Friday, May 28, 2010

Holiday drivers will see a drop at the gas pump

Washington Post article "Holiday drivers will see a drop at the gas pump" (May 29th, 2010) makes a seemingly innocuous claim that doesn't appear to have empirical support. Specifically, the article claims that the approaching summer is a time in which gasoline prices increase.

Memorial Day weekend, the traditional start of the summer driving season, typically means higher prices at the gas pump. But drivers hitting the road for the holiday this year will be greeted with falling prices instead, thanks in part to investor jitters about the global economy that have sent crude oil prices tumbling in recent weeks.

The average monthly price data on regular unleaded gasoline does not appear to support the claim that "driving season" has meant higher prices historically. First, for reader edification, we display the inflation-adjusted relative price changes in regular unleaded gasoline since 1976, with the 1982-1984 price set to 1 (click to enlarge).

It is, of course, difficult to discern whether or not there is a modulo 12-month pattern from the time series display. Therefore, we produce the monthly deviations from a linear December-to-December trend (click to enlarge). If there is indeed a trend, we should see an increase during the Summer months, which certainly doesn't jump out at Corrections.

Viewing the data in a different way, we can see general monthly effects by looking at year-month price data (click to enlarge).

Finally, we can look at the same graph, but of off-December-to-December deviations (click to enlarge). If a trend is present, it should be apparent from a continual up-slope heading north-northwest on the figure between April and August--something that, if it is present, we cannot see.

In summary, a quick graphical examination of the data appears to belie the Post's statement that regular unleaded gasoline prices go up during the Summer. Corrections would further suggest that though we undoubtedly have a shift in the demand for gasoline, we also are likely to have a shift in supply. Indeed, we would suggest that one shouldn't expect a perpetual increase in price much above the cost for storing gasoline for a few months--otherwise a predictable and significant arbitrage opportunity is present--a thing abhorrent to nature.

No comments:

Post a Comment