Chapter
2

Gasoline Prices and Vehicle Markets

The longer gasoline prices remain high, the broader the scope of actions consumers will take in response—in part because the longer high prices are sustained, the more they affect consumers’ expectations about future prices. Those expectations influence consumers’ long-term choices in several areas—including their decisions about what kinds of automobiles to drive and how many miles they are prepared to commute to work. All of those choices impinge on gasoline consumption and are in contrast to consumers’ immediate, short-term, largely behavioral adjustments to high gasoline prices, which involve how fast or how much to drive, for example.

A shift in the kinds of vehicles consumers buy can affect overall gasoline consumption, but only gradually because many vehicles already in operation when the shift occurs remain in use for a dozen years or more. In contrast, signs of a shift can be detected relatively quickly in vehicle sales data, as was apparently the case with automobile sales after the average price of a gallon of gasoline first rose above $3, in September 2005. That price threshold was exceeded again in the spring and summer of 2006 and yet again through much of 2007. Since 2005, the sale of cars relative to light trucks has increased, after declining for several decades. After stagnating for a comparable period, the average fuel economy of new cars has increased, as has that for light trucks.

Whether those effects will be sustained in future vehicle model years depends on whether real gasoline prices remain substantially above their historic average, how those prices affect automakers’ future product decisions and the demand for those products, and how quickly consumers’ real income grows in relation to the growth in their fuel costs of driving. If gasoline prices drop back to earlier levels, the effect of high prices on the overall passenger vehicle fleet might be limited to vehicles sold while consumers expected gasoline prices to remain high. If high gasoline prices persist, however, and if consumers continue to respond as they have in the past few years, the composition of the passenger vehicle fleet eventually could resemble that of the vehicles sold in the 2005 and 2006 model years, as described below.

Some of the recent decline in market share for light trucks could be attributable to the imposition of stricter fuel economy standards for those vehicles, which include crossover and sport–utility vehicles (SUVs).1 The new standards could have led to price or attribute changes in some light-truck models that induced some consumers to purchase cars instead. The phase-out of some of the favorable tax treatment for business use of light trucks also could have changed some companies’ buying patterns. Some tax benefit is still provided for those purchases, however, and the effect of the phased-out benefit may not have been large.2 Still, the contribution of either factor to the loss in the light-truck market share remains uncertain.

The relationship between gasoline prices and the demand for automobiles was the subject of several economic studies in the 1980s in the aftermath of the gasoline price shocks of the 1970s. Those studies found that higher gasoline prices increased the demand for smaller, more-fuel-efficient vehicles relative to larger, less-efficient vehicles.3 That relationship continues to hold with recent vehicle sales, according to several current economic studies.4

Market Shares for Cars and Light Trucks

Sales of new cars as a share of all passenger vehicle sales increased noticeably in late 2005. Through August of that year, the share of cars was unchanged from its year-to-date share in August 2004. But light-truck sales in autumn 2005 were weaker than usual as businesses and consumers first experienced gasoline prices in excess of $3 per gallon after Hurricane Katrina.5 For the full year of 2005, cars constituted 45 percent of new-vehicle sales, almost a full percentage point higher than in 2004. The share since then has been even higher, around 47 percent (see Figure 2-1).

Figure 2-1. 

Market Share of Cars versus Light Trucks, 1976 to 2007

(Percent, new passenger vehicle sales)

Source: Congressional Budget Office based on data from the Bureau of Economic Analysis.

