Understanding the Impact of High Food Prices in Latin America

Since the late 1980s, almost all Latin American countries have adopted a series of far-reaching economic reforms, especially trade, financial, and capital account liberalization. Increased economic openness has gone hand in hand with large financial inflows—particularly in the first half of the 1990s—and has brought new sources of economic growth. As a result, economies grew, inflation declined, and there was a big surge in foreign capital inflows. Although overall growth slowed after 1995, the region has experienced strong growth in the past five years, the best sustained performance since the 1970s. With the exception of a handful of countries, this economic growth has been accompanied by relatively modest inflation.

Despite these positive results, virtually all Latin American countries share similar problems: uneven economic growth, unacceptably high poverty and malnutrition rates, and lagging agricultural growth. More than 60 percent of the region's poor live in rural areas, where slow economic growth, unequal distribution of assets, inadequate public investment and public services, and vulnerability to natural and economic shocks are major policy issues.

The 2007-08 food price crisis exacerbated these problems. Prior to the crisis, the region was considered relatively stable and capable of absorbing external shocks, thanks to its higher foreign exchange liquidity; decreased public sector and external borrowing needs; exchange rate flexibility; lower exposure to currency, interest rate, and rollover risks in public sector debt portfolios; and improved access to local-currency loans. Nevertheless, the food price crisis severely affected most of the Latin American countries in terms of inflation, especially food inflation.

The impact was greatest on net importing countries (specifically, Central America and Mexico) and on poor consumers in peri-urban and rural areas. Most Central American countries are net consumers of basic food. For example it is estimated that the share of the households that were net consumers before the crisis was 68.2 percent in El Salvador, 83.3 percent in Guatemala, 88.8 percent in Honduras, and 90.2 percent in Nicaragua. As a result, a rural household in El Salvador in 2008, for example, was able to buy only 56 percent of what it used to buy eighteen months before with the same amount of money (U.S.$0.17).¹

Before the crisis, most Latin American countries were on track to reach the Millennium Development Goal of halving the proportion of people who suffer from hunger by 2015. A significant number of countries have since had to revise their ability to accomplish this goal. Furthermore, the fear of more permanent inflationary pressures coming from food prices alerted most central bankers in the region.

This paper examines the effects of food price changes and the distributional impact within countries. It analyzes the price transmission mechanism from world markets to local markets for different types of countries to better understand the net effects, and then examines the effect of food price changes on household welfare and consumption across different types of households in both urban and rural areas. In estimating welfare effects, we account for direct or first-round effects, as well as substitution effects. For the latter, we provide estimates of own-price and cross-price demand elasticities. In this respect we depart from other cross-country studies in which substitution effects are absent.² This is an important distinction, as substitution effects are far from negligible given the size of the observed food price changes. To our knowledge, no other cross-country study incorporates rigorous estimates of substitution effects in analyzing the impact of high food prices. Chávez, del Campo, and Villarreal Paez include these estimates for Mexico, in contrast to Valero-Gil and Valero; our methodology is closely related to the former.³ Ivanic and Martin analyze the impact of high food prices on poverty rates in Nicaragua and Peru using household surveys from 2001 and 2003, respectively. ⁴ Our data are closer in time to the food crisis (2006 in both cases). This difference is also important, as production and consumption structures within countries matter for poverty and welfare impact estimates...