Sam Peltzman on Regulation

Episode Summary

Executive Summary: Russ Roberts interviews Sam Peltzman about his influential work on regulation and unintended consequences. Peltzman argues that safety and drug rules often change incentives, causing offsetting behavior, delays, and higher costs that can blunt or even reverse intended benefits. The conversation also covers why such regulations persist despite evidence and how market, legal, and private certification mechanisms can sometimes substitute.

Main Topics: Automobile safety regulation and risk compensation (Priority: 5/5): Peltzman explains his 1975 study of safety mandates like seat belts and collapsible steering columns, arguing that lower perceived accident costs can lead drivers to take more risks, offsetting benefits and shifting harm to non-occupants. Reaction to Peltzman’s auto safety findings (Priority: 4/5): The paper triggered intense hostility from the safety establishment, which saw it as a threat to the rationale for regulation and resisted the idea that drivers adjust behavior in response to safety devices. Drug regulation, efficacy standards, and delayed innovation (Priority: 5/5): Peltzman discusses the 1962 FDA amendments requiring proof of efficacy, arguing that longer testing and higher R&D costs delay beneficial drugs, reduce innovation, and can cost more lives than they save. Regulatory persistence and political economy (Priority: 5/5): The discussion explores why regulations survive: rational ignorance among voters, strong bureaucratic incentives, industry accommodation, and the difficulty of evaluating counterfactuals. Wealth effects vs. regulation in improving safety (Priority: 4/5): Peltzman contrasts mandated safety changes with the natural increase in consumer demand for safety as societies get richer, arguing many improvements would have happened anyway without regulation. Private and market alternatives to regulation (Priority: 4/5): He outlines how reputational incentives, tort liability, and private certification systems can help ensure safety and efficacy without government control, especially in drug markets.

Key Arguments: Safety regulation can reduce harm directly but also lower the perceived cost of risk, inducing behavior that offsets some or all of the intended safety gains. In the auto case, fatalities for occupants fell, but deaths of pedestrians, bicyclists, motorcyclists, and people in other cars rose, leaving little or no net reduction in highway fatalities. Drug efficacy regulation improves information and screens out bad products, but the dominant effect is delaying beneficial drugs and raising development costs, which can increase total mortality. Much of the apparent improvement in safety over time would have occurred even without regulation because rising wealth raises consumer demand for safety and induces market-led innovation. The correct evaluation of regulation is counterfactual: compare outcomes with regulation to what would likely have happened without it, not just pre/post trends. Voters tend to be rationally ignorant about counterfactual effects, while regulators and industry groups have stronger incentives to preserve existing systems and avoid disruptive reform. Private mechanisms—reputation, legal liability, and certification—can discipline producers and partly replace government regulation in many settings.

Data Points: Auto safety study period: about 10 years - Peltzman evaluated the first wave of automobile safety regulation roughly a decade after implementation. Subsequent research duration: 30+ years - He describes the literature following his auto safety paper as spanning more than three decades. Drug testing timeline: 10 years plus - Peltzman says bringing a new drug to market now takes about a decade. Drug development cost: about $1 billion - He cites the cost of bringing a new drug to market under FDA regulation. FDA delay effect: 4 to 5 years - He says many drugs that would otherwise be available are delayed several years by regulation. Drug approval innovation effect: dramatically lower rate - He says the rate of innovation dropped sharply after the efficacy requirement. Speed limit example: 55 miles per hour - Used as an illustration of misleading before/after comparisons in evaluating regulation. AIDS policy response: terminal patients - He describes pressure to relax testing rules for AIDS drugs because terminal patients could not wait for full-length trials.

Pivotal Quotes: "The price of an accident had gone down." — Sam Peltzman: He summarizes the core mechanism behind risk compensation in auto safety regulation. "This kind of regulation kills more people than it saves." — Sam Peltzman: His blunt summary of the net effect of drug efficacy regulation when delay and reduced innovation are included. "The correct comparison is not did it go down absolutely. But did it go down relative to a world without regulation?" — Russ Roberts: He frames the counterfactual method needed to assess regulation properly.

Implications: Listeners should expect well-intended regulation to create incentives, delays, and unintended harms. Policymakers need counterfactual analysis, and industries may benefit from more flexible, market-based, or private certification approaches.