The upturn in the market share of cars since 2004 is particularly noteworthy because that share had been in decline for 25 years, falling lower every year after 1981, when more than 80 percent of new passenger vehicles were cars. The decline reflects the increasing popularity of minivans and, later, of SUVs and crossover vehicles, and the declining popularity of some types of cars, particularly station wagons. The adoption of corporate average fuel economy regulations contributed to that shift by requiring higher average fuel economy for cars than for light trucks, which led to the development of the minivan, a light-truck alternative to the station wagon. Because most minivans had relatively good fuel economy compared with that of other light trucks, and most station wagons had relatively poor fuel economy compared with that of other cars, the shift in the demand for high-capacity passenger vehicles from station wagons toward the more popular minivan helped automakers satisfy CAFE standards both for cars and for light trucks.6

Between 2004 and 2006, every major category of car gained and every category of light truck lost market share (see Table 2-1). The biggest gain was for large cars, which went from 8.0 percent to 9.4 percent of the market. The share of midsize cars also increased considerably, and although the market shares of compact and subcompact cars also increased, those gains were smaller. The underlying sales data indicate that those gains in market share for the most part reflect a decline in sales of light trucks (in every category) rather than an increase in car sales. Overall, the number of cars sold actually declined by about 1.5 percent each year from 2004 to 2006, or by a little more than 100,000 cars per year. (Sales in the large-car category increased by 10 percent.) However, over that same period, sales of light trucks fell by 10 percent, with nearly 1 million fewer new vehicles sold per year. (For 2007, August year-to-date unit sales of light trucks were down an additional 2 percent compared with the same period in 2006. However, car sales were 4 percent lower.)

Table 2-1. 

Market Shares of Different Types of New Vehicles, 2004 to 2006

(Percent)

Source: Congressional Budget Office based on data published in Automotive News.

Notes: SUV = sport–utility vehicle.

Market shares total 100 percent.

Despite the relatively large decline in light-truck sales, the data do not indicate the extent to which the increase in market share for cars occurred because would-be truck buyers merely delayed their truck purchases versus buying a car instead. The small decline in sales of cars could even be consistent with a large number of would-be truck buyers switching to cars and averting what might otherwise have been a larger decline in car sales. Such a possibility also is consistent with the considerable gains that occurred for large-car sales, because among all car types, large cars are the closest substitute for light trucks in terms of seating, storage capacity, and engine size. To the extent that some truck buyers have delayed purchasing a truck rather than switching to a car, sustained high gasoline prices could eventually lead to more car sales because consumers must eventually replace their existing vehicles.

Gasoline Prices and Vehicle Market Shares

A comparison of monthly data on the market shares of different types of vehicles against monthly average retail gasoline prices suggests that some light-truck buyers may have briefly delayed purchasing a new truck while gasoline prices were increasing. Price spikes in the spring of 2005, in October 2005 (after Hurricane Katrina), and in the spring of 2006 all coincided with sharp increases in the new-car market share (see Figure 2-2). Market shares for leading categories of light trucks—especially SUVs—went the opposite way, dipping as gasoline prices rose. When the price of gasoline has dropped, the market share of each vehicle type has tended to return to previous levels. Some of those movements in market share could be influenced by manufacturers’ sales incentives, which vary from one month to the next and sometimes are relatively stronger for cars, sometimes for light trucks. However, exploratory analysis of recent data on incentives, described later, suggests the incentives account for little of the variation in the relative market shares of cars and light trucks.

Figure 2-2. 

Passenger Vehicle Market Shares and the Real Price of Gasoline, 2004 to 2006

(Percent, new passenger vehicle sales)                                                                   (2007 Dollars per gallon)

Sources: Congressional Budget Office based on data from Automotive News and the Department of Energy, Energy Information Administration.

Note: The right-hand scale shows the average inflation-adjusted price per gallon for all grades and formulations of gasoline.

To assess the degree to which gasoline prices and vehicle market shares are related statistically, the Congressional Budget Office analyzed 36 months of data from January 2004 through December 2006 on sales of each type of vehicle. CBO computed the market shares from monthly sales data for specific vehicle models (for example, Toyota Camry), aggregated into total sales by vehicle category (for example, midsize cars). The aggregations are based on vehicle classifications from the fuel economy ratings published by the Environmental Protection Agency (EPA).

