Charles River Associates

Collusion, information sharing and the winner's curse in the North Island livestock market

Email address: jmellsop@crai.co.nz

JEL Classifications: D44, L41

Abstract:
The New Zealand meat processing companies have a long history of exchanging information about livestock prices and other variables that affect livestock prices, such as predicted export prices and capacity changes. After raids in April 1995, the Commerce Commission brought proceedings against several North Island meat companies alleging that they fixed livestock prices between October 1992 and April 1995. The litigation was eventually settled, with penalties totalling $5.510 million being approved by the High Court in August 1998. However, analysis intriguingly indicates that the “agreed” price was almost invariably a minimum price, not a maximum one as might be expected if the companies were acting collusively in this input market. This fact suggests that there might be an alternative explanation for the behaviour of the meat companies to that charged by the Commerce Commission. The explanation that this paper proposes is that the North Island meat companies were exchanging information to reduce uncertainty. Spot trading in the livestock market is carried out by an auction mechanism; in particular, farmers explicitly compare the meat companies’ offers. Accordingly, this research uses auction theory as the tool for analysing the meat companies’ behaviour. It argues that the meat companies face the “winner’s curse” when they bid for livestock from farmers; therefore, they bid cautiously. However, information exchange between the meat companies that reduces their uncertainty will mitigate the winner’s curse, resulting in more aggressive bidding. Accordingly, the primary hypothesis of this research is that farmers were paid higher prices for livestock when the meat companies exchanged information. This prediction is the opposite of what we would expect if the meat companies were successfully colluding. The paper develops an econometric model to test the theoretical predictions. The results generally support the hypothesis that information exchange leads to higher livestock prices. A secondary hypothesis of this research is that information exchange will also reduce the variance of the livestock price. The statistical results support this proposition.

PDF file of paper: mellsop.pdf

Session: Environmental and Resource Economics

Time: Saturday, 7 July, 2:15pm - 3:45pm

Room: F