Last week on vacation we stayed in a paddle-in cabin 20 miles northeast of Ely, Minnesota. No phone, no television, no computer, hence no information on the horrific consequences of hurricane Katrina.
We ventured into town twice for dinner, once on Monday (when we saw a newspaper and saw the looming threat) and once on Wednesday. When we first drove into town on Saturday, gas was $2.54/gallon. On Monday it had increased to $2.59. On Wednesday it had increased to $2.99.
As soon as I saw the 40-cent increase, I said “the hurricane hit the Gulf refineries”. Sure ’nuff, when we bought a paper we saw what had happened.
By the time we left the area on Friday, gas prices had increased to $3.29.
The severity of price spikes like this reflects the magnitude of the shock (including the transportation time to such far away places), and the integration and sophistication of global energy markets to transmit that much information that quickly. It also reflects the consistent tendency to form expectations of the worst case scenario. If that worst case is not realized, as it was not in this case, prices tend to decline. To the extent that the price spike persists, it could do so because of uncertainty about just how bad the refinery damage is, as well as underlying scarcity.
Anecdotally, I have heard that $3+ gasoline induced dramatic curtailment in driving over the weekend. Some people wonder what it takes to get people to drive less; now we have some more information on that. Furthermore, retail prices have started to fall since Monday.
Bottom line: prices are very parsimonious, yet very sophisticated, transmitters of information about scarcity and expectations. They seem to have done their job in the past week.