This is's Typepad Profile.
Join Typepad and start following's activity
Join Now!
Already a member? Sign In
Recent Activity
More on poverty and food prices - a response from Derek Headey I’d like to thank Will and Hassan for replying to my guest blog. Their comments are fair, but I have a few corrections. First, while they are perfectly right to say that the poverty simulations they and others conduct pose a different question – what are the impacts of rising food prices on poverty, all else equal? – the interpretation that these studies are given by international institutions and the media is that global poverty actually rose. My study contradicts that (erroneous) interpretation, but does not contradict the studies themselves. Indeed, in one of my sensitivity analyses I use a backcasting approach to decompose the estimated change in self-reported food insecurity into a food price component and an income growth component. The food price rise of 2006-2008 is estimated to raise self-reported food insecurity by 128 million, but the economic growth component reduces food insecurity by 215 million (for a net decrease in food insecurity of 87 million). The 128 million rise is broadly consistent with the figures that Will and Hassan derive in their respective papers. And of course there is an obvious caveat here – these are only estimates and they come with sizeable confidence intervals. However, the fact that economic growth seems to have such a large effect on self-reported food insecurity (especially in China and India) does raise doubts about the “all else equal” assumption that Will, Hassan and others apply to their models. I suggest that if they are using an LSMS survey for 2004 and modeling the effect of price shocks in 2008, they also shock their 2004 model with mean income growth over 2004-2008. Maybe they will then find that economic growth often provides considerable compensation for rising food prices (???). This is what Mason and Jayne found in their study of urban wage and food price movements in Kenya and Zambia (their paper is published in the journal Food Policy). Second, in the paper I don’t emphasize the 400 million number as my preferred estimate because that number is driven by a huge reduction in self-reported food insecurity China (prevalence drops by 16 points). Although one would expect a large reduction in China given its phenomenal economic growth and very limited food inflation (until recently that is), that number is too big. I therefore emphasize alternative estimates which give more reasonable numbers. For example, the backcasting approach mentioned above puts the reduction in self-reported food insecurity at 87 million. Third, yes, the Gallup surveys for 2008 nearly all cover the international peak of the crisis. Footnote 20 in my paper covers this. “In 2008 only one sampled survey was conducted before April (Indonesia, where the survey finished on March 25, when international food prices were already very high). Hence all the 2008 values for food insecurity—which are 12-month retrospective answers—cover the first half of 2008, and most cover the last few months of 2007 as well.” Of course, domestic prices could have followed international prices with a lag, so the coverage is not ideal, but it’s the best we have. Moreover, the Gallup data for 2009 would deal with the lag issue (and the onset of the financial crisis), but the 2009 data also don’t show a huge rise in self-reported food insecurity. Finally, I should re-emphasize that I certainly don’t think self-reported data are the ideal. My current work is looking at how we should optimally measure food insecurity in response to shocks, largely using the Indonesian 1997/1998 financial crisis as an example since it was a major shock with well documented effects on income, employment and food prices. Though very preliminary, that work suggests that if we wanted to measure food insecurity in real time (as opposed to simulations) we should be measuring anthropometrics (weight for height for young children, and body mass index for adults), and very indicators of diet quality (anemia and dietary diversity scores). Those measures would capture both calorie deficiencies and micronutrient deficiencies, both of which are very important for human wellbeing. The Indonesian crisis adversely affect both, especially micronutrient deficiencies (see the wonderful paper by Stephen Block et al on this: That’s where we should be heading if we are serious about gauging the real-time impacts of this type of crisis, and the WFP is indeed moving in that direction. However, they don’t disseminate their data well, and there is huge scope for coordination and collaboration between WFP, FAO, the World Bank and the Demographic Health Surveys to increase the coverage of shorter, quicker and cheaper food security surveys. is now following The Typepad Team
Jun 17, 2011