How hard has the food price shock hit vulnerable households?

8 January 2009 13:00 - 14:15 GMT+00
Public event


Samuel Hauenstein-Swan - Action Against Hunger


John Barrett - Head of food group, Policy and Research Division, DFID


Steve Wiggins - Research Fellow and Programme Leader - Protected Livelihoods and Agricultural Growth, ODI


To date most assessments of the impacts of the food price shocks on vulnerable households in developing countries have been simulated in models. Field evidence of impacts is only just emerging. This meeting reported the findings and experiences of Action Against Hunger from Central African Republic, Ethiopia, Kenya, Liberia, Nepal and Sierra Leone.


1.            In terms of malnutrition and hunger, what does the impact of the global food price crisis look like on the ground? To date, much of the work examining this question has been based on theoretical models.

•              Samuel Hauensteinswan’s presentation Impact on the local level of high and volatile food commodity prices on malnutrition rates presented empirical work from three country case studies

i)             Central African Republic (urban setting)

ii)            Ethiopia (two rural regions)

iii)          Sierra Leone(urban setting)

•              ACF also has reports on field studies from Liberia and Nepal.  These studies focus on the acutely malnourished—of whom there are an estimated 19 million worldwide.

2.            The Central African Republic country study was conducted in the urban context of Bangui. Staple prices increased by 20 percent—modest compared to trends in world prices. Information came from three surveys conducted January 2006, January 2007, and September 2008.

•              For global acute malnutrition (weight-for-height ratios of two standard deviations below the mean or less) and severe acute malnutrition (weight-for-height ratios of three standard deviations below the mean or less) researchers were unable to prove a statistically significant increase over the time period in question, although it looks as though they may have increased.   

•              Stunting (a low height-for-age index) showed a slow increase over the 2006-2007 period, but it actually went down from January 2007 to September 2008—this is most likely a reflection of the different time of year at which the surveys were conducted.  Difficulties in statistically associating wasting and stunting with food price increases could point to a need for other malnutrition indicators.

•              One indicator showing a clear impact was the percentage of people reporting eating two or more meals a day—which decreased quite drastically from 50 percent of households before the crisis to only 24 percent of households after. Decreasing number of meals is a common coping strategy, but one that tends to fall behind both 1) eating less preferred/ expensive/ diverse foods and 2) reducing portion sizes.

3.            The Ethiopian study was conducted in two regions—one dominated by subsistence farming and coffee cash cropping, and the other a pastoralist/agropastoralist region stressed by conflict.  Prices of the staple of low nutritional value (kocho) remained relatively stable. Certain commodities did experience price rises, thought to be due less to global drivers than local environmental conditions.

•              Anecdotal evidence from focus groups suggested people were resorting to selling more assets to compensate for declines in terms of trade between coffee and maize, oxen and maize, and most crucially, daily wages and maize.  Children of wage earners tend to be the ones who end up in ACF feeding centres.

•              At a country level it is difficult to see impacts, but at a district/village level certain increases were observed in both acute malnutrition and under 5 mortality.

4.            The case of Sierra Leone, which ranks 6th in the FAO’s assessment of national vulnerability to global food prices, was conducted in the urban context of Freetown. Rice is a very political commodity here, and focus groups demonstrated concern early on regarding the trend of global price rises.

•              In response to the price rises, a sequence of coping strategies were adopted—all pointing to depletion of assets. Quantity and diversity of diet also suffered significantly. Household level consumption of the staple (rice) decreased and large percentages of people gave up meat and dairy.  Empirical evidence suggests acute malnutrition rates (which vary drastically across different districts of the city) did not increase to the same degree as they have done in the past due to seasonal trends.  Findings emphasise prices as only one factor of livelihoods.

5.            Discussant Tim Waites emphasised the need to differentiate between shocks and (economic / livelihood) stresses in this type of analysis. He described malnutrition as a primary effect of the price shock and reminded attendees of the importance of examining secondary round shocks such as school attendance and effects of migration.  He also stressed the difficulty of separating attribution of malnutrition into different categories including higher prices, social exclusion and environmental effects.

•              He described a study on Ethiopian safety nets which estimated food price inflation having a stronger impact on rural poor than failure of rains. 

•              Waites concluded by citing: 1) The importance of context and 2) The need for analysis to inform policy responses. 

6.            Points raised in the Q&A session included:

•              Vulnerability of particular marginalised groups to insensitive aid. --Hauensteinswan responded by stressing the necessity of giving people voice. Programs such as Right to Food are going forward with the idea of making the basic human right to food a responsibility not only of national states but the international community. He explained that certainly the international community must take responsibility for their support of often controversial programs—for example biofuels subsidies. Results of high level meetings need to be translated to the country level.  Waites in response cited malnutrition rates in children in South Asia being in the 50 percent rage, whereas in Africa the proportion is around 30 percent. This he explained is about social exclusion, rights of women, etc—and contrasted to Asia’s good performance elsewhere it is striking

•              Methodological issues surrounding different income elasticities of food groups were highlighted (demand for superior goods increasing with income; demand for inferior goods decreasing with income). Hauensteinswan responded by referring to the illustration of Kocha as an inferior good in the case of Ethiopia. Shift in diet away from normal and superior goods is also part of the description of coping strategies.

•              Susanne Jaspers (ODI) wondered if there is a difficulty in getting funding for coordinated twin-track (short/long term) responses despite an understanding that increasing levels of malnutrition often have structural underlying causes. Waites responded that there is a tendency for ‘Humanitarian’ and ‘Development’ “pots of money” to be allocated differently. Aid architechture is supposed to be allowing for a “handshake” between emergency and development, but its effectiveness to date is questionable. Hauensteinswan commented that social protection is a mechanism to link humanitarian and development work. He reiterated the case for seasonality because it is a predictable phenomenon, citing the need for “predictable funds for predictable problems”.

•              Andrew Dorward (SOAS) expressed interest in the conclusion that global shifts were not as important as seasonal ones and questioned whether the timing of the price rise was of essence (an interesting question for further research is to ask how much it coincided with harvest times and if there is any Northern/Southern hemisphere difference that emerges).

•              Steve Jones (DFID) explained that in the DFID case study of impacts in Bangladesh (results available next month), to control for certain variables, the same children were checked for malnutrition at different times, and it increased by something like 17 percent.

7.            Steve Wiggins drew three broad conclusions.

i)             The moment we begin to look at this situation in detail, highly complex, very important intellectual challenges arise.

ii)            When looking at additionally hungry and poor—modelling increases are small, but 1 or 2 percent of a large number is a large number in terms of human cost.

iii)          Reminder: The large differences from village to village, and at district versus city levels highlight the very strong case for existing and flexible safety nets that can be scaled up if and when need arises. This is a clear indication that social protection issues need to be taken more seriously.