World - Is it demand or costs that are driving food inflation?

Raghav Gaiha & Varsha Kulkarni

The tyranny of old ideas is often harder to overcome than developing new ideas, lamented John Maynard Keynes in The General Theory of Employment, Interest and Money (1935). The Asian Development Bank’s Asian Development Outlook 2008 Update offers a rich and insightful analysis of the growth and inflation experience in nine Asian countries of which India is one. In its analysis of the surge in food prices, however, it falls prey to a diagnosis of underlying factors that is not just simplistic but also misleading. In fact, in a subsequent chapter in part 2 of this report, this analysis is abandoned and the focus is on the dynamics of food and oil prices and a different and more nuanced policy stance.
The Outlook 2008 projects inflation in Asia to rise to 7.8 per cent in 2008, up sharply from 4.3 per cent in 2007 and 3.3 per cent in 2006. The benign paradigm of strong growth and subdued inflation, it emphasises, has ended.
The key question addressed is: How did it happen? The Outlook 2008 begins by rejecting the view (referred to as ‘popular misconception’) that this region is suffering from cost-push inflation. The sheer rapidity of the rise in oil and food prices and hence input costs impart credibility to this view. If this view is validated, a tighter monetary policy is unlikely to curb food inflation. By contrast, if the surge in food prices is largely homegrown and due to excess aggregate demand and inflationary expectations, a tight monetary policy would be effective. In fact, loose monetary policies pursued throughout Asia (including India and China) were the main culprits in fuelling unsustainable growth of aggregate demand. That these policy prescriptions follow from standard macro-economic theory is not in dispute or controversial. As argued below, what is indeed controversial is the econometric analysis performed and the conclusions drawn from it.
Let us first review the empirical evidence. The first important result is that excess aggregate demand and inflationary expectations account for much of the consumer price inflation in the nine Asian countries studied —about 60 per cent. The second is that external cost-push factors — specifically, (global) food and oil prices — are more important in explaining producer price inflation. These factors explain about 50 per cent of the producer price inflation (for example, China and Korea). In countries where foreign exchange rates are stable (Malaysia), oil prices account for 50 per cent of the inflation. In others with not-so-stable exchange rates (Indonesia), the exchange rate accounts for almost 40 per cent of producer price inflation. India’s is a case apart as more than 50 per cent of producer (wholesale) price inflation is due to excess aggregate demand and inflationary expectations, while external shocks (global oil and food prices) account for about 25 per cent .
But this analysis is fraught with serious measurement and interpretational problems. Excess aggregate demand is measured as the gap between actual GDP and estimated GDP obtained from a trend. This is referred to as output gap. In China and India the ratio of actual GDP to estimated GDP rose from 0.98 in 2005 to 1.02 in 2008. The fact that this ratio exceeded 1 since 2006 is taken to imply that aggregate demand exceeded the rate of utilisation of productive capacity, consistent with non-accelerating inflation. Excess liquidity due to expansionary monetary policies and balance of payment surpluses helped fuel aggregate demand growth and widened the output gap. All of this of course rests on the presumption that the output gap is correctly measured by the gap between actual and estimated output. Whether this gap is in part due to measurement errors — there is a growing consensus that GDP estimates are highly suspect because of high measurement errors in the contribution of services — is glossed over. The measurement of inflationary expectations in terms of a lagged inflation measure is equally suspect as no light is thrown on how such expectations are formed and evolve over time. We fear, therefore, that the relationship established between current and lagged inflation is just a statistical one (for example, reflecting serial correlation between the two).
But far more serious is the failure to understand fully the dynamics of oil and food prices. Some key inter-relationships and the sequence of effects over time are succinctly summarised in the subsequent chapter: “Causes of High Food Prices” by C P Timmer- a leading scholar at Stanford. The more important point, however, is that this exposition flies in the face of the simplistic and misleading diagnosis of the preceding chapter (based on a background paper by two ADB economists).
Timmer emphasises that there are layers of cumulative causation of food price inflation. These comprise:

Rising living standards in China, India and other emerging economies leading to improved diets — especially greater consumption of vegetable oils and livestock products (and the feedstuff required).
Rapid deterioration of the dollar driving up the commodity prices quoted in dollars, and greater attractiveness of commodities as hedging options.
Mandates for corn-based ethanol in the US (and biodiesel fuels from vegetable oils in Europe).
Massive influx of capital into commodity markets in search of speculative gains.
Underlying all these demand drivers is the high price of oil.
Although these factors have been operating for several years, the last two have appeared more recently with the potential for changing the price dynamics in rapid and unexpected ways.
Some of the commodities (wheat, maize and soyabeans) are ‘multi-end-use’ commodities. Which particular use is driving market prices thus depends on the supply and demand structure of all alternative commodities, macroeconomic conditions and trade policies in importing and exporting countries. Multiple end-uses lead to “parameter instability” in the relationship between supply, demand and price. To illustrate, in one month demand for maize/corn to make ethanol may drive up the prices of corn, soyabeans and palm oil while in the next their prices may be delinked and driven more by their own supply and demand. In fact, much depends on substitution possibilities in both production and consumption of these commodities. So the assessment that food price inflation is largely of the demand-pull variety and a rigid monetary policy stance as the main remedial option for developing Asia is as unimaginative as it is specious.
The authors are, respectively, Professor of Public Policy, Faculty of Management Studies, University of Delhi and Visiting Scholar from University of Wisconsin-Madison