Saturday, 2 August 2008

Asymmetric price adjustment and petrol prices

Petrol prices are in the news more often than not these days. One particular point that has been made is that when retail petrol prices are going up, they tend to go up quickly, but when the same prices are heading down, they tend to go down slowly. This asymmetry leads some critics to berate oil companies for not passing along the savings when crude oil or wholesale petrol prices fall. For example the consumer.org.nz website writes
Oil companies led by BP - New Zealand's largest - have given motorists a hard time since the New Year. They've raised prices at the pump as soon as crude oil prices rise and then failed to pass on savings from falling prices and the high New Zealand exchange rate.
This is just? May be not. It turns out that the pricing phenomena may be more appropriately pinned on petrol consumers. When retail petrol prices are rising, consumers search harder for the best price; when prices are falling, consumers ease up on search.

Matthew Lewis of the Ohio State University has a paper on the Asymmetric Price Adjustment and Consumer Search: An Examination of the Retail Gasoline Market (pdf). The abstract reads
It has been documented that retail gasoline prices respond more quickly to increases in wholesale price than to decreases. However, there is very little theoretical or empirical evidence identifying the market characteristics responsible for this behavior. This paper presents a new theoretical model of asymmetric adjustment that empirically matches observed retail gasoline price behavior better than previously suggested explanations. I develop a “reference price” consumer search model that assumes consumers’ expectations of prices are based on prices observed during previous purchases. The model predicts that consumers search less when prices are falling. This reduced search results in higher profit margins and a slower price response to cost changes than when margins are low and prices are increasing. Following the predictions of the theory, I use a panel of gas station prices to estimate the response pattern of prices to a change in costs. Unlike previous empirical studies I focus on how profit margins (in addition to the direction of the cost change) affect the speed of price response. Estimates are consistent with the predictions of the reference price search model, and appear to contradict previously suggested explanations of asymmetric adjustment.
The outcome is that if consumers are diligent about seeking out the lowest available price, maybe retail petrol prices will drop more more quickly when wholesale prices fall.

No comments: