Marketing margins and price transmission on the Hungarian beef market

Abstract

There is a wealth of literature on farm-retail price spread for different commodities and countries. However, research on price transmission and marketing margins in the transition economies is still limited. The paper analyses two specific aspects of transition: the larger probability of asymmetric price transmission and structural changes in the case of Hungarian beef chain. The article identifies the date of structural break applying the Gregory and Hansen procedure with recursively estimated breakpoints and ADF statistics. Exogeneity tests reveal the causality runs from producer to retail prices. Homogeneity is rejected, suggesting a mark-up pricing strategy. Price transmission analysis suggests that, despite the common belief, price transmission on the Hungarian beef meat market is symmetric on both long and short run.

Keywords:

• Marketing margins
• asymmetric price transmission
• cointegration in the presence of structural breaks
• error correction
• beef market

The authors gratefully acknowledge financial support from the CERGE-EI Foundation in the framework of research grant ‘Price Transmission on Hungarian Agri-Food Markets’ (RRCV-23). The content of the publication is the sole responsibility of the author and in no way represents the views of the CERGE-EI Foundation.

Notes

1. As Bojnec and Günther (2005) point out, the constant margin might also depend on various other factors (e.g. existence of returns to scale, mark-up changes, technological or other input cost changes) beyond the farm component of the retail good.

2. Consider the first order autoregressive process, AR(1):The process is considered stationary if |ρ|<1, thus testing for stationarity is equivalent with testing for unit roots (ρ= 1). Rewriting to obtain:

3. There is some evidence of a structural break in the retail prices series occurring in mid 1994. Test results however, were highly dependent on the deterministic specification, choice of lag length and selected break date search methods, sometimes providing un-interpretable results (e.g. stationary level but non-stationary first differences).

4. With the inclusion of an intervention dummy corresponding to an outlier in 1992:10, residuals can be rendered normal.