Market Integration with Transaction Costs in Developing Country Staple Food Markets: the Case of the Malawi Maize market

Abstract

We investigate measurement of market integration of staple food markets in developing countries. The analysis takes the Parity Bound Model as starting point and modifies this model by parameterizing and estimating transaction costs. The specification of transaction costs takes account of transport costs, fixed source costs, fixed destination costs, ad valorem taxes & levies and seasonality an is implemented on the basis of a specific sub-sample of price differentials. Price differentials combined with predicted transaction costs enable the measurement of market integration for each location and each period. The proposed method is applied to the Malawi maize market with monthly data from June 1999 to October 2009 for 26 districts. This period covers two major food shortages (2001-2003 and 2005-2006) as well as the impact of the global rise in food prices in 2008. Predicted transaction costs are shown to be in the same order of magnitude as survey observations of transaction costs of domestic maize trade in Malawi. The evidence indicates that markets are particularly well integrated during food shortages and substantially less in periods without food shortage. However, this result is largely due to high transaction costs during food shortages that make trade economically unattractive. Market integration develops markedly different between regions and districts: various districts in the south show signs of barriers to trade while districts in the north suffer from frequent high transaction costs. Low levels of market integration observed outside periods of food shortages, as well as high transaction costs during food shortages both suggest that trading infrastructure is not sufficiently developed.