Marketing and Pricing Dynamics in the Presence of Structural Breaks: The Hungarian Pork Market

Abstract

The study of marketing margins and price transmission on various commodity markets has been a popular research topic of the past decades (see Meyer & Von Cramon-Taubadel, 2004, for a recent survey). However, with a few exception these studies focused on developed economies. This article examines this phenomena on the Hungarian pork market. The Johansen (maximum likelihood, 1988) or Engle and Granger (2-step, 1987) cointegration tests do not reject the no-cointegration null hypothesis between the Hungarian pork producer and retail price series. Therefore, we applied the Gregory and Hansen (1996) procedure with recursively estimated break points and Augmented Dickey Fuller (ADF) statistics and found that the prices are cointegrated with a structural break occurring in April 1996. Exogeneity tests reveal the causality running from producer to retail prices both in the long and short run. Homogeneity tests are rejected, suggesting a markup pricing strategy. Price transmission modeling suggests that price transmission on the Hungarian pork meat market is symmetric in the long run but asymmetric in the short run; that is, processors, wholesalers, or retailers might take temporary advantage should price changes occur.

KEYWORDS:

• Hungarian pork market
• marketing margin
• price transmission
• pricing
• structural breaks

Notes

*Significant at 5%; **Significant at 1%.

†Nonnormality—implies that the test results must be interpreted with care, although asymptotic results do hold for a wider class of distributions (Von Cramon-Taubadel, 1998).

Source: Hungarian Central Statistical Office.

RATS 6.0 code and EVIEWS 5.0 software was used to test the order of integration.

The Gregory and Hansen (1996) cointegration tests in the presence of structural breaks were carried out using a GAUSS code.

Results were substantiated using the Johansen et al. (2000) maximum likelihood cointegration procedure in the presence of structural breaks. MALCOLM code, in RATS programming language, is available to test cointegration with up to two structural breaks.

With the inclusion of intervention dummies, they can be rendered normal.