An econometric analysis of food inflation in South Africa

Abstract

Rapid commodity and food price increases were driven by the commodity super-cycle experienced between 2005 and 2008. This subsequently resulted in a wave of literature on food prices and food inflation. Two main lines of thought can be identified in the vast body of literature on this topic. The first is the so-called “agricultural economics” approach that considers food prices and inflation, in an aggregated or disaggregated form. Here the disaggregated form should be understood as product-specific analysis aimed at determining the transmission dynamics between two prices either in two locations or at two different nodes of the supply chain. In this type of analysis, the inflationary implications of the transmission process are considered either implicitly or explicitly. In an aggregated form, total food inflation is considered and the objective of these studies is usually to determine what the main fundamental drivers and their associated dynamics are. The second is a “monetarist” approach, that specifically considers the role of food inflation in headline inflation and other macroeconomic variables. Conceptually, these lines of thought allow one to think about food inflation at three different tiers, namely product specific inflation, aggregate food inflation and headline inflation. As a result, the three articles contained in this study are organised according to the above-mentioned tiers. Essay 1 considers vertical price transmission and its inflationary dynamics for two productspecific value chains in South Africa, namely wheat to bread and maize to maize meal. The consideration of product-specific value chains also assists in drawing conclusions on how the salient features of these value chains impact on price movements and inflationary dynamics. This study uses time series econometric techniques, that account for possible asymmetry, and results indicate that the price determination and price transmission processes in these chains are not contributing factors to the inflationary pressures that the above-mentioned chains have experienced over the past decade. Symmetric price transmission in both chains seems to further suggest no opportunistic behaviour on the part of firms to exploit situations where commodity prices decrease. Essay 2 examines the factors driving South African food inflation. It estimates the impact of key fundamental variables on aggregate food inflation, in terms of magnitude and duration by means of time series econometric techniques and Monte Carlo simulations. It concludes that the exchange rate, world food/commodity prices and local agricultural prices are the main drivers which, in turn, implies that food inflation in South-Africa is driven by structural supply shocks. Essay 3 reflects on the role of food inflation on headline inflation and adds to the growing body of literature that proves that food inflation should be included, and regarded in its own right, when monetary policy is devised. The study uses OLS methods to determine the presence of second-round effects associated with food inflation and find significant evidence thereof. In addition to this, persistence in food inflation, in an aggregated and disaggregated form is also established. The findings support advice for a more holistic policy approach to managing food inflation. This could include industry-specific policies, aimed at curbing price increases that result in first round food inflation, combined with monetary policy geared towards minimising the effect of second round food inflation. This study contributes to the existing literature on price transmission and food inflation in several ways. Each of the aforementioned articles makes a contribution in its own right. The first article explores how price transmission and price determination impacts productspecific inflationary dynamics. The second paper explores the key drivers of food inflation for a food net importing emerging market. The last article considers industry specific inflationary dynamics and explores possibilities for a combination of structuralist and monetarist policies that can be used to curb food inflation. Overall, the study links various ‘tiers’ of inflation conceptually. This link can prove useful when structural or industry related shocks need to be translated to inflation on a more aggregate level.