Blame crude oil and the calendar.
Gasoline prices have been moving up quickly around the country over the last few months, putting us back up to the highs of last summer. If you’re feeling a sense of deja vu, yes, it’s true, something similar seems to happen every spring. There has always been a strong seasonal in gasoline prices, with prices highest in the summertime when demand for gasoline is highest. The blue line in the graph below shows what this seasonal looked like before 2000. The height of the graph for each week is the ratio of that week’s average U.S. retail gasoline price (data source: EIA) to the value it had been at the start of the third week in September of that year, with this ratio then averaged over 1994-1999 to summarize the average seasonal trajectory. In a typical year over this period, gas prices ended up the year about 5% higher than they started. However, this price increase was not monotonic, but instead started out as a 5% decline in gas prices through January. That 5% decline, which resulted from the end of the summer driving season, when combined with the 5% cumulative increase over the year, meant that in a typical year, retail gasoline prices would rise 10% between January and May.
With the introduction of reformulated gasoline requirements in 2000, the process of converting from winter to summer fuel blends has become more costly, and the seasonal is much more pronounced. As shown in the red line (the average seasonal over 2000-2005) above, gasoline prices in recent years have dropped about 15% between September and January, and therefore exhibited an even sharper subsequent rise between January and April. The current NYMEX futures price is betting on the same thing happening next year, as evidenced by the 15% price difference between the May 07 and Jan 08 NYH RBOB contracts.
It’s interesting to calculate what you’d predict would have happened so far this year if the seasonal from 2000-2005 had simply been repeated exactly for 2006-2007. That is shown in the dashed red line in the figure above (which just reproduces the red curve from the previous figure, scaled to the September 2006 price of $2.62/gallon). Also shown for comparison is the actual price (in blue). It appears that a good deal of what has happened this year is just following exactly along the recent seasonal.
Of course, another important factor is the price of crude oil. Although crude oil prices do not themselves display a clear seasonal, this year they happened to decline in the fall and resurge in the spring, reinforcing the usual seasonal pattern for gasoline prices. One rough guide is that a one dollar change in the price of a barrel of crude oil means 2 cents per gallon of gasoline at the pump. One can then get a crude measure (if you’ll forgive the pun) of the contribution of this effect by multiplying the change in the WTI price (data from EIA) since September 11, 2006 by 0.02, and adding this to the 2000-2005 seasonal. This calculation is reported in the red line in the graph below.
The combined contribution of the usual springtime run-up in gasoline prices plus the effects of higher crude oil prices would lead us to expect something like a 60 cents per gallon increase in the retail price of gasoline in the U.S. between January and April of 2007, close to what actually happened.
I used to regularly get phone calls about this time of year from reporters asking me why gasoline prices had gone up. My answer was always, “because crude oil prices are higher and it’s springtime.” I haven’t received any calls so far this year, though. Maybe reporters have finally figured out there’s nothing particularly surprising about the recent rise in gasoline prices. Either that, or they’ve figured out that if they call somebody else, they get a more exciting soundbite.
Via GasBuddy, here’s what gasoline prices in your area look like at the moment: