Despite the drop in oil and gasoline prices this past fall, oil companies are once again reporting record profits. Last week, ExxonMobil reported $39.5 billion for 2006, breaking its own record for the largest annual profit by a U.S. company for the third year in a row. At least one consumer group is calling on Congress to investigate possible manipulation of prices.
But this concern is not new. Declining gasoline prices in the run up to Election Day in 2006 caused many voters to think there was deliberate manipulation involved. In a September Gallup poll, 42 percent of respondents agreed with the statement that the Bush administration “deliberately manipulated the price of gasoline so that it would decrease before this fall's elections.”
Were Americans correct in assuming manipulation? Or was the gasoline price decline before the 2006 mid-term elections following historical patterns, as many industry insiders claimed?To answer that question we developed an index using gasoline price data from the U.S. government’s Energy Information Administration to examine the variation in the national average price for regular gasoline from Memorial Day, the start of the summer driving season when demand for gasoline historically rises, to Election Day in early November.
As the graph above shows, gasoline prices in 2006 followed a very different path than they have over the past 15 years and in the last seven election years.
The gasoline market rule of thumb predicts relatively high prices starting at Memorial Day, a peak at Labor Day and then a decline through the fall. The average of all years since 1991 supports this generalization of market behavior.
But if you examine just the election years, you see a deviation from the norm with the loss of the Labor Day peak and a slight overall decline from the Memorial Day price. In those election years average gasoline prices declined only slightly more than in non-voting years but then tracked the overall decline in prices in the fall between 1991 and 2006.
In 2006, however, gasoline prices rose higher than the typical year, then started a steep drop in the first half of August to the November 7 elections. The $2.20 average on the day prior to the election was the lowest recorded gasoline price of 2006. The year then ended with an uncharacteristic rebound of gasoline prices.
Since 1991, the average gasoline price change from Memorial Day to the first Monday of November has been a decline of 2.9 percent. In 2006, the drop was 23.3 percent. The only other comparable decline was 29.2 percent in 2001 following 9/11 as demand shriveled amid sharply curtailed traveling across the country.
Furthermore, the price of gasoline went up after the election in 2006—unlike in previous years. This past November gasoline prices climbed 4.4 percent, to $2.30, but in previous years the average change after the first Monday in November is a 2.6 percent drop.
Only 1999 experienced similar growth, gaining 3.5 percent in November. But in that non-election year, gasoline prices experienced rare steady growth throughout the summer and fall. Since the Energy Information Agency started keeping weekly data on gasoline prices, no other year had the extreme swings that 2006 had prior to and after the elections.
But was 2006 so unusual? To help answer that question, we examined five other years that had some characteristics similar to 2006 as seen in the following figure.
The EIA data shows that after Memorial Day 2006 gasoline prices rose at a similar rate (0.45 percent per week) over the same period in 1999 (0.63 percent per week) until the first week in August, when the prices sharply declined at a rate (-2.44 percent per week) comparable to the decline after the 2001 terrorist attacks of September 11 (-2.70 percent per week) and Hurricane Katrina in 2005 (-2.76 percent per week).
As for the atypical rise in prices in November 2006, two other years, 1991 and 1996, also experienced a similar trajectory. However in each of those years, gasoline prices were relatively stable throughout the year and never varied more than 4.9 percent and 6.4 percent, respectively, from the Memorial Day price. Returning to our initial question, it is clear that 2006 did not follow the traditional gasoline price pattern. The steep price decline in the fall of 2006 was unique for an election year and only comparable to two other years both of which experienced major domestic disasters—the terrorist attacks in 2001 and the debilitating hurricane season in 2005—neither of which were national election years.
Supply and demand issues, of course, certainly influenced the oil and gasoline markets last year. Oil prices started their decline in August, for example, after it became clear that a total shutdown of the Prudhoe Bay oil pipeline was unnecessary, as you would expect from simple supply considerations.
Still, there’s no getting around the fact that only in 2006 was there such a sharp drop in prices prior to the mid-term elections followed by such a rise in prices immediately thereafter. The data cannot tell us whether manipulation was involved in the price drop before the November elections, but it does suggest that something other than supply and demand was influencing the gasoline market in 2006.
Even Lee Raymond, the former Chairman and CEO of Exxon Mobil agrees. He said in a November 9, 2005 Senate hearing on energy prices and profits that, “I do not think that the fundamentals of supply/demand, at least as we have traditionally looked at it, have supported the price structure that is there.”
