Ideas to Action:

Independent research for global prosperity


Global Health Policy Blog


I follow Richard Horton, the editor of The Lancet, on Twitter (@richardhorton1). I was concerned to see his post yesterday: “China faces a food crisis that threatens child survival. A UN FAO special alert reports wheat/water shortages. The worst in 60 years.”

I navigated to the FAO alert.  Winter wheat crop yields may be affected by an on-going drought and there are already reported shortages of drinking water.  FAO also reports that the national average retail price of wheat flour in China has grown 16% since January 2010, suggesting that the latest price hike is anticipating the drought’s possible effect on the June 2011 wheat harvest.

With this news, food security on the G-20 agenda, and panic in the air, what do we really know about the effect of food prices on health and nutrition?  What can we say about policy responses? Here are 10 points that I wish the media and donors would consider when analyzing food price hikes and their impact on people:

  1. The poverty and nutrition impact of a food price increase depends greatly on the composition of diets among the poor and near-poor. For example, an increase in the price of wheat in China in 2011 may not be as important as the increase in the price of rice may have been in 2008, in spite of the noodle argument made by WSJ today.
  2. The poor devote a larger share of their already small consumption to purchasing food. If food prices increase, you might expect to see substitution away from more nutritious, more expensive foods towards less expensive, less nutritious foods. These effects have been documented in the literature on consumption shocks in general; however, the literature on the impact of price increases on nutrition is thin. Skoufias and co-authors have recently looked at the income elasticity of micronutrients in Indonesia and how they change during a shock, finding that for some micronutrients such as iron, calcium, and vitamin B1, elasticities are significantly higher in the crisis year compared with a normal year. However, this was not the case for Vitamin C, suggesting that micronutrient supplementation is just as important. You might also observe reductions in other kinds of human capital-related spending –on schooling, health care, etc., but less evidence there.
  3. Pre-price hikes, nutritional status among the poor is already problematic. In China, for example, UNICEF reports that 40% of children in rural counties are stunted.  The prevalence rates of iron deficiency anemia in China’s poorest rural areas range between 25% and 60% (Chen at al 2005; Wang 2007). Both stunting and anemia affect cognitive development, school performance and, in some studies, later labor market earnings.  Most developing countries look similar or worse. This argues for a permanent nutritional safety net effort, not a food price specific response.
  4. If you care about nutritional impact, the important population to reach is poor or near-poor pregnant and lactating women and children under three years old. Not the median voter, not affected farmers, and not the population in general.
  5. If you care about poverty impact, the important population is still the poor and near-poor. Not necessarily food producers who may benefit from price increases.  There is still no empirical literature on the poverty impact of 2008 food price crisis; we only have simulations of the poverty impact based on household surveys and there has been a lot of noise around these measurements related to oil prices and recession effects.
  6. To respond to political pressures, countries have continued to resort to tax cuts, import tariffs and other measures that do not mainly benefit the poor or the nutritionally-affected. Peru, for example, just announced a one-point reduction in the VAT equivalent to 0.5% of GDP, a measure that will benefit the entire population in a context where spending on health is only 2.5% of GDP, 35% of children under 5 are stunted and 50% are anemic (data here).
  7. Many donors continue to tout agricultural innovation programs –like Feed the Future–as the solution to malnutrition. Yet, there is limited empirical evidence to connect these programs to nutrition and poverty impact. In a welcome move, the International Food Policy Research Institute (IFPRI) has launched an agriculture and health research program, noting that many countries with good economic and agricultural growth don’t actually see reductions in malnutrition (for example, India). As a result, for the moment, there is good reason to tone down the donor rhetoric on the use of Feed the Future and related as a response to malnutrition and poverty. [Though the argument that increasing agriculture productivity should help to eventually lower prices is taken.] Instead, focus on #9 below.
  8. Food-based safety net programs continue to be common (for example here), in spite of their high administrative costs, nutritional limitations (frequently inappropriate products), and poor targeting (to school aged children generally, in urban areas). See the World Bank nutrition tool kit for more examples, or almost any national safety net assessment.  Although scaling up cash and vouchers, the World Food Program (WFP) is still obliged by its donors to sponsor in-kind programs (2.7 million tons in direct transfers in 2009), as if food availability instead of food prices was always the problem. Instead focus on #9.
  9. Permanent conditional and other cash transfer programs (CCT) plus micronutrient supplementation –targeted to poor women and young children–seems a first best option for mitigating both the poverty and nutritional effects of food price increases and shocks in general. CCT also help improve health (see my paper here). The Lancet names the CCT plus supplementation strategy as “what works” in their maternal and child undernutrition series. Policy recommendations and donor support should be directed to these programs, which are among the most extensively evaluated in the development enterprise. If a CCT already exists, benefits can be indexed to food inflation at more regular intervals. If coverage is partial, it can be expanded the entire eligible population.  With the boom in mobile telephony and banking in Africa as safe and relatively fraud-free ways to transfer monies, it is time to move away from in-kind support (see my colleague Alan Gelb’s podcast).  The current environment is perfect for rationalizing the untargeted, poorly performing, in-kind safety nets and royalties schemes in favor of CCT and the like. A concern could be the predominantly rural focus of existing CCT in the context of affected urban populations; another modification could consider a temporary transfer or food stamp scheme in urban areas. Nothing is perfect, but CCT and their ilk are better than alternatives.
  10. Lots of money has been plowed into measuring food prices. I count at least five initiatives between WFP, FAO, EU, IFPRI and USAID.  All the media reporting focuses on prices. Yet relatively little money has gone to monitoring nutritional status and health facility data in real-time, which would help us to determine whether families are really suffering in China or not.

I know this is a political economy problem.  Least-best policy responses occur “when we run the risk of food riots in the poorest countries” (Sarkozy).  There is a lot of uncertainty. But the CCT and similar ideas are both sellable and effective as responses, and have lots of nice spillover effects (for a future blog…).

Related Topics:


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.