Spatial Integration of Milk Markets in Uganda

Abstract

This paper examines whether milk prices within Kampala are cointegrated with those in other major towns of Uganda. The
period of estimation is from July 2005 to March 2015. We use retail monthly prices for raw milk. We obtained the dataset
from Uganda Bureau of Statistics. Using Engel Granger’s two‐step error correction approach, we examine the relationship
that exists between prices for raw milk in Kampala and the regional markets. Our analysis provides evidence of a long‐run
cointegration relationship between Kampala and Gulu, Mbarara and Masaka milk markets. On the contrary, we failed to
reject the null hypothesis of no cointegration between Kampala and Arua, Jinja and Mbale milk markets. Regarding the
Granger causality tests, the results reveal that causality mainly originates from the milk supply towns. This implies that milk
prices in Uganda are supply driven. The speed of adjustment of the model is 50%, i.e. half of the disequilibrium in the
system is corrected in one month time period. These results have important implications for the dairy sector in Uganda. The
paper discusses these implications.