Skip Navigation

Environment Magazine September/October 2008


July-August 2007

ResizeResize Text: Original Large XLarge Untitled Document Subscribe

Examining New U.S. Fuel Economy Standards

After decades of futile attempts to increase U.S. fuel economy standards for passenger cars, which have remained unchanged since enactment of the Corporate Average Fuel Economy (CAFE) Standards in Title V of the 1975 Energy Policy Conservation Act, it seems increasingly likely that new and tougher standards will be enacted in the near future—especially after the Senate’s 21 June passage of energy efficiency bill H.R. 6. As this magazine went to press, the bill, which calls for a 40 percent increase in vehicle fuel economy by 2020 among other efficiency and alternative energy goals, was headed to the House of Representatives for more debate. Congress has seen proposals like this since the 1980s, but this is the first time that one of them has passed in the Senate. The Bush administration has also weighed in with a proposal to increase new vehicle fuel economy by 4 percent per year from 2011 to 2017,1 and the administrator of the National Highway Traffic Safety Administration (NHTSA) has asked Congress to grant the Secretary of Transportation the authority to restructure and increase CAFE standards for cars, a power denied by the original CAFE legislation.2

A confluence of events has led to this change of political climate, including: the failure of world oil production and refining capacity to keep pace with rapidly growing demand, especially from China and other emerging economies, which has led to the highest oil prices since the 1980s and growing fears that world production of conventional oil3 may be close to its peak and rapid decline; the escalating influence of oil resources on geopolitics as China seeks to guarantee its future access to supplies;4 enhanced revenues from the higher prices, which prop up authoritarian regimes in Iran, Venezuela, Russia, and elsewhere and allow them increasing freedom of action; the enhancement of the role of climate change in political decisionmaking by new reports from the Intergovernmental Panel on Climate Change (IPCC), with much strengthened language about the probability and severity of climate change and man’s influence on it;5 and a recent Supreme Court decision rejecting the Environmental Protection Agency’s assertion that it has no authority to regulate greenhouse gas emissions.6

New fuel economy standards will represent an ambitious and expensive undertaking on the part of the automobile industry and the nation, and proposals for new standards deserve careful congressional and public scrutiny.

Do Fuel Economy Standards Make Sense?

The CAFE standards adopted by the United States in 1975 played a crucial role7 (assisted strongly by rising oil prices) in raising the average fuel economy (EPA test values) of new U.S. passenger cars from 15.8 miles per gallon (mpg) in 1975 to 28.0 mpg in 1987. During the same period, light trucks went from 13.7 mpg to 21.6 mpg.8 Similar standards or their equivalent are now in place in multiple countries throughout the world, and one of the United States’ closest neighbors, Canada, is actively pursuing new standards of its own. Nevertheless, not everyone agrees that fuel economy standards are a good idea.

Aside from domestic automakers, who fought the original CAFE standards and continue to fight new attempts to strengthen them, there are numerous opponents of new standards, among, for example, members of Congress, economists, and auto enthusiasts. Their opposition centers around a range of arguments about the limitations of new standards and their impacts on oil use, public safety, consumer choice, vehicle markets, and the economy.

Fuel economy standards are, as their critics suggest, limited in their effect. They do not affect the current fleet, which will be an important factor in vehicle fuel use for decades. They do not reduce levels of driving and may actually increase driving somewhat—the latest research indicates that roughly 10 percent of the fuel savings benefits of new standards will be lost to the increased driving that results from lower fuel costs associated with more efficient cars.9 Current standards did not reduce absolute levels of automobile fuel use, and new ones may not either, but the current standards did reduce fuel use from what it would otherwise have been, and new standards will at least do that as well. In addition, fuel economy standards do not affect public transport or non-motorized transport (walking and biking), nor do they affect driving behavior. Opponents use these limitations to argue that standards are not effective. A better argument would be that standards are not enough by themselves to fully address energy security and environmental problems.

Standards do push automakers to build vehicles that consumers might not otherwise choose, as opponents claim. That is what regulations are for: When market forces alone will not provide the outcome desired by society, regulations override the markets. If consumers wanted fuel economy improvements badly enough, automakers would almost certainly provide them. The problem here is that vehicle buyers do not value fuel economy all that highly, for a variety of reasons. One is that they do not (and should not be expected to) take account of the climate change and energy security implications of their purchases. Also, purchasers of new vehicles are, as a group, wealthier than average, and the potential net savings from higher fuel economy represents only a small fraction of their annual income, even if fuel prices go up. Perhaps more important is that vehicle buyers have shown that they place a high value on vehicle characteristics such as higher power and larger size—characteristics that compete directly with fuel economy for the potential benefits of energy-
saving technology. Over the past 20 years, energy-saving technology has been used to allow vehicles to get bigger, heavier, faster, more luxurious, and safer without reducing fuel economy, when perhaps the goal should have been increasing fuel economy instead.

Economists have tended to focus on the advantages of using the market—specifically, higher taxes on fuel—as a more efficient means to reduce fuel use. However, high gasoline prices by themselves do not seem to be enough to obtain the very high fuel economy levels and reduced oil use desired by many in the policy community (nor are they likely to be politically acceptable). Even the Europeans and Japanese, who have much higher gasoline prices than those in the United States, rely on fuel economy standards or their equivalent and an array of vehicle registration fees (tied to efficiency) to reduce consumption.10 Higher gasoline taxes, however, would certainly help make standards more economically rational for consumers.

Another argument used vigorously by automakers and other CAFE opponents in the past is that CAFE kills. Past studies by NHTSA have concluded that vehicle downsizing associated with previous CAFE legislation caused more than 2,000 traffic fatalities yearly.11 CAFE opponents have argued that new standards would force vehicle weight downward and cause a wave of new fatalities. This argument has been vigorously disputed, and an evaluation of its merits deserves at least a lengthy paper all its own. Recent analyses of vehicle size and safety12 and the variation of fatality statistics across the fleet13 indicate that vehicle size is more important than weight in affecting vehicle safety; vehicle design and safety equipment is more important than either size or weight; the early NHTSA studies suffered greatly from their failure to separate out the effects of weight and size, which are strongly correlated; and strong CAFE standards are unlikely to have significant adverse effects on safety as long as sufficient time is given to automakers to carefully redesign their fleets and as long as new standards are structured to avoid incentives for design decisions that might have adverse safety impacts (the new NHTSA light truck standards are a good step in this direction). However, it is quite certain that the argument about fuel economy standards and safety is not dead and will be vigorously argued in any future debate on new standards.

How Ambitious Should New Standards Be?

It would be useful if there were a way to calculate an optimum level for fuel economy standards. Unfortunately, due to the complexity of the market and conflicting interests, there is no such method. Instead, it may make sense to try a few different approaches to setting new standards to get a broad perspective for what options might be open to policymakers.

“Cost-effective” Standards

A common method of identifying fleet targets for a new standard is to identify a fuel economy level that would create fuel savings over the vehicles’ lifetimes that would be greater than the added cost of fuel-saving technologies. For example, the U.S. National Academy of Sciences (NAS), in a 2002 study of fuel economy standards,14 identified “cost-effective” fuel economy gains of 12 to 27 percent (depending on vehicle size) for passenger cars and 25 to 42 percent for light trucks assuming gasoline prices at $1.50 per gallon (the values clearly would be higher at today’s fuel prices). NAS researchers arrived at the targets by establishing baseline vehicles and theoretically adding, one by one, a series of fuel-saving technologies in order of their cost-effectiveness (highest first), until adding the next technology on the list would cost more than would be saved in reduced fuel consumption.15 Using standard economic methods, future fuel savings were “discounted” to the present. The Office of Technology Assessment16 and others used similar methods. This method is useful for getting a general sense for what is achievable by available technologies, but it has several problems. First, the method treats the problem as if it had only two variables, technology cost and fuel savings. In this formulation, the vehicle designer and purchaser are simply deciding whether adding fuel economy technology to a vehicle is worth the cost in fuel savings. In reality, however, all fuel-saving technologies are dual purpose: They can be used to save fuel, or they can be used to gain attributes like better performance, larger size, more luxury, or even greater safety without having to use more fuel. Thus, an engine improvement that allows more power to be squeezed out of an engine can lead to a more powerful vehicle without increasing engine size or a more fuel-efficient vehicle with a smaller engine and the same power. Or, the vehicle designer can compromise and get some of each—more power and better fuel economy, but less than the maximum possible for each. The box below illustrates the tradeoff between fuel economy and other vehicle features. Vehicle purchasers attach real value to the attributes that “compete” with greater fuel economy for the benefits of efficiency technology. Consequently, although some in the environmental community have suggested that fuel economy standards can be cost free when fuel savings outweigh technology costs, the costs associated with asking consumers to forgo valued improvements in other vehicle attributes should be part of the consideration of new standards.

A second concern with the method is that there is strong evidence that the great majority of vehicle purchasers simply do not perform even rudimentary analysis of the tradeoff between higher first cost and fuel savings over time17; in other words, the method by which analysts estimate “reasonable” levels of fuel economy improvement bears little relationship to how vehicle purchasers actually value fuel economy. Further, when consumers respond to surveys that ask direct questions about how they value fuel savings, their answers imply that they want any added purchase cost to be repaid within just a few years. If translated into potential fuel economy savings, this criterion would yield very little improvement. For example, NAS did an alternative analysis of fuel economy potential using a three-year payback as a criterion. The average improvement was estimated to be -3 to 3 percent improvement for cars and 2 to 15 percent for light trucks.18

A third concern is that this method has tended to focus only on currently available technology and generally fails to account for likely improvements in technology performance and cost over time and the development of new technology that conceivably might play a significant role during the time period of the analysis (if this is 10 years or more). Such a focus leads to conservative results, although these factors are hard to quantify.

Finally, the targets identified by the “cost-effective” method depend on fuel prices over the lifetime of the vehicles, which is uncertain, as well as on the discount rate chosen to represent the value of savings in the future, which is a variable figure; estimates of technology costs, which are hotly debated; and whether or not the value of externalities such as climate change damages and energy security costs, the magnitudes of which are highly contentious, should be included in the calculation.

“Top Runner” Method

The Japanese avoided these debates over cost-efficiency details by setting standards based on the idea that vehicles representing the “best in class” of the current fleet—weeding out vehicles that that are anomalous in performance or that have especially expensive technology—can be exemplars of what the average vehicle should be in 8 to 10 years. With these “top runner” examples, policymakers identified a series of fuel economy targets for vehicles in different weight classes and represented a 23 percent increase (over 1995 levels) in new gasoline-fueled passenger vehicle fuel economy by 2010 (assuming there would be few changes in average vehicle weight over the period). Although this method, or at least the Japanese version of it, is conservative in that it ignores the potential for newer technologies (such as hybrid electric/combustion drivetrains) to become more common through cost reductions, it does provide another potential fuel economy target that can inform the ongoing debate. Further, one can use the method to extrapolate into the future by conjuring up a vision of a future top runner vehicle—that is, the best mass-market vehicle that could be available a number of years in the future—and calling for the fleet average several years later to achieve the same fuel economy as the top runners.

The U.S. Environmental Protection Agency (EPA) has performed top runner analyses for the new 2006 U.S. car and light truck fleet,19 answering the question: What would the fuel economy of the new fleet be if the current fleet were replaced by the best four vehicles in each size class20); the best dozen vehicles in each size class; and the best dozen vehicles in each inertia weight class? The answer is that the car fleet would be 17 to 20 percent more efficient, and the truck fleet would be 14 to 24 percent more efficient. However, the fleet would be slower (for the largest boosts in efficiency, cars would take 10.2 seconds to go from 0 to 60 miles per hour (mph) versus the actual fleet’s 9.5 seconds, though the higher-efficiency trucks would actually shave one-tenth of a second off of their times); trucks would move sharply away from 4-wheel drive; the share of hybrid drivetrains would grow sharply, from 1.6 percent to 14 percent for cars and from 1 percent to 36 percent for trucks (but only 5 percent for cars and 12 percent for trucks for the next best case, with only a 1 mpg loss in fuel economy); and many automatic transmissions would be exchanged for continuously variable transmissions and manual transmissions. Unfortunately, this mixing of the effects of efficiency technology and utility-oriented vehicle attributes limits the usefulness of this type of analysis in setting standards—but the analysis nevertheless can offer a useful added perspective if interpreted cautiously.

What might the “top runners” be in the year 2020? Over the next 10 to 15 years, large and small changes in the technology embedded in cars and light trucks could bring about an improvement in fuel economy greater than 50 percent for a best-in-class vehicle with a conventional drivetrain and perhaps as much as a doubling in fuel economy for such a vehicle with a hybrid drivetrain. This assumes, of course, that the technology is used primarily for fuel economy rather than for performance and other attributes. To understand fuel-saving technology and the potential for improving it, it helps to understand a bit about why vehicles need energy and power and how they obtain it. This is discussed in the box below.

The major part of industry’s focus on raising fuel economy has been on the powertrain, but vehicle load reduction can play an important role. As noted in the box at the end of this article, reducing vehicle weight through sophisticated design and use of enhanced materials (such as high strength steels, aluminum, plastics, and composites) has considerable leverage on vehicle efficiency because weight reduction reduces inertial loads and rolling resistance losses. The U.S. Department of Energy’s FreedomCAR program has established the ambitious goal for 2015 of reducing the weight of the vehicle structure and subsystems by 50 percent.21 However, over the past decade, a considerable portion of the weight reduction potential of structural redesign and materials substitution has been used for improving vehicle stiffness and structural strength rather than for reducing weight. These attributes yield consumer benefits in better crash protection and a more solid feel that is highly valued by vehicle buyers. Assuming that some further gains in these attributes will be sought, weight reductions of 20 percent or so may be a more realistic estimate for what might be achieved by 2020, assuming strong pressure to maximize fuel economy. More drastic reductions might be possible if vehicle structures of carbon composites become practical for mass-market vehicles in this timeframe. A 20 percent weight reduction could yield a 12 to 14 percent fuel economy improvement if vehicle performance were unchanged.

Improvements in vehicle aerodynamics are hard to predict because aerodynamic drag is closely tied to vehicle appearance, and consumer acceptance becomes a key issue. By 2020, a coefficient of aerodynamic drag about 15 percent lower than the current fleet’s lowest may be possible for mass-market cars with side mirrors replaced by cameras, continued improvements in manufacturing tolerances for body panels, undercarriage smoothing, and careful design.

Reducing rolling resistance by improving tire design and materials is also possible. However, a tire’s design and materials affects not only its rolling resistance characteristics but also its resistance to wear and its handling performance. Because there is little publicly available information, achieving widespread use of tires with efficiencies about 25 percent better than today’s best mass-produced tires should be considered an educated guess at a realistic goal for 2020.

Engines have improved dramatically over the past two decades, and they will continue to improve. Recent presentations by a number of automakers and suppliers at the 2007 Society of Automotive Engineers World Congress presented a fairly unified picture of the potential future evolution of the gasoline engine. Currently, the most efficient gasoline engines have direct injection fuel systems with continuously variable valve timing on inlet and exhaust valves and variable valve lift and duration on intake valves. Because engine downsizing will yield significant benefits in efficiency, a “best-in-class” 2020 gasoline engine will probably use a turbocharger with variable geometry vanes. An alternative method to improve efficiency in larger engines is to shut down a third or half of their cylinders at low load. Improvements in emissions control should allow high air/fuel ratios (“lean burn”) that will further improve efficiency, although the availability of lean burn operation will depend on the stringency of future emission standards. Overall, efficiency gains of about 25 percent should be possible from engine improvements alone.

Advanced direct injection diesel engines currently are about 30 percent more efficient than gasoline engines of similar performance. Diesels will improve further with improved combustion chamber designs and higher pressure injection systems, but their gains relative to gasoline engines should shrink as gasoline engines become more diesel-like.

Hybrid drivetrains will certainly be an important part of the fleet in 2020, but the magnitude of their role is highly uncertain, dependent on fuel prices and on reductions in component costs. Hybrid sales have grown rapidly since the 1999 introduction of the Honda Insight, and May 2007 sales totaled a record 45,000 vehicles.22 In the near future, a variety of new hybrid systems, from simple stop-start mechanisms to the General Motors/Allison two-mode full hybrid system, will be introduced to the fleet. However, the more efficient systems currently can pay for themselves with fuel savings only if gasoline prices remain high and only for high-mileage drivers who spend much of their time in urban stop-and-go driving where hybrids maximize their efficiency advantage over conventional vehicles. The key to making them into a dominant technology is to substantially reduce costs, shifting to lithium ion or other energy storage technologies that may be less expensive than current nickel-metal hydride batteries (which have limited cost reduction potential because of high nickel prices), and reducing the cost of their expensive electronic controls.

Although plug-in hybrids (cars with larger batteries and motors that can fuel some of their daily miles with electricity from the grid) are not yet commercially available, they might begin to play a role in the new vehicle fleet by 2020 if their battery costs decline. Their batteries are considerably larger than those used in today’s hybrids, but battery costs will be not go up by the same multiple as their size. Also, as production ramps up, batteries will achieve substantial economies of scale. A new report by the California Air Resources Board23 projects that lithium ion batteries capable of 20-mile range (about 7 kilowatt hours of capacity) would cost about $5,000 each at a production rate of 20,000 batteries per year and less than $3,000 each at a production rate of 350,000 per year. However, the likelihood that these batteries can last a vehicle lifetime remains uncertain.

Although there will certainly be an argument about what a 2020 top runner midsize passenger car might look like, a reasonable guess, which assumes a very strong focus on fuel economy and a vigorous research and development program, might be as follows: full hybrid drivetrain, assuming battery and electronics costs are reduced sufficiently for hybrids to become fully mainstream; curb weight reduced by about 20 percent from today’s cars; rolling resistance of the tires at 0.006, compared to about 0.008 for today’s mainstream tires; aerodynamic drag coefficient of 0.22, compared to today’s best-in-class 0.26; downsized gasoline engine with direct injection, limited mode switching from Homogeneous Charge Compression Ignition to Atkinson cycle to Otto cycle depending on load, turbocharging and perhaps super-charging, and possibly camless valves; and automated manual transmission.

 A 2003 Massachusetts Institute of Technology study estimated that such a car would get about 60 (adjusted24) mpg compared to a 26 mpg car in 2001, a 130 percent improvement; a conventional counterpart, without the hybrid drivetrain, would obtain about 42 mpg, about a 60 percent improvement.25 Simulation runs from an ongoing Argonne National Lab study using the Powertrain Systems Analysis Toolkit (PSAT) model for a similar midsize car found fuel economy improvements of about 55 percent and 127 percent, respectively.26

Note that it will take a number of years before the top runner vehicle can become the average vehicle in the fleet. Although several of the technologies will likely have been introduced several years earlier in other vehicles, standards focused on this vehicle’s attained fuel economy should probably aim at least 5 to 10 years later, with the lower end of this range implying a higher degree of technical and market risk.

Adding It Up

The availability of NAS-style calculations of cost-effective fuel economy targets and visions of future top-runner vehicles will not add up to a certain view of a “correct” fuel economy target, but they are valuable in informing target decisions. For a more comprehensive view, policymakers need to combine the perspectives gained from these analyses with a careful consideration of how urgently society needs to combat climate change and the economic security problems associated with U.S. dependence on an unstable fuel supply. Policymakers must also carefully consider their views on consumer freedom of choice, because a future shift to faster acceleration capability and increased weight (associated with more size or other features) will significantly reduce the fleet’s fuel economy improvement potential.

The NAS-style calculation offers a way to get a sense for a conservative view of what an economically rational consumer might want if size or power were not important considerations for the driver, or if policymakers were determined to push the fleet away from the “performance race” that characterized the last 20 years. On the other hand, fleet targets might be more ambitious if automakers could promote smaller cars by emphasizing safety and comfort in their design. Similarly, sales of four-wheel and all-wheel drive vehicles, which suffer significant weight and fuel economy penalties, might slow as universal penetration of electronic stability control and traction control reduce the perceived safety and traction advantages of four- and all-wheel drive. Conclusion: the type of fuel economy improvement goal derived from an NAS-style calculation (roughly a 20 to 30 percent improvement over 10 years) may be a decent starting point for negotiations. Technological optimism and a strong sense of urgency in reducing oil use and greenhouse gas emissions would tend to push the goal upward; pessimism about trends in performance and other efficiency-reducing vehicle attributes and concerns about adequate lead times for some weaker automakers would tend to push in the opposite direction.

For a longer-term and less conservative perspective, projecting future top-runner vehicles provides a good view for what developing technology could do for fleet fuel economy. For 2025 or 2030, a doubling of passenger car fleet fuel economy and somewhat less for the light truck fleet (because towing requirements limit the benefit of hybrid drivetrains) would be quite possible assuming either strong reductions in the cost of hybrid drivetrains or simply the willingness to treat reduction in oil use and greenhouse gas emissions as societal requirements in the same way that reductions in emissions of criteria pollutants are treated. A more conservative goal of a 50 to 60 percent improvement would reflect less willingness to impose costs on vehicle purchasers, less technological optimism, or both.

An added consideration will come into play if it becomes important for the United States to make a strong shift to oil substitutes. The most straightforward substitutes are alternative liquids from unconventional oil sources (such as tar sands and heavy oil), natural gas, and coal. These will yield substantial increases in per gallon emissions of greenhouse gases, and large increases in vehicle efficiency will be needed to avoid sizable increases in total emissions. Biomass liquids can represent a strong alternative if they can be obtained from cellulosic materials, but they will provide a large share of U.S. requirements only if fleet efficiency is greatly improved. And hydrogen and electricity have severe onboard fuel storage problems that are likely to be solved only if less fuel (or less battery storage capacity) is needed—that is, only if overall vehicle efficiency is very high. In other words, greatly increased vehicle efficiency is a crucial requirement if the United States needs to move dramatically away from its dependency on imported oil.

The Structure of a New Standard

The economic impacts of a new standard will depend not only on the stringency of the standard (the mpg target) but also on its structure.

Opponents of fuel economy standards have long complained about the various market distortions that the standards appear to have created, including the virtual death of the station wagon and its replacement by minivans and sports-utility vehicles; the advent of very large SUVs whose weight put them outside the light-duty fleet and free from CAFE standards; pricing of some small car models among the “big three” U.S. manufacturers that appeared to be below production cost; and deliberate foreign sourcing of key components of some full-size cars and their inclusion into the import fleet. These distortions appear to have little to do with the stringency of the standards and much to do with their structure, particularly the separation of passenger cars and light trucks with very different fuel economy targets, the separation of domestic and import fleets, and the assignment of a uniform fuel economy target to every automaker regardless of the mix of vehicles they produce. For example, the car/truck separation, with light trucks having a much lower standard (until recently, 20.2 mpg versus 27.5 mpg for cars), produced a strong incentive for automakers to find a way to move their least efficient passenger cars into the light truck fleet. This incentive should not take sole responsibility for the rise of strong markets for minivans and SUVs, however—minivans turned out to be extraordinarily attractive vehicles for suburban families, and SUVs showed high profit margins for automakers.

Many of the problems of the current system can be overcome by eliminating separate domestic and import fleets (which are an anachronism in an age of multinational automakers); insuring that artificial weight ceilings do not allow vehicles to escape from compliance; and moving away from uniform standards to standards based on the attributes of each automaker’s fleet, as long as the attributes are reasonably related to vehicles’ fuel economy potential. The central idea of attribute-based standards is that they provide individual fleet targets to each automaker that reflect the degree of difficulty faced by that automaker to comply with the standard. This can greatly reduce a problem associated with the current standards: Manufacturers of small vehicles may be able to comply with the standard without any action to improve efficiency design and technology, while manufacturers of larger vehicles, or a mix of vehicles, may have to take strong measures for compliance. It also may allow combining car and light truck fleets because such a standard can shrink differences in the degree of difficulty of compliance—the primary reason for keeping the fleets separate.

The attribute most closely related to fuel economy is vehicle weight, and Japanese and Chinese fuel economy standards are weight-based standards (that is, automakers producing larger, heavier vehicles have lower fuel economy targets than automakers making primarily small, light vehicles). An outcome of moving to weight-based standards, however, is that automakers tend to eliminate weight reduction as a strategy for compliance because while reducing weight improves fuel economy, it also makes the vehicle’s fuel economy target more stringent, with no net regulatory benefit to the company. Because weight reduction can be an important component of fuel economy improvement, weight-based standards limit the degree of improvement that a new standard can demand.

In setting new standards for U.S. light trucks, NHTSA chose standards based on vehicle “footprint”—track width multiplied by wheelbase. Footprint is much less closely correlated with fuel economy than is vehicle weight.27 However, footprint is attractive as the basis of a standard because any tendency to increase either track width or wheelbase will be limited by the need to essentially redesign the vehicle (which is not the case with weight), and because increasing either of these dimensions would tend to be beneficial to vehicle safety. Wider track width will reduce a vehicle’s potential to roll over, and a longer wheelbase may provide more space for crash management and improve directional stability.

On-road versus Tested Fuel Economy

As currently structured, fuel economy standards will improve the tested fuel economy of the vehicle fleet. The actual on-road fuel economy of the fleet will tend to follow the direction of these tests, but with important differences that should be understood in considering new standards.

EPA discovered that their fuel economy tests28 gave fuel economy values that were considerably higher than drivers were actually obtaining and, using the data available at the time, reduced the city test result by 10 percent and the highway result by 22 percent for the fuel economy estimates actually communicated to consumers. The value calculated this way is the one that now appears on the window sticker of new cars and light trucks. However, even this adjusted fuel economy has proved to be optimistic for most drivers,29 especially as congestion has spread, highway speeds have increased, and air conditioning has become almost universal. EPA has now proposed to require the “window sticker values” on new cars to be based on a series of five driving cycles, some of which are driven with air conditioning on or at cold temperatures (20˚F), some of which duplicate driving that is considerably faster (up to 80 mph) and more aggressive (2.5 times the acceleration on the original tests) than the original two cycles. This new method is expected to reduce estimated city fuel economy values by an average of 12 percent (with a maximum drop of 30 percent) and highway values by 8 percent (25 percent maximum).30

Although there remain doubts about whether the new testing series will yield accurate results, they will at least take some account of measures that manufacturers can take to improve real world fuel economy but that will not make a difference on the formal two-cycle test. For example, improving the efficiency of the air conditioning system, insulating the vehicle, or adding special coatings to the windows to reduce heat gain during the summer will all improve actual fuel economy but will be ignored by the two-cycle test. In other words, energy-saving measures that will not count on the test will appear on the sticker as a benefit to automakers that include them.

If new standards are formulated, regulators could strengthen this modest incentive to include test-invisible technologies and designs by awarding credits toward satisfying the standards. Although it might seem more logical to simply change the official test-driving cycles to accurately reflect these factors, such a change is problematic without considerably more confidence in the new tests.

Maintaining Fuel Economy After the Sale

Discrepancies in the fuel economy on the sticker and the driver’s observed fuel economy occur because of failures in testing, but there are other reasons why a driver may come up with a different figure than a lab. Fuel economy standards generally have been aimed at new vehicles and have not tried to affect what happens to vehicles after they are sold.

Vehicle fuel economy can degrade significantly after a vehicle is sold. For example, poorly maintained vehicles will lose fuel economy through loss of engine efficiency or added friction in worn drivetrain components. Also, underinflated tires increase rolling resistance and, because the added resistance causes more tire heating, can adversely affect safety as well. Automakers have a strong incentive to install high efficiency tires to maximize reported fuel economy values, but replacement tires generally are less efficient than original equipment tires. There is no rating system for tire efficiency, and no way for the vehicle owner to know the added price in increased fuel costs of a less efficient replacement. 

In addition to maintenance issues, drivers commit other fuel conservation gaffes: Added weight from heavy materials left in the trunk add to inertial and rolling resistance losses, and vehicle body add-ons such as ski and bicycle racks add weight and reduce aerodynamic efficiency, leading to reduced fuel economy. Driving style greatly affects fuel economy as well. Aerodynamic loads grow with the square of velocity, so high-speed driving can be very inefficient. Rapid acceleration and a driver’s general failure to stay even with the flow of traffic can have adverse effects on fuel economy.

Technology requirements can address some of these issues. Requirements for automakers to incorporate tire safety warning systems should reduce the incidence of severely underinflated tires; however, the current requirements do not demand actual measurements of tire pressure, so mild underinflation is unlikely to be affected.

Efficiency requirements for tires may be regulatory overkill, but NHTSA or EPA could try to develop a tire efficiency rating and labeling system that communicates the likely value of excess fuel use over the tires’ lifetime. This value would have to be tied to the vehicle’s rated fuel economy.

Another possibility is to give a fuel economy credit to vehicles that incorporate real-time fuel economy indicators on their vehicles’ dashboards. U.S. and European studies have indicated that fuel economy improvements on the order of 10 percent or more can be obtained if drivers are aware of the effects of their driving style on efficiency and adjust their driving accordingly.31

Vehicle inspection systems tied to emissions and safety should tend to reduce some of the maintenance problems, but these inspections are limited in geographic coverage and may be a difficult sell, politically. And convincing vehicle owners to remove unnecessary material from trunks and dismantle detachable vehicle racks when not in use may be difficult, though it certainly is worth an information campaign to communicate just how much fuel these changes can save.


The process of developing new fuel economy standards is inherently too complex to do it justice here. The timing of standards was not discussed in depth, but timing is very important: Redesigning vehicles is a time-intensive and very expensive process that requires large engineering teams. For the industry to redesign the large part of its fleet will require about a decade, and automakers like to proceed with caution when introducing new technologies to avoid maintenance and operational disasters. Another issue not discussed here is the economic impact of new standards. In the past, economic analyses of proposed standards have tended to follow a common script: The industry and its consultants forecast huge negative impacts, and the environmental community forecasts large positive impacts. In most cases, the results flow primarily from the input assumptions, not from robust analysis.32 And as noted, safety has been and will be a crucial factor in negotiations about new standards, but the intricacies of the subject cannot be addressed here.

The decisionmaking process that will create new fuel economy standards is intensely political, as it should be. Scientific analysis can define the possibilities, but in the end the process is about trading off competing societal values—the relative importance of global warming and energy security concerns, the value of the free market, the ability of Americans to drive whatever kind of vehicle they want, and the value of future savings versus present costs. Scientists can inform this process, but they should not rule it. Further, as anyone who has watched this debate over the years knows intensely well, there are strongly variable scientific positions about all of the issues in the debate. What does seem certain, however, is that the extremes of the debate—that fuel economy standards do not work and do not save fuel, or that fuel economy standards can be cost-free—are both incorrect.

The extent to which fleet fuel economy can be improved is controlled not only by technology but also by consumer desires. Over the past 20 years, in the absence of more stringent standards, many fuel efficiency technologies have been widely disseminated in the fleet, but their potential to improve fuel economy has been totally cancelled by changes in vehicle attributes desired by buyers, especially increased performance and larger size. New standards might constrain trends to larger, heavier, more powerful vehicles, but vehicle manufacturers (through advertising and design decisions) and government and civic leaders (through their ability to inform and influence the public) have a strong role to play in the future of automotive design.

In the near term (10 years), fuel economy improvement goals of 20 to 30 percent seem to be a reasonable starting point for negotiations between government and industry, though higher values would be possible if the federal government felt that the urgency of energy security and climate change issues justified asking consumers to pay more for new technologies than they would likely save in future fuel savings. In the longer term, goals of 50 percent improvement and perhaps considerably more appear quite feasible, especially if adverse vehicle attribute trends are stopped and if progress continues in cutting the costs of hybrid drivetrains and other new technologies.

It has been more than 30 years since the original CAFE standards were put into place. It may finally be time to update and strengthen them.

Steven Plotkin is a transportation energy analyst with the Center for Transportation Research at Argonne National Laboratory who recently has specialized in analysis of transportation energy efficiency and new fuels. He has worked extensively on automobile fuel economy technology and policy as a consultant to the Department of Energy, and was a consultant to the National Research Council’s study, Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. He is a lead author on the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report on mitigating climate change. For 17 years, he was a Senior Analyst and Senior Associate with the Energy Program of the Congressional Office of Technology Assessment (OTA) and prior to that he was an environmental engineer with the U.S. Environmental Protection Agency. He is the 2005 recipient of the Society of Automotive Engineers’ Barry D. McNutt Award for Excellence in Automotive Policy Analysis. He can be reached at This article is product of Argonne National Laboratory, operated by UChicago Argonne, LLC, for the U.S. Department of Energy under Contract No. DE-AC02-06CH11357.


1. The White House, “Twenty in Ten: Strengthen America’s Energy Security,” policy initiative (Washington, DC, 2007).
2. General Accountability Office, Passenger Vehicle Fuel Economy, Preliminary Observations on Corporate Average Fuel Economy (CAFE) Standards, GAO-07-551T (Washington, DC, 2007),
3. The term “conventional oil” refers to liquid hydrocarbons that can be extracted by drilling to the rock in which it resides and pumping it to the surface, possibly stimulated by injecting water or carbon dioxide to push the liquid to the wellbore.
4. See H. Lee and D. A. Shalmon, “Searching for Oil: China’s Initiatives in the Middle East,” Environment 49, no. 5 (June 2007): 8–21.
5. Intergovernmental Panel on Climate Change (IPCC), Climate Change 2007: The Physical Science Basis. Contribution of Working Group I of the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (Cambridge: Cambridge University Press, 2007); and IPCC, Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II of the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, April 2007, (accessed 13 June 2007).   
6. Massachusetts v. EPA, 549 U.S., 1 (2007), (accessed 30 May 2007).
7. D. L. Greene, “CAFE or Price? An Analysis of the Effects of Federal Fuel Economy Regulations and Gasoline Price on New Car MPG, 1978-89,” The Energy Journal 11, no. 3 (1990) 37–58.
8. R. M. Heavenrich, Light-Duty Automotive Technology and Fuel Economy Trends: 1975 Through 2006, U.S. Environmental Protection Agency EPA420-R-06-011 (Washington, DC, July 2006).
9. K. A. Small and K. Van Dender, “Fuel Efficiency and Motor Vehicle Travel: The Declining Rebound Effect,” University of California at Irvine Economics Working Group Paper 05-06-03, Irvine, CA, March 2006, Opponents often cite old data to argue that increased driving will be much greater, with adverse effects not only on fuel use but also on pollutant emissions. Recent studies of the effect of driving costs on the volume of driving show that income gains and driving “saturation” make it highly unlikely that more efficient vehicles will spur large increases in driving.
10. S. E. Plotkin, “Fuel Economy Initiatives: International Comparisons,” in Encyclopedia of Energy, Volume 2, (Amsterdam: Elsevier, 2004) 791–806.
11. C. J. Kahane, National Highway Traffic Safety Administration, Vehicle Weight, Fatality Risk, and Crash Compatibility of Model Year 1991–99 Passenger Cars and Light Trucks, DOT HS 809 662 (Washington, DC, October 2003), (accessed 30 May 2007).
12. R. M. Van Auken and J. W. Zellner, A Review Of The Results In The 1997 Kahane, 2002 DRI, 2003 DRI, And 2003 Kahane Reports On The Effects Of Passenger Car And Light Truck Weight And Size On Fatality Risk, Dynamics Research Inc. report DRI-TR-04-02, March 2004, NHTSA Docket 16318-7.
13. M. Ross and T. Wenzel, An Analysis of Traffic Deaths by Vehicle Type and Model, American Council for an Energy Efficient Economy report number T021, March 2002,
14. National Research Council, Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards (Washington, DC: National Academy Press, 2002).
15. Note that this analysis looks at the marginal cost of technology, not the total cost. It makes no sense to pursue technologies until the total cost just equals the fuel cost savings, because the last few technologies might be actually reduce the savings to the consumer because they are not cost-effective.
16. Office of Technology Assessment, Improving Automobile Fuel Economy: New Standards, New Approaches, OTA-E-504 (Washington, DC: U.S. Government Printing Office, October 1991).
17. T. S. Turrentine and K. S. Kurani “Car Buyers and Fuel Economy?” Energy Policy 35 no. 2 (2007): 1213–23.
18. National Research Council, note 14 above.
19. R. M. Heavenrich, note 8 above.
20. There are nine size classes in both the car and light truck fleets.
21. For more information, see the FreedomCAR website. Fifty percent weight reduction would be available for use in a top-runner vehicle; the 2015 date does not assume that the new vehicle fleet could achieve such gains at this time.
22. See the Green Car Congress website.
23. F. R. Kalhammer, et al., Status and Prospects for Zero Emissions Vehicle Technology; Report of the ARB Independent Expert Panel 2007, paper prepared for State of California Air Resources Board, Sacramento, CA, 13 April 2007.
24. Adjusted fuel economy is about 15 percent less than the EPA test values for the combined city and highway tests.
25. J. Heywood, et al., The Performance of Future ICE and Fuel Cell Powered Vehicles and Their Potential Fleet Impact, Massachusetts Institute of Technology, Laboratory for Energy and the Environment, MIT LFEE 2003-004 RP (Cambridge, MA: MIT, 2003).
26. Personal communication with study principal investigator, 4 June 2007.
27. S. Plotkin, D. Greene, and K. G. Duleep, Examining the Potential for Voluntary Fuel Economy Standards in the United States and Canada, Argonne National Laboratory report ANL/ESD/02-5 (Washington, DC, October 2002).
28. EPA fuel economy tests involve operating the vehicle on a dynamometer (something like a treadmill for cars) while a driver uses the accelerator and brake to match a speed/time profile called a driving cycle. There are two profiles on the test, a relatively slow cycle designed to simulate city driving, and a faster one designed to simulate highway driving. However, partly because of the limited capabilities of dynamometers at the time the tests were designed, both driving cycles are “gentle” cycles with modest rates of acceleration and braking, and the highway cycle never tops 60 mph. The tests are conducted with heating, air conditioning, lights, and other accessories turned off, and the temperature is held at 68˚F–86˚F. To obtain an “average” fuel economy, it is assumed that 55 percent of driving is on the city cycle and 45 percent on the highway cycle.
29. Fuel economy is extremely sensitive to driving styles—how gently one brakes and accelerates, how much the driver anticipates speed changes and avoids unnecessary braking—and the type of driving one does. As a result, multiple drivers using the same vehicle model typically will get a wide range of fuel economy results. Other factors that affect fuel economy results are average temperature and accessory use. Fuel economy values typically drop substantially in severely cold weather, for example.
30. D. Edmunds, “Explained: 2008 EPA Fuel Economy Ratings,” 7 March 2007, (accessed 30 May 2007).
31. European Congress of Ministers of Transport/International Energy Agency, 2005: Making Cars More Fuel Efficient; Technology for Real Improvements on the Road (Paris: OECD Publishing, May, 2005).
32. Massachusetts v. EPA, note 6 above.
33. Office of Technology Assessment, note 16 above.

The Tradeoff Between Fuel Economy and Other Vehicle Attributes

Since 1987, the tradeoff between fuel economy and other vehicle attributes has allowed the U.S. fleet of cars and light trucks to add a staggering array of fuel efficiency technologies, including

 • supercomputer design of vehicle body structures coupled with new lightweight materials and higher strength steels;
 • significant improvement in aerodynamics and tires;
 • and new engine technology such as valves that adjust their timing and lift (degree of opening) in response to changing power demand, and fuel injection systems that can respond instantly to changes in cylinder conditions monitored by sophisticated sensors and controlled by more onboard computer power than was available in the lunar module.

However, the net effect of this technology on fleet fuel economy has been zero. Every bit of fuel economy potential represented by this technology has been traded away for other things. There is no ideal way to measure the impact of this tradeoff. There is only the answer to this simple question: How efficient would the fleet have been had it remained at the average acceleration performance and weight of the 1987 fleet? The U.S. Environmental Protection Agency has concluded that the tradeoff cost of the years 1987–2004 has been about 5.5 miles per gallon (mpg), or 22.5 percent, for the combined car/light truck fleet.1

Toyota’s stable of hybrid gas/electric cars shows this tradeoff explicitly. In the Prius, Toyota designers chose to use the hybrid technology primarily to increase fuel economy. They use a small, very efficient engine and use the added power of the electric motor to achieve performance similar to other vehicles in Prius’s size category, with much better fuel economy (city/highway fuel economy of 60/51 mpg versus 30/38 mpg for the smaller, conventional Corolla). In the Camry hybrid, the emphasis is still on fuel economy, but the designers chose to forgo downsizing Camry’s four-cylinder engine, creating performance a bit better than the hybrid’s conventional sibling (187 horsepower (hp) versus 157 hp) but with clearly superior fuel economy (40/38 mpg vs. 24/33 mpg for the conventional four-cylinder with automatic transmission).2 And in the Lexus GS450h, the designers pushed the tradeoff considerably more toward performance (0 to 60 mph in 5.2 seconds, versus 5.7 seconds for the GS350 with the same engine), creating an ultra-powerful luxury car with fuel economy comparable to or a bit better than a less powerful car of the same size (25/28 mpg versus 21/29 mpg for the GS350).3

1. C. H. Hellman and R. M. Heavenrich, United States Environmental Protection Agency, Light-Duty Automotive Technology and Fuel Economy Trends: 1975 through 2004, EPA420-R-04-001 (Washington, DC, 2004).
2. Toyota corporate website,
3. Lexus corporate website,

A Short Primer on Vehicle Energy Use

All fuel-saving technology is designed either to reduce the power needed at the wheels to move the vehicle and power to run accessories or to improve the efficiency by which the vehicle obtains power from its energy source—generally gasoline or diesel fuel.

Vehicles need energy to provide the power at the wheels to overcome the force of inertia when accelerating, the force of gravity when climbing a grade, and the forces of air drag and tire friction. Energy is also needed to power the accessories that maintain comfort (such as air conditioning and heating), provide entertainment (such as a radio or CD player), or enhance safety (such as lighting).

These forces vary a great deal with the type of driving. Air drag, for example, is high at high speeds on the highway and very low in the city. Tire rolling resistance, which varies directly with vehicle weight, is important in all kinds of driving. Inertial forces also vary directly with weight and are a function of changes in speed: They are low in smoothly flowing traffic and high if there is frequent slowing down and speeding up.

What this means is that weight reduction is an excellent way to reduce the energy needed by a vehicle, because weight is directly proportional to two of the three primary sources of energy use in driving (inertial losses and tire rolling resistance). A weight reduction of 10 percent can improve fuel economy by as much as 7 percent. Improving the efficiency of tires and aerodynamic performance by the same 10 percent is less effective but will still achieve about 2 percent increases in fuel economy for each, measured by the standard EPA fuel economy test.1

Improving the efficiency of accessories will also help improve fuel economy, although much of this improvement will not show up on the U.S. Environmental Protection Agency (EPA) test, which does not include use of heating, air conditioning, lights, or entertainment systems. A 10 percent reduction in accessory energy use could improve fuel economy by about 1 percent.

The other way to improve fuel economy is to improve the efficiency with which the vehicle translates fuel energy into power at the wheels. An average passenger car or light truck powered by a gasoline engine loses more than 80 percent of its fuel energy between its fuel tank and its wheels. The most losses occur inside the engine, through friction of air and fuel pumped through tubes and valves (“pumping losses”), friction of moving surfaces, heat losses through cylinder walls, loss of heat in the exhaust, and so forth. A wide variety of technologies and operating strategies can act to reduce these losses. For example, variable control of engine valves can reduce pumping and other losses, with valves with individual activators (“camless valves”) being the ultimate goal. And because the conventional Otto thermodynamic cycle used in gasoline engines is not the most efficient possible, the ability to shift to more efficient modes, at least under certain favorable load conditions, is being pursued vigorously by engine developers. Further, because engines are most efficient when the power from them is a large fraction of their maximum power, strategies to keep engines small (turbocharging or supercharging) or to allow them to behave as if they were small (shutting down cylinders when they are not needed) add to efficiency. Adding speeds to the transmission allows it to keep the engine operating nearer its most efficient mode.

Finally, a hybrid drivetrain, combining an engine with a battery and electric motor, saves energy through multiple means, including allowing the engine to be turned off during idle, allowing downsizing of the engine (since the motor boosts total power), capturing braking energy by using the electric motor to brake,2 and using the motor to power the vehicle when using the engine would be inefficient.

1. In each case, it is assumed that the improvements are used to increase fuel economy rather than to increase performance. For example, if vehicle weight is substantially reduced, a smaller engine would be used so as to maintain the same level of performance. If engine power is kept the same, the fuel economy benefit would be substantially lower.
2. The motor then acts as an electric generator, feeding electricity into the battery.

SubscribeBecome a Subscriber

In this Issue

On this Topic

Taylor & Francis

© 2018 Taylor & Francis Group · 530 Walnut Street, Suite 850, Philadelphia, PA · 19106