The Labor Day holiday marks the end of the peak driving season in the United States and provides an opportunity to review year-to-date gasoline consumption. U.S. Energy Information Administration (EIA) current data indicate gasoline consumption in the first half of 2012 was 0.3 percent lower than in the comparable 2011 period. Economic growth, gasoline prices, and vehicle fleet efficiency are key determinants of gasoline use. So far in 2012, year-over-year economic growth that would generally lead to increased gasoline consumption has been more than offset by increases in retail gasoline prices and higher fuel efficiency of the in-use vehicle fleet.

Prior to February 2012, the United States had consumed less gasoline year-over-year for 13 out of 15 months since November 2010 (Figure 1). In 2011, U.S. gasoline consumption decreased on average almost 240,000 barrels per day (bbl/d), a 2.7-percent reduction from its 2010 level. Consumption continued to weaken, but at a slowing rate, into the first quarter of this year, declining by about 110,000 bbl/d (1.3 percent) from the comparable year-earlier level.

What is driving U.S. gasoline consumption?

In February 2012, however, almost 20,000 bbl/d more gasoline was consumed than in February 2011, the first year-over-year increase in 11 months. While gasoline consumption through June was slightly lower year-over-year, consumption in the second quarter increased by almost 60,000 bbl/d compared with 2011.

Real GDP for the first six months of 2012 is estimated to be 2.3 percent higher than for the same period in 2011. According to EIA's Short-Term Energy Outlook (STEO) model, the short-term income elasticity for total vehicle miles traveled by light- and heavy-duty vehicles together is about 0.58 (meaning a 1-percent increase in income results in a 0.58-percent increase in miles traveled). This elasticity estimate suggests motor fuel consumption should have been 1.3 percent higher this year than in the same period in 2011, assuming no year-over-year change in fuel prices, the efficiency of the vehicle fleet and the light-duty vehicle share of total travel.

In its STEO model, EIA estimates the short-term elasticity of vehicle miles traveled with respect to the price of gasoline to be about -0.04 (a 1-percent increase in prices leads to a 0.04 percent decline in highway travel). The average pump price of regular-grade gasoline during the first six months of 2012 was 3.3 percent higher than the same period last year. Motor gasoline consumption would therefore have been slightly more than 0.1 percent lower than in 2011, assuming no year-over-year change in economic activity or improvement in the average efficiency of the vehicle fleet. Combined, the estimated net effect of income and price changes and their respective elasticity estimates during the first half of 2012 imply a 1.2-percent increase in gasoline consumption over the comparable year-ago period assuming a constant share of light-duty vehicle travel.

A rough estimate of average gasoline use per mile traveled, a measure of the efficiency of the in-use vehicle fleet, can be obtained by dividing total gasoline consumption reported by EIA by preliminary total highway miles travelled reported by the Federal Highway Administration. Based on this measure, the implied average fuel efficiency of the in-use light-duty vehicle fleet rose by roughly 1.1 percent in the first half of 2012 versus the comparable year-ago period. Efficiency gains likely reflect both increasingly stringent Corporate Average Fuel Economy (CAFE) standards that were implemented for light-duty trucks starting in model year 2008 and for passenger cars starting in model year 2011 and consumer vehicle choices in an era of higher gasoline prices.

The net effect of the income, price, and fuel efficiency impacts as modeled by EIA implies almost flat year-over-year gasoline consumption. Keeping in mind that measures of freight activity suggest some actual decline in the light-duty vehicle share of total vehicle travel between the first half of 2011 and the first half of 2012, the STEO modeling appears to be roughly consistent with, the 0.3-percent year-over-year decline in gasoline use reflected in EIA's current data for January through June 2012.