Food prices: The need for insurance

Esther Duflo

25 April 2008



Throughout last week, violent riots in Haiti – provoked by Haitians’ fury over the increased price of basic foodstuffs – brought the issue of agricultural prices to the forefront. Other incidents occurred in Indonesia, Guinea, Mauritania, Mexico, Morocco, Senegal, Uzbekistan and Yemen. Several large rice producers (e.g. Vietnam, India, Egypt) put severe limits on rice exports.

After being stable for several decades, foodstuff prices began to rise again in 2005, and in 2007, their increase was phenomenal. From March 2007 to March 2008, the average world price for corn increased 30%; for rice, it increased 74%; for soybeans, 87%; and for wheat, 130% (the BBC has a nice summary of what is happening to food prices).

High prices

Several reasons explain the upward trend in prices, including the demand for biofuels (which consume a significant part of the corn produced worldwide), and the growth and enrichment of the world population (particularly the increased demand for meat in China – paradoxically, it takes more grain to produce a calorie in meat form than it does to produce a calorie in grain form).

Several short-term factors also help to explain the recent price peak. Because the main consumers of rice are also producers, the volume of rice traded is low compared to the volume of rice produced (only seven percent of produced rice is traded). Restrictions by big producers (such as India) can thus have large impacts on the world price of rice, since they affect a large proportion of the volume of rice traded. It keeps the prices in India relatively insulated from what is happening on the world market, however. Wheat harvests have been poor in several major producing countries, rice is suffering from a mysterious parasite in Vietnam, and following accusation of corruption and mismanagement in the last few years, there have been sharp declines in grain stocks maintained by governments to stabilise prices (they are at their lowest levels since 1984). As a result, prices are not only high in general, but also more volatile (another drop in rice prices is expected after the harvests in Indonesia and India); even the financial crisis plays a role: in the current meltdown, commodities offer an interesting betting opportunity.

Winners and losers

Robert Zoellick (President of the World Bank), Jacques Diouf (the chairman of the UN Food and Agriculture Organization), and many others, have now rung the alarm bells. Zoellick even waved a loaf of bread at the annual meetings of the IMF and the World Bank to drive the point home. According the analysis of the World Bank Living Standard Surveys (which Abhijit Banerjee and I have exploited in our article “The Economic Lives of the Poor”1 ), food accounts for between 50% and 77% of the budget for a family that lives on less than a dollar a day per capita (the world poverty line), depending on the country. This leaves these families very sensitive to food prices, and it is no surprise that the recent increase can make a serious dent in their budget.

Yet, two or three years ago, rich countries’ farm subsidies, and even food aid, were strongly criticised: by keeping prices artificially low, the argument went, they prevented African farmers from selling their products at a good price, keeping them in poverty.

These two arguments may seem contradictory at first. Unfortunately, they are not. An increase in food prices benefits net producers (those who produce more than they consume), to the detriment of net consumers. This is true at both the national and individual levels. Thus, at the national level, the rise in grain prices will improve the trade balance of exporting countries and worsen that of importing countries, including those of Sub-Saharan Africa. At the individual level, the poor who are most affected are in urban areas, but even in rural areas, a number of the poorest people are in fact net consumers of grain as well. A 1989 study of Thailand by Angus Deaton2 showed that, on average, rural households benefited from an increase in the price of rice, but with large variations from one household to another. Those households who benefited most were neither too poor nor too rich.

So when grain prices rise, some of the poor gain, and some lose – in the short-term. In part, the uproar over food prices reflects a fundamental political economy reality: when some gain and some lose, the voice of those who lose is always louder. This is particularly true of an increase in food prices, which hurt primarily the urban poor. In the medium-term, however, an increase in price volatility is damaging for everyone. Poor families in developing countries are already facing enormous risks (they are often self-employed, they are subject to the uncertainties of weather, and their health is fragile), and there is very little insurance against such risks, apart from their own savings or informal solidarity. Furthermore, these risks are more serious for households struggling to provide the bare minimum. A setback might mean sacrificing the children’s education, or not being able to save a little girl from a diarrhea attack (on this, see a beautiful paper by Elaina Rose, which showed that during drought, the relative survival probability of girls dramatically declines).3 A passing difficulty can leave a permanent scar.


Beyond emergency relief, which the world community is scrambling to organise at the moment, it is essential to establish effective insurance against variability in food prices for the poorest. Many argue now that the way to do it is to stop trade and encourage countries be “self sufficient” in food. This seems of course a third best solution, since the countries would then be entirely dependent on the vagaries of the weather at home. It is also not clear what countries that are net food importers would do (in the long run, the argument goes, they will improve their productivity… however, given Senegal’s climate, producing rice there will always be much more expansive than doing it in Thailand). The traditional method used by developing country governments – maintaining large stockpiles of grain by buying when the price is low and selling when the price is high – has its share of problems. In India, it was said that at some point that there were enough bags of rice in those storage facilities to go to the moon and back. The losses in storage and to corruption were important. Alternatively, the governments can manipulate prices using taxes and subsidies. Or perhaps it is time to be creative and make the international financial services actually work for the poor: governments could provide price insurance for the poor (in the form of transfers to some when price are high, and others when price are low). Countries that are neither net sellers nor net buyers could do this internally, and countries that are either net sellers or net buyers should be able to sell this insurance on the world market. There is nothing straightforward about these solutions. The key point, however, is that it is urgent to think of something.


1 Journal of Economic Perspectives 21(1): 141 – 167, Winter 2007

2 Angus Deaton (1989). "Rice Prices and Income Distribution in Thailand: A Non-parametric Analysis" Economic Journal, 99(395): 1 – 37.

3 Elaina Rose (1999). "Consumption Smoothing and Excess Female Mortality in Rural India," Review of Economics and Statistics, 81(1): 41-49, February.



Topics:  Development Poverty and income inequality

Tags:  insurance, food prices, poor people, agricultural producers

Professor of Economics at MIT and a CEPR Programme Director