In August 2020, the Inspector General for the US Department of Transportation announced it has initiated an audit of the National Highway Traffic Safety Administration’s oversight of Federal Motor Vehicle Safety Standards. An IG memorandum regarding the audit stressed the importance of ensuring that all new vehicles and equipment meet federal safety standards. Further noting that “more than 36,000 people died in motor vehicle crashes in the United States in 2019,” the IG broadly stated that it would review “NHTSA’s FMVSS process,” suggesting a potentially wide scope of review that could include the extensive process through which NHTSA develops and adopts auto safety standards. Seemingly confirming this intention, the IG’s memorandum concludes, “Our objective for this. . . .audit is to assess NHTSA’s efforts to set and enforce Federal Motor Vehicle Safety Standards” (emphasis added).
The same day, the US Department of Justice issued a report proposing significant changes to the federal Administrative Procedure Act (APA), the foundational and overarching statute that has long provided procedures and standards governing agency promulgation and application of federal regulations. The report contends that, for several reasons, substantial revisions to the APA are necessary, and provides ideas and proposals for such systemic change.
Together, these two announcements provide an occasion to consider a broader overall review of federal auto safety regulation. While an IG audit may not be the appropriate vehicle or forum, the time is ripe for a general, thoughtful, and balanced official review of US auto safety regulation. Among other things, such a review could generate recommendations for regulatory system changes to better address potentially transformative new motor vehicle systems and technologies.
Time for an objective general review of the auto safety regulatory system?
The IG announcement did not further explain the context or impetus for its “self-initiated audit.” However, the existing federal auto safety regulatory regime, and NHTSA’s development, application, and enforcement of safety standards, have been the subject of divergent criticisms from industry, Congress and other government agencies, and a variety of policy, advocacy, and interest groups. For example, DOT and NHTSA have begun to explore regulatory alternatives and initiatives to reduce “regulatory barriers” to motor vehicle innovation. Those efforts have largely focused on revising existing conventional vehicle standards and tests that may impede the commercial introduction of certain autonomous vehicles and innovative equipment and designs. Response to those initiatives has been mixed, including industry support and opposition from some members of Congress and safety advocates.
At the same time, the National Transportation Safety Board and others have criticized NHTSA for failing to issue new mandatory standards and regulations to govern automated and autonomous vehicles and testing. NHTSA’s handling and oversight of the complex ongoing recall of potentially lethal airbag inflators manufactured by the former supplier Takata – the largest recall in US history – has also been controversial. The pace of some of NHTSA’s regulatory activity is also controversial. For example, it often takes a number of years for NHTSA to develop and issue a single new or revised safety standard (FMVSS).
These and other significant auto safety regulation debates and challenges have roots in the structure, requirements, and limitations of the federal auto safety regulatory system, established 54 years ago by the National Traffic and Motor Vehicle Safety Act. By many relevant measures, the Act has been a tremendous public safety success. After five decades of experience and motor vehicle change (including today’s unprecedented technological developments), the regulatory system is due for review and potential revisions to continue that success and aid in advancing modern auto and highway safety.
Existing regulations often do not fit emerging technologies, equipment, and functions
If there is a consensus among industry stakeholders and observers, it is that existing motor vehicle safety regulation and standards are, at best, an imperfect fit for many new and developing automotive technologies, systems, and modes of operation. Increased vehicle automation and self-driving and connected cars hold real promise for transformative improvements in road and motor vehicle safety (including potentially saving tens of thousands of lives annually in the US alone), mobility, human productivity, vehicle emissions and even land use. But existing regulations are often a poor fit or even illogical when applied to such emerging technologies, and trying to fit a square “peg” innovation into round regulatory “holes” can unnecessarily impede safety and performance advances. Simultaneously, those same complex technologies threaten to outstrip the capacity of the existing regulatory system and the federal auto safety agency to expeditiously generate and enforce meaningful and effective safety standards.
For example, the current safety standards system (FMVSS) is designed primarily to set parameters and objective tests to assure the safety of vehicles’ physical equipment, components, mechanical controls and occupant protections. It is not designed for the different-in-kind challenge of assuring the safe performance of the entire actual driving task (otherwise performed by human drivers) by an autonomous driving system, which involves a very complex set of interdependent tasks and functions and dynamic decision-making under a nearly infinite variety of circumstances and conditions, all coordinated and controlled by computers and computer code. Timely development of effective, comprehensive and universally applicable objective safety standards and tests for a vehicle’s performance of the full driving task may not be possible under the current regulatory approach and framework. A new and significantly different alternative policy or regulatory approach (to the existing FMVSS approach) will likely be necessary to provide appropriate safety assurance for AVs allowed to be commercially deployed on public roads.
Even elements of AVs that may be amenable to conventional standards and tests face significant regulatory obstacles under the current system. Numerous existing federal standards effectively prevent reasonable, and practicable deployment at scale of fully autonomous vehicles without provisions for human drivers. For example, vehicles manufactured for regular use on public roads without operative conventional brake and accelerator pedals violate FMVSS and are thus illegal. Because it often takes years for NHTSA to issue new and revised FMVSS, the time it would take to make necessary revisions to dozens of such provisions poses a significant potential obstacle to deployment of such vehicles.
NHTSA, private standards organizations, and others are presently exploring and developing potential new standards and tests (for automated driving systems and other new technologies) that might be made to fit within the structure, requirements, and limitations of the existing regulatory system. Those are valuable and important efforts, but most work within the existing regulatory framework and do not consider material revisions to the overall regulatory structure, tools and authorities, or approach, in order to better and more efficiently promote and protect motor vehicle, highway, and public safety.
Appropriate regulatory revision should further promote safety and foster innovation
Rapid and unprecedented automobile technology innovation make this a propitious time to seriously consider balanced, informed, and workable changes to update our half-century-old US auto safety regulation system. Such changes should seek to improve safety regulation by making it more flexible and adaptable to new technologies, systems, applications, equipment and even fundamentally different modes of operation; incorporating advances in safety science, analysis, methodologies, and techniques in the last five decades; and reducing unnecessary regulatory delay, barriers, and burdens. The goal should be to identify revisions to the regulatory system that will promote and enhance motor vehicle and roadway safety while fostering innovation. Based on a careful review, elements of the system that work well, perhaps including some of the present system for regulation of conventional vehicles, could be retained. Where appropriate, new regulatory approaches, methods, and standards could be recommended for better-fitting and more effective methods of safety assurance and regulation, particularly for new and materially different technologies and modes of operation.
Both legislative and administrative changes in motor vehicle safety regulation likely will be necessary to appropriately address the challenges and opportunities presented by today’s conventional, automated, and autonomous vehicles, equipment, and technologies as well as future innovations. That process should start with a comprehensive review of existing motor vehicle safety regulation, its strengths and weaknesses, and potential alternatives and changes. A balanced and objective official review of the present system, structure, and implementation of US auto safety regulation, including concrete and workable recommendations for any regulatory system changes that may be appropriate, could point the way for future policies to help save lives and promote motor vehicle safety and innovations with widespread social benefits. Such an important review cannot come too soon.
 See Modernizing the Administrative Procedure Act (US Dept of Justice, Aug. 11, 2020). The Report’s contributors share a similar general perspective on contemporary executive agency regulation and its perceived shortcomings, and the Report’s proposals for change reflect that particular perspective. It does not canvas a broad spectrum of different views and ideas on how the APA might be revised to meet modern regulatory challenges and opportunities. Citation to the Report here is not meant to endorse its views and ideas on the APA or the contemporary role of administrative agencies, but rather as an indicator of the high level of interest in the Executive and Legislative branches and other quarters in making changes to foundational administrative law, processes, and functions.
 In evaluating the regulatory actions of DOT/NHTSA, it is useful to remember that their actions, options, and available tools are directed and limited by authorizing statutes and regulations. They may act only as directed, authorized and constrained by their regulatory authority, and broad or systemic regulatory change often may be accomplished only through statutory change, which is the province of Congress. Presumably an IG audit will also largely be confined to evaluation of the agency’s performance of its statutory and regulatory responsibilities using available tools, and not broader questions of what other (de)regulation or different policies or tools might be desirable or appropriate to advance motor vehicle and highway safety and other relevant objectives.
 While there have been a number of amendments and additions to the Act in the intervening decades, the basic safety regulatory system and structure has remained largely the same since the 1970s. The most recent significant system revision was adopted in 2000 by the TREAD Act, which primarily created manufacturer “early warning” data collection and reporting requirements and expanded recall obligations.
 Compounding the regulatory challenges for these new technologies and modes of operation are overlapping state and federal regulatory jurisdiction and activities; lack of clarity as to which level of government is responsible for regulating what aspects of AVs and their operations; sometimes inconsistent state and federal policies and requirements; and limited coordination of federal, state, and local regulation and enforcement activity.
 In addition to the direct benefits of effective modern safety regulation of autonomous driving systems, effective and reliable safety regulation will likely be essential to foster public confidence and widespread adoption of autonomous vehicles that will be necessary to realize their full promise for broad and diverse benefits to society (starting with a quantum leap in motor vehicle and roadway safety).
 Greater participation by private companies and organizations in necessary research and proposed standards could significantly expedite the regulatory development process, including the development and promulgation of new regulatory standards and tests.
 Recent congressional efforts to revise auto safety laws and regulation have largely focused on regulatory exceptions and other incremental or temporary adjustments to the existing regulatory system and requirements (such as expanded exemption authority) in order to facilitate safe initial deployment of autonomous vehicles in the short term. They generally have not proposed broader regulatory system changes to address new and evolving auto safety challenges and opportunities in the longer term.