Comments on v0 of the Report on Price Volatility

As per my post yesterday on how the High-Level Panel of Experts (HLPE) appointed by the Committee on World Food Security is seeking comments on “version zero” of its Report on Price Volatility, my coauthor Chris Barrett and I sent the following comments. Remember that if you are interested in commenting, you have until June 3 to do so.


Having both written, read, and taught extensively on food policy as it relates to developing countries, we read version zero (v0) of the Report on Price Volatility prepared by the project team for the High-Level Panel of Experts (HLPE) on Food Security and Nutrition with great interest and  offer some reactions for the HLPE’s consideration.


The strength of v0 is that not only does it address the possible causes — excessive food price volatility, cyclical food crises, and emerging scarcities — and presumed consequences of the food crises of 2008 and of 2010-2011, it also considers several policy options at both the international and national levels in view of preventing such food crises in the future. We applaud this focus on policy-oriented inquiry.


The major weakness of v0 is that it reinforces an error of conflating ‘price volatility’ with ‘food price rises’, an error that pervades much popular discussion of the global food crises and which subtly encourages policymakers to tackle the wrong problem. Let us explain.

Defining food price volatility as the fluctuations (specifically, the moving average standard deviation) around a given food price level and using the Food and Agriculture Organization (FAO) of the United Nations’ food price index, we show in a recent paper (Barrett and Bellemare, 2011) that increases in the food price level are statistically significantly associated with higher levels of political instability while increases in food price volatility are actually associated with lower levels of political instability.

These findings at the worldwide, macro level actually echo those at the micro level. Indeed, in another paper (Bellemare et al., 2011), we show that increases in food price volatility, although detrimental in aggregate to rural households in Ethiopia, may actually slightly benefit the poorest households.  It is the wealthiest households in rural Ethiopia who suffer from food price volatility because they are net food sellers who have to invest in land, seed and other inputs long before they know what price their crop sales will fetch.  For poor buyers, high food price levels are harmful; but price volatility, defined as variation around the level, is not intrinsically bad.  It can even offer some opportunities for timing purchases and substituting among commodities to reduce food costs.

This discrepancy between our findings and the beliefs commonly held by policy makers and portrayed by the media arises from a subtle but crucial distinction. We define food price volatility in its precise, statistical sense (i.e., as a measure of the movements around a given level, or as the variance or standard deviation of food prices) whereas they seem to define volatility in another sense (i.e., as large increases in the level, or as increases in the food price level). Is this all just semantics? Of course not. Words have distinct meanings, and the right policy responses require one to think clearly using the right concepts.

Lastly, even if we were to assume for the sake of argument that food price volatility is an issue, a forthcoming paper by Jacks et al. shows that although commodity prices have always been more volatile than the prices of manufactured goods, commodity price volatility has not increased significantly over the past 300 years.


Given the foregoing, our view is that v0 focuses on the wrong issue by using the wrong definition of food price volatility. Our work at both the micro and macro levels shows that the real issue is rising food prices, and not food price volatility.

Indeed, as a result of the hardship they induce, rising food prices often sow discontent in developing countries. The 2008 food price spike helped bring down governments in Haiti and Madagascar. Recent political unrest in North Africa and the Middle East coincides strikingly with this most recent period of rising food prices. In Tunisia, protesters in the initial demonstrations in December 2010 brandished loaves of bread as they launched political unrest that toppled a dictatorial regime that had been in place for almost 25 years.

By conflating high food price levels and high price volatility, policy makers and members of the press make three important errors. First, while it is clear that food price levels are at historic highs, food price variability, although high these past few years, is not out of line with historical experience and is generally lower than in the 1970s.

Second, the effects on the well-being of the poor of price levels and of price volatility differ. Rising food price levels hurt food consumers by reducing their purchasing power while benefitting food producers by increasing farm profits. By contrast, food price volatility hurts net seller food producers, who make irreversible investments in crop inputs at the start of the growing season and routinely reduce such investments as food price risk increases. Perhaps not coincidentally, these same large farmers enjoy tremendous taxpayer-funded support programs from G-20 governments presently expressing concern about food price volatility.

Third, volatile food prices — as distinct from high price levels — do not necessarily harm the poor and are not associated with increased political unrest. Indeed, our statistical results suggest the opposite.

As the HLPE pursues a policy-oriented investigation of the causes and consequences of recent global food crises and of prospective policy preventatives for further crises, we strongly urge clarity in the definition of key terms and a tight focus on the phenomena of concern — historically high food prices — rather than on food price volatility that is neither high in historical terms, nor necessarily problematic for the poor nor for political stability.