days and days. New Yorkers believe that they can stand up to whatever life throws at them, yet this experience caught even the proudest natives off-guard.

Networks diagrams of wheat and rice trade

Network diagrams show the complexity of the wheat and rice trades. Line widths are proportional to the trade flow volume, while the size of the nodes is proportional to the average export degree of the node. The largest export links together account for 80% of the total trade for each network. (Image credit: Columbia University/CCSR/Michael Puma)

In developed regions around the globe, people expect and even demand plentiful food when and where they want it. To meet this demand in a place like New York (with its regional population of over 23 million people) requires a logistical marvel that is among the greatest achievements of modern society. Food arrives in New York from all corners of the globe: fruits and nuts from California, seafood from China, pasta and olive oil from Italy, vegetables from Peru, and beef from Australia. Highly interdependent and complex systems involving food production and processing, energy, and transportation make this possible.

Taking a step back, should we really be surprised by a disruption to our regional food system? More importantly, how concerned should we be about the stability of our global food system in a world marked by geopolitical, economic, and climatic uncertainties?

In 2008, the world experienced spikes in international cereal prices, with wheat and corn prices doubling over about two years and rice prices tripling over just a few months in late 2007 and early 2008. While the impacts in New York were limited to some higher food prices, food riots spread across countries in Africa (14 out of 53 countries experienced mass disturbances) and other developing regions around the world.

What led to this food crisis? Surprisingly, it was not the simple case of a production shortfall leading to a spike in food prices; there was only a small decline (0.7%) in the harvest of 2007. Instead, it was a consequence of unforeseen interactions in the world's highly interconnected global food system.

In a study recently published in Environmental Research Letters, we explored the changing characteristics of the global food system. Using annual staple food production and trade data from 1992 to 2009, we showed that the global system has increased in complexity over time. In particular, the number of wheat and rice trade connections has doubled, and trade of wheat and rice has increased by 42% and 90%, respectively. We also found that the global food system is relatively homogeneous; 85% of countries have low or marginal food self-sufficiency. These findings are quite important, as highly interconnected and homogenous systems (like the Internet and global financial systems) are thought to have increased susceptibility to system-wide, cascade-like disruptions.

Photo of drought-stressed corn

Corn shows the effect of the drought in Texas in August 2013. The state had gone through one of the worst drought's in its history, causing significant economic damage to farmers and ranchers. (Image credit: U.S. Department of Agriculture/Bob Nichols)

We next simulated the impacts of continental-scale disturbances such as weather extremes or major outbreaks of crop disease on the wheat and rice trade networks. We found greater impacts due to disruptions in European wheat and Asian rice production in the more interconnected networks. Importantly, we also found that least developed countries (e.g., Senegal and Haiti) experienced greater import losses in more highly connected networks through their increased dependence on imports for staple foods.

There are many options that could increase resilience in our global food system. First, we need to build more redundancy into the global food system, moving away from a narrow focus on the efficiency of international trade. This can be done by distributing the production of staple crops over multiple regions (as is done for wheat) and broadening the genetic diversity of staple food crops. We should also balance each country's reliance on food trade with the need for self-sufficiency in staple food production, to the extent that it is possible.

A renewed focus on national level food reserves is also important, as reserves can help to reduce volatility in food prices, for example, by improving the "confidence in trade" of policymakers who might otherwise consider trade restrictions during price spikes. Lastly, we need to move towards a network- and interaction-oriented view of the global food system. For example, network information on food production and trade can be merged with ongoing efforts like the Global Agricultural Monitoring (GEOGLAM; a NASA partner) to better understand vulnerabilities in the global food system.

In our globalized work, it is critical to recognize that major food shortages can occur anywhere in the world at any time. Famines — thought by many to be a concern in only poor, developing countries — are a worldwide danger considering that large complex interconnected systems such as the global food system are vulnerable to systemic disruptions. A renewed policy focus on both U.S. and global food security is therefore needed, especially given the current tight relationship between global food supply and demand.

Thus, the conclusions of the General Accountability Office's report Grain Reserves: A Potential U.S. Food Policy Tool remain as true today as they were when written back in 1976:

  • We cannot be certain that adverse weather shocks, similar to those in 1972 and 1974, will not occur in the future. Such shocks would tax existing food supplies and the United States would be faced with making decisions on domestic price increases and allocations of food abroad.
  • Rather than face these future decisions as crisis decisions, a grain reserve that is built during years of plenty and made available during lean years could act as a buffer against unpredictable shocks to the food system.
  • Because a food reserve would be a physical source of food, it deserves serious attention by the Congress as part of a package to meet U.S. food policy objectives.


Puma, M., S. Bose, S.Y. Chon, and B. Cook, 2015: Assessing the evolving fragility of the global food system. Environ. Res. Lett., 10, no. 2, 024007, doi:10.1088/1748-9326/10/2/024007.

U.S. General Accountability Office, 1976: Grain Reserves: A Potential U.S. Food Policy Tool, OSP-76-16.


Please address all inquiries about this research to Dr. Michael J. Puma.