This is a an interesting piece by Tim Requarth on the development of the concept of risk compensation in the 1970s, and the insidious effects on social policy this almost always false theory continues to have.
Risk compensation is the brainchild of University of Chicago economist (uh oh) Sam Peltzman, who came up with the kind of silly-clever hypothesis so beloved of ambitious academics looking to get paid for defending the social status quo. Basically, the idea is that safety measures and the regulations that require implementing them are bad, because they make people careless.
What if requiring cars to have seat belts in them — not requiring people to actually wear them, mind you; this isn’t Communist Russia — actually increased the number of traffic fatalities, because people would drive more recklessly? What if child safety locks on medicine bottles just led parents to leave uncapped pills lying around everywhere? What if requiring skiers to wear helmets actually made skiing more dangerous because people would attempt runs beyond their competence to the point that the safety effects of the helmets would be cancelled out? What if giving poor people money just makes them poorer in the long run because it makes them even lazier than they were already? Etc. etc.
This is what Albert Hirschman identified as the “perversity thesis” in his The Rhetoric of Reaction: you may think you’re helping by requiring rich natural and corporate people to spend a little money on making their products safer and poor people slightly less poor, but ACTSHULLY it makes the products more dangerous and the wretched refuse of our teeming shores more irresponsible. So it’s better for everybody if the rich people get to keep their money.
It turns out that this thesis, so delightful in every way to the Lords of Capital, is rarely, to use an old-fashioned term, true:
Peltzman’s findings gave serious quantitative ammunition to the anti-regulatory enthusiasm of the 1970s.
Subsequent analyses of Peltzman’s work, however, found it riddled with errors. Other researchers showed his model couldn’t predict traffic fatality rates before regulation. As one critic wrote in 1977, Peltzman failed to perform even “rudimentary checks on the validity of his model.” Decades of traffic data now leave little doubt that, overall, safety regulations have indeed reduced traffic-related fatalities. These days you would, with good reason, not even consider getting behind the wheel of a car that did not have working seat belts. . . .
Safety measures are not, of course, inherently beneficial. The effectiveness of a precaution that is going to be widely mandated should be studied—human behavior is complex and unpredictable. There are certainly examples of safety measures that don’t quite live up to the hype; anti-lock brakes, for instance, may have had no overall effect on fatal crashes (though it’s difficult to say risk compensation has anything to do with that). At best, risk compensation is something that happens at the level of the individual but rarely, if ever, fully offsets the social benefits of an effective safety regulation. At its worst, risk compensation is just kneejerk libertarianism masquerