Much is being made of the U.S. Environmental Protection Agency’s announcement last month that it will consider anew the greenhouse gas standards for vehicle model years 2022 to 2025. This announcement, while generating controversy, was not unexpected. The mid-term assessment of the standards for the upcoming model years was, since its start in 2012, expected to be a thorough review of market conditions, available technology and economic considerations.[1]

The EPA’s program operates in connection the U.S. Department of Transportation’s Corporate Average Fuel Economy, or CAFE, program. Congress first enacted the CAFE program in 1975 in response to the Organization of Petroleum Exporting Countries, or OPEC, oil embargo.[2] The EPA program began more than three decades later, after the U.S. Supreme Court greenlighted the EPA’s regulation of automotive greenhouse gas emissions in Massachusetts v. EPA[3] in 2007 and the EPA issued its endangerment finding in 2009.[4] Because California has been allowed to set its own emissions standards, the auto industry faced an amalgam of potentially conflicting regulations governing the same aspects of performance.

The Obama administration addressed this situation by negotiating the One National Program, or ONP. Under the ONP, the federal agencies and California retained the pre-existing testing protocols but set new uniform standards.[5] The industry gained a national program and agreed not to challenge the standards in court.[6] In 2012, the ONP was extended to cover model years 2017 to 2025. Doubting whether fuel efficiency standards could be set up for that long into the future, the industry agreed to this extension only after the EPA, National Highway Traffic Safety Administration and California consented to a “mid-term review” of the standards for model years 2022 to 2025.[7]

Because NHTSA is statutorily limited to setting standards for no more than five years in advance, it published “augural” standards predicting what the standards could be if the agencies predictions proved mostly accurate. For its part, the EPA established standards subject to an independent review to determine whether the standards remained appropriate. The outgoing EPA administrator decided in January 2017 that the standards should be continued,[8] but the new administration subsequently initiated its own review. In early April, the EPA published Administrator Scott Pruitt’s official finding that the standards initially set in 2012 are no longer appropriate and that EPA will therefore conduct a new rulemaking in conjunction with NHTSA’s ongoing rulemaking to set CAFE standards for model years 2022 to 2025.[9]

Rollback or Review?

The notice published in the Federal Register, which according to the EPA is not a final agency action, is carefully constructed to ensure that all administrative legal requirements are followed.[10] In the notice, the administrator found that his predecessor’s earlier determination had been “based on outdated information.”[11] According to the document, “more recent information suggests that the current standards may be too stringent.”[12]

The findings made (and the administrator’s other public comments) suggest that the EPA has already concluded that some rollback of the standards is likely appropriate. Whether that “rollback” will occur and what the standards will be, however, are unknown until after the EPA completes a full rulemaking. The notice itself states that while the administrator has found the original standards to be “inappropriate,”[13] they will not be revised until after the rulemaking process in conducted in conjunction with NHTSA.[14] A separate, but related, fight is looming over the ongoing role of California in establishing its own, independent set of standards and programs adding layers onto, or creating potential conflicts with, the revised federal program.[15]

The CAFE Dilemma

Fuel efficiency regulation differs from most product regulatory programs because it does not impose a per se limit on any individual vehicle. An automobile manufacturer’s compliance is based on a complicated calculation of target fuel consumption values based on each vehicle’s “footprint” (a measure of vehicle size) weighted according to the manufacturer’s sales mix.[16] Compliance is based on the company’s overall fleet, which is known only after the model year ends and consumers have contributed to the determination through choosing which vehicles to purchase.[17] Consumer preferences reflect gas prices (when gas prices rise, there is more demand for fuel efficiency; when they fall, truck sales soar) and general economic conditions impacting the market for new automobiles.[18]

Because compliance depends in part on market conditions and consumer demand, it has long been difficult to impose standards reaching far into the future. In the 1980s, the government found it necessary to reduce fuel economy standards in response to unexpected economic conditions — a move that was upheld by the courts.[19] In 1994, the Clinton administration published an advance notice of proposed rulemaking suggesting setting CAFE standards for a number of years into the future.[20] At that time, minivans and SUVs were popular — giving rise to new vehicle platforms that both met consumer demand and allowed carmakers to take advantage of the relatively lower fuel economy standards applicable to “trucks” over “passenger cars.” Congress responded to the Clinton-era suggestion of setting longer-term fuel economy standards by imposing an appropriations freeze prohibiting any new CAFE standards.[21] The freeze was extended year after year, until it was finally lifted in early 2002 as part of the aftermath of the World Trade Center attack.[22]

The recent activity faces the same dilemmas as in the past. Whether cast as fuel economy or greenhouse gas standards, the influence of broader economic and market conditions makes it difficult for the government to predict what particular level of fuel economy or fuel consumption can be met over the long term. A major accomplishment of the Obama-era regulatory program was that the EPA and NHTSA were able partially to divorce the setting of the standards from the very short-term product plans of a small number of larger volume manufacturers, and instead have developed computer models to predict the ability of technology to advance and the relative societal costs and benefits that inform regulatory decisions. While NHTSA and the EPA have competing models, the advent of computer modeling itself greatly enhances their ability to make date-driven decisions about the automotive fleet in the time frame covered by the mid-term review — about three to six years into the future.[23]

Opportunities for Reform

The upcoming rulemaking need not only be about the stringency of the standards — or the role of California vis-à-vis the federal government. This rulemaking presents an opportunity also to take a look at the structure of the programs — and to revise the structure to reduce greenhouse gas emissions, to promote fuel economy and to advance motor vehicle safety.

This would not be the first structural reform of the programs. In 2005 to 2006, NHTSA reformed the light truck CAFE program by moving from a “corporate average” to an “attribute-based” compliance program.[24] The following year, Congress amended the CAFE statute to require “attribute-based” standards for passenger cars and for light trucks.[25] The EPA also subsequently applied the attribute system when developing its own greenhouse gas program.[26]

When amending the CAFE statute in 2007, Congress also for the first time allowed companies to generate over-compliance credits that could be transferred between their passenger car and light truck fleets (subject to some limitations) or traded to other companies.[27] The EPA’s program is even more generous because it allows unlimited credit transfers between the passenger car and light truck fleets.[28] These relatively simple allowances create an economic incentive to develop and deploy technology regardless of whether the company needs that technology to comply with the standards. Since allowing this flexibility, the number of companies needing to pay fines to comply with CAFE standards (a compliance option allowed mostly to accommodate companies with more limited fleet mixes) has dwindled.[29]

Similar structural reforms should be considered during this upcoming rulemaking. First, the agencies should consider adding support for crash avoidance technologies. Regardless of current views on the readiness of autonomous driving systems for use on the public roads, crash avoidance technologies are increasingly effective in promoting safety and reducing emissions. One estimate from 2009 found that approximately 25 percent of the traffic congestion in the United States is due to collisions which impede the flow of traffic.[30] It has also been estimated that eliminating traffic congestion can substantially reduce carbon dioxide emissions. While the industry has already voluntarily committed to the widespread deployment of automatic emergency braking,[31] and the regulatory programs already provide certain credits for stop/start technology,[32] the agencies should consider whether more gains could be had through the further encouragement of crash avoidance. In the current environment, there is no ongoing need to consider emissions reductions and safety to be in conflict. The regulatory programs should be able to adapt to technology that can simultaneously reduce collisions, congestion and emissions.

Second, the environment in which these regulatory programs are operating is already substantially different than that in which they were first developed. The fundamental structure of the CAFE and greenhouse gas programs is dependent on a market in which manufacturer’s sell motor vehicles to individual consumers for their personal transportation needs. That transportation model, however, may become quickly outdated. Shared mobility and transportation on-demand is prevalent throughout the United States, and manufacturers are experimenting with automotive subscription services. These changes are occurring now, and the upcoming rulemaking is an opportunity for the agencies to consider whether and how the structure of the regulatory programs should take into account modern trends in transportation.


Fuel economy (CAFE) and fuel consumption (greenhouse gas) standards depend not only on the deployment of technology, but also on a broad set of market and economic conditions. For that reason, it has been difficult throughout the history of the CAFE program to establish longer-term standards without the need for subsequent review and possible revision. The advent of computer modeling in recent years enhances the ability of the agencies to predict beyond the manufacturer’s very short-term product planning. But the midterm review originally contemplated in 2012, and now put into full swing by Administrator Scott Pruitt’s decision to conduct a full rulemaking in conjunction with NHTSA’s CAFE rulemaking, is largely consistent with the history of fuel efficiency regulation.

The upcoming rulemaking — and subsequent judicial challenges to it — will likely put much focus on whether the ultimate stringency of the standards for model years 2022 to 2025 and the various factors informing those standards (the rebound effect, the discount rate, predictions about electrification of the fleet, etc.) were more or less appropriate than those originally used in the 2012 rulemaking. The ultimate standards emerging from that rulemaking must be based on the data and science presented, and appropriate findings made through the rulemaking process. Subsequent court challenges will determine the extent to which the data supports the findings, and the findings support the decisions made.

This upcoming rulemaking need not be only about the stringency of the standards. The agencies should also consider whether and how the programs should change to respond to the adaptations in the way light-duty vehicles are being used in the U.S. and to incorporate technologies that can simultaneously reduce crashes, congestion and emissions. Politics aside, the agencies have an opportunity through this rulemaking to move beyond the conflicts of the past and to help structure a constructive vision of the future.