Current gasoline prices and the market shares of specific types of vehicles should be related to the extent that current prices influence consumers’ expectations about future prices and to the extent that vehicle operating costs (primarily fuel costs) influence consumers’ decisions about which vehicles to purchase.7 CBO’s analysis of the relationship between gasoline prices and vehicle market shares accounts for seasonal effects, a possible source of correlation between market shares and gasoline prices if both exhibit seasonal variation.8

On the basis of that reasoning, CBO constructed a model in which the market shares of different vehicle types are influenced by the current average inflation-adjusted price of gasoline and the current season (see Box 2-1).9 According to the model, the effect of gasoline prices on U.S. new-vehicle market shares has been such that a price increase of 60 cents per gallon (a 20 percent increase if the base price is $3 per gallon) is associated with an average increase in the market share of new cars of 2.6 percentage points (the sample average is 46.4 percent). That result reflects recent experience, although consumers’ future responses also would be affected by changes in automakers’ product offerings and pricing (which are also responsive to rising gasoline prices).

Box 2-1. 

Modeling the Influence of the Price of Gasoline on Vehicle Market Shares

All major car categories—from two-seaters and subcompacts to large sedans and wagons—have gained market share as the price of gasoline has risen, with gains of between 4.5 percent and about 9 percent for every 60 cent increase in the price of gasoline above $2.30 per gallon. At the same time, the market shares of all types of light trucks, from minivans and SUVs to pickup trucks and passenger or cargo vans, have fallen by 4 percent to 6 percent. For example, at average values, a 60 cent increase in the price of gasoline would have increased the market share of midsize cars by about 0.8 percentage points, which is a 5 percent increase over its average value of 16.6 percent. That price increase also would be associated with a decline of 1.2 percentage points or 4.5 percent in the share of new SUVs, on average, from a baseline share of about 27 percent (see Table 2-2).

Table 2-2. 

Estimated Effect of a 20 Percent Increase in the Price of Gasoline on U.S. Market Shares of New Passenger Vehicles

(Percent)

Source: Congressional Budget Office based on vehicle sales data from Automotive News and gasoline price data from the Department of Energy, Energy Information Administration.

Note: ** = significant at 1 percent; * = significant at 5 percent; SUV = sport–utility vehicle; n.a. = not applicable.

a. Market share averages are not sales weighted.

b. Over all vehicle categories, the net change in market share must be zero. Thus, a value of -0.1 for the market share of "passenger or cargo van" is not estimated directly but is derived from changes in the market shares of other vehicles.

c. Percentage increase in market share given in first column when percentage points in second column are added to it.

The results in Table 2-2 might not reflect the full shift in demand toward greater fuel economy, because automobile manufacturers can respond to higher gasoline prices—and to slower sales of light trucks—by offering incentives to purchasers to reduce inventories of new vehicles. CBO examined data on finance rate and price incentives for February 2006 through October 2006 and found that, during the summer of 2006, when gasoline prices were above $3 per gallon, the average financing rates that automakers were offering on some models of light truck did decline somewhat relative to the rates offered on some cars. (Incentives were not offered on all models every month, and for some models, no incentives were offered in any month.) Some consumers thus might have been encouraged to purchase a light truck when they otherwise would have chosen a car, partially counteracting the effect of higher gasoline prices on consumers’ decisions about which new vehicle to purchase and leading to slightly smaller gasoline price effects.

Changes in New-Vehicle Fuel Economy and Pricing

The 2005 model year might have marked a turning point not only in the long, steady decline in the new-car market share but also in the declining average fuel economy of new passenger vehicles. The recent increase in market share for new cars is responsible for some of the increase in the average fuel economy for the vehicle fleet; on average, a car can travel about 6.5 miles farther on a gallon of gasoline than a light truck can, according to data from EPA (see Figure 2-3).

Figure 2-3. 

Average Rated Fuel Economy for New U.S. Passenger Vehicles, 1975 to 2006

(Miles per gallon)

Sources: Congressional Budget Office based on data from the Environmental Protection Agency (EPA).

Note: Data are for sales-weighted fuel economy. Although a vehicle’s fuel economy performance is determined by EPA, either through EPA’s laboratory results or in test data submitted by the manufacturer, it is the National Highway Traffic Safety Administration (NHTSA) that determines compliance with corporate average fuel economy (CAFE) standards. NHTSA considers CAFE credits the automaker has earned, including those for hybrid and dual-fuel vehicles. NHTSA’s CAFE data are similar to those illustrated here, although the averages are slightly higher because of the credits.

However, the average fuel economy of new cars has itself been rising. Beginning with the 2001 model year, when the average fuel economy rating for new cars was 28.4 miles per gallon, the average began to increase, peaking at 29.2 mpg for the 2005 model year before slipping back slightly in 2006.10 Another reason for the increase in average new-vehicle fuel economy is that since 2005, automobile manufacturers have had to meet stricter CAFE standards for light trucks. An increasingly popular new kind of light truck, often called a crossover vehicle because it is built on a car chassis but with the higher clearance of a light truck (and thus is so classified for CAFE purposes), also is partly responsible for the rise in the average fuel economy of light trucks and may reflect, in part, a response by automakers to higher CAFE standards for light trucks. The CAFE standard for cars did not change. (Legislation signed in December 2007 requires a substantial increase in the average fuel economy of all new passenger vehicles sold in the United States. By 2020, the CAFE standard will increase to 35 mpg.11)

Although the average fuel economy ratings of cars and light trucks alike have been increasing since 2002, the average fuel economy for all new passenger vehicles declined through 2004 because the market share for light trucks was still rising at that time. Higher gasoline prices cannot explain why average new-car fuel economy ratings began to increase in 2001, because gasoline prices did not start rising until 2003 (see Table 2-3). However, gasoline prices have risen steadily since then, helping to increase the demand for vehicles that are more fuel efficient.

Table 2-3. 

National Average Gasoline Prices, 2000 to 2006

(Dollars per gallon)

Source: Congressional Budget Office based on data from the Department of Energy, Energy Information Administration (for nominal prices) and from the Department of Commerce, Bureau of Labor Statistics (real prices calculated from the consumer product index for all urban consumers).

Note: Prices are for all grades and all formulations.

That increased demand and the resulting improvement in vehicle fuel economy appear to have affected the pricing of some new vehicles. In the past, some automakers might have set lower prices on fuel-efficient vehicles—and higher prices on other vehicles—as part of their overall strategy to meet CAFE standards; an increase in demand for fuel-efficient vehicles would permit automakers to rely less on pricing to meet the standards. If so, prices of new fuel-efficient vehicles should have increased relative to prices of larger, less-efficient vehicles. There is some evidence that that has occurred (see Box 2-2), meaning that the rise in average fuel economy for new vehicles does not fully reflect the shift in consumer preferences toward greater fuel efficiency.

Box 2-2. 

CAFE Implications of Rising Average Fuel Economy

Changes in the Used-Vehicle Market

The recently observed price shifts for new vehicles are reflected in used-vehicle prices as well. Average prices of fuel-efficient used vehicles have been rising, and those of less-efficient vehicles have been falling. That is as expected: In both markets, consumers’ preferences for fuel-efficient vehicles should be similarly affected by rising gasoline prices—which should affect prices similarly in both markets. Even if consumers’ preferences were not affected in exactly the same way by gasoline prices, price increases on fuel-efficient new vehicles would cause the prices on similar used vehicles to rise as some consumers shifted from buying new vehicles to buying similar used vehicles. Similarly, smaller price increases on less-fuel-efficient new vehicles would limit how quickly prices could rise for comparable used vehicles. New-vehicle prices serve as a ceiling on consumers’ willingness to pay for used (noncollectible) vehicles.12

Consistent with a shift in consumer demand toward greater fuel economy, monthly average prices in the United States for used vehicles in the large-SUV and luxury car categories declined steadily over the 49 months for which CBO collected data, through July 2006. Over the same period, average prices increased for sporty, full-size, and compact or midsize used cars (see Table 2-4).13 It is not possible to attribute those price changes definitively to vehicles’ fuel economy ratings, because the data do not include information on other characteristics (such as the age of the vehicles or prices when new) that also could have contributed to the price trends. However, those trends are consistent with the new-vehicle market: Prices for small vehicles have been rising while those for larger vehicles have fallen or, for new cars, have risen more slowly (see Figure 2-4).

Table 2-4. 

Average Monthly Change in Price of Used Vehicles, 2002 to 2006

Source: Congressional Budget Office based on monthly average used-vehicle prices and vehicle classifications from Automotive News.

Note: *** = significant at <0.01 percent; ** = significant at <5 percent; * = significant at <10 percent; n.s. = not significant at conventional levels; SUV = sport–utility vehicle.

a. Includes SUVs, luxury sport wagons, and small sport wagons.

Figure 2-4. 

Average Wholesale Prices for Used Vehicles, July 2002 to July 2006

(Thousands of dollars)

Source: Congressional Budget Office based on data published in Automotive News.

Notes: SUV = sport–utility vehicle.

Automobile category names are from Automotive News. Data are monthly sales figures. Straight lines are trend lines added by CBO.

Figure 2-4 displays average prices for the vehicle categories with the strongest and most statistically significant price trends. Although the data do not include information about the average size or weight of vehicles in each category, the category names are an indication of likely differences in vehicle size. Based on category names alone, the price trends in Figure 2-4 suggest that vehicles with declining prices are larger than average and thus are likely to have fuel economy ratings below the applicable CAFE standard. The vehicle categories with increasing prices seem likely to include somewhat smaller-than-average vehicles with higher fuel economy ratings. (Compliance with CAFE standards is determined on the basis of an automaker’s fleetwide average fuel economy, so vehicles with ratings lower than the applicable standard do not subject the automaker to penalties if they are balanced by sales of more-fuel-efficient vehicles.) The average monthly dollar changes in vehicle prices corresponding to the sloping trend lines in Figure 2-4 are given in Table 2-4.


1

For a concise introduction to corporate average fuel economy standards, see Congressional Budget Office, The Economic Costs of Fuel Economy Standards Versus a Gasoline Tax (December 2003), p. 2.


2

See Congressional Budget Office, Budget Options (February 2005), Option 32, p. 308. The phaseout was mandated by the American Jobs Creation Act of 2004.


3

John Greenlees discussed the effects on new-vehicle demand in "Gasoline Prices and Purchases of New Automobiles," Southern Economic Journal, vol. 47, no. 1 (1980), pp. 167–178. For evidence that used-vehicle demand responds similarly to gasoline prices, see George G. Daly and Thomas H. Mayor, "Reason and Rationality During Energy Crises," Journal of Political Economy, vol. 19, no. 1 (1983), pp. 168–181; and James Kahn, "Gasoline Prices and the Used Automobile Market: A Rational Expectations Asset Price Approach," Quarterly Journal of Economics, vol. 101, no. 2 (1986), pp. 323–340.


4

See, for example, Joshua Linn and Thomas Klier, "Gasoline Prices and the Demand for New Vehicles: Evidence from Monthly Sales Data" (working paper, March 2007), http://tigger.uic.edu/~jlinn/vehicles.pdf; and Sarah West, "The Effect of Gasoline Prices on the Demand for Sport Utility Vehicles," (prepared for a session on "Demand Estimation in Environmental Economics," Midwest Economics Association Meetings, March 2007), www.macalester.edu/~wests/SarahWestMEA2007.pdf. For a more general treatment, see Walter McManus, "The Link Between Gasoline Prices and Vehicle Sales: Economic Theory Trumps Conventional Detroit Wisdom," Business Economics (January 2007), pp. 53–60.


5

According to data published in Automotive News,a trade publication, in other recent years, the share of light-truck sales was higher in the autumn than in the rest of the year. For 2003, 2004, and 2006, the August year-to-date market share for light trucks was1 to 2 percentage points lower than the full-year share, indicating stronger light-truck sales for September through December.


6

See Congressional Budget Office, The Economic Costs of Fuel Economy Standards Versus a Gasoline Tax. Automakers’ CAFE ratings depend not only on the mix of vehicles they sell but also on the fuel economy rating of each model. As automotive technologies have become more fuel efficient, automakers have been able either to improve fuel economy or to add fuel-consuming features (such as more horsepower) without reducing fuel economy. With lower gasoline prices, consumers tended to favor other attributes over fuel economy, and many manufacturers’ CAFE ratings stayed flat, despite technological advances. Higher gasoline prices have led to a shift in consumer preferences toward greater fuel economy.


7

Current gasoline prices could reflect consumers’ best forecast of likely future prices—what economists call "static" expectations. (For empirical evidence, see Daly and Mayor, "Reason and Rationality During Energy Crises," an examination of vehicle prices in the wake of the 1973 and 1979 oil price increases.) Consumers also could implicitly make more sophisticated, "dynamic" price projections by, in effect, averaging prices observed in recent months. It is straightforward to model such expectations, but it is not considered here. Given the recent upward trend in gasoline prices and the downward trend in the market share for light trucks, such a model would probably yield results similar to those of the static model.


8

Gasoline prices typically rise in the warmer months as longer daylight hours and vacation travel increase demand. Light-truck sales recently have tended to be higher in the autumn, when gasoline prices usually are below their annual peak.


9

A more thorough analysis would incorporate the effects of relative vehicle prices. Automakers have responded to rising gasoline prices by raising the prices of smaller vehicles relative to those for larger ones. Thus, CBO’s analysis could understate the effects of gasoline prices on the market share of a given type of vehicle, because some of those effects would have been partly neutralized by changing vehicle prices.


10

From 2003 to 2006, midsize cars made the biggest gain in sales-weighted average fuel economy, increasing from 23.9 mpg to 25.3 mpg. Among light trucks, the biggest increase was for SUVs, which went from 18.3 mpg in 2003 to 19.5 mpg in 2006. Sizable improvements also were made by compact cars and minivans. The averages reflect EPA’s adjustment of about 15 percent below the vehicles’ CAFE ratings to account for estimated differences between laboratory-measured and actual fuel economy. EPA will begin using a new method of adjustment in 2008. See Robert M. Heavenrich, Light-Duty Automotive Technology and Fuel Economy Trends: 1975 Through 2006, EPA420-R-06-011 (Environmental Protection Agency, Office of Transportation and Air Quality, Advanced Technology Division, July 2006), p. A-10, www.epa.gov/otaq/cert/mpg/fetrends/420r06011.pdf; and Environmental Protection Agency, Office of Transportation and Air Quality, Compliance and Innovative Strategies Division and Transportation Compliance Division, Light-Duty Automotive Technology and Fuel Economy Trends: 1975 Through 2007, EPA420-R-07-008 (September 2007), p. A-9, www.epa.gov/otaq/cert/mpg/fetrends/420r07008a.pdf.


11

See the Energy Independence and Security Act of 2007. For model years 1996 to 2004, the CAFE standard for light trucks was 20.7 mpg. It was raised to 21 mpg for 2005 and 21.6 mpg for 2006. The standard is now 22.2 mpg for light trucks in model years 2007 and later. The CAFE standard for cars, unchanged since 1990, is 27.5 mpg. See National Highway Traffic Safety Administration, "CAFE Overview, Frequently Asked Questions," www.nhtsa.dot.gov/cars/rules/cafe/overview.htm.


12

Although CAFE standards apply only to new vehicles, they also can affect prices of used vehicles: Where some manufacturers’ pricing is affected by the standards, as when just-compliant manufacturers offer lower prices on their fuel-efficient vehicles to boost sales and CAFE ratings, those lower prices will attract some potential buyers of used vehicles, thus reducing the demand for (and average prices of) fuel-efficient used vehicles.


13

The used-vehicle classifications are from Automotive News.



Previous Table of Contents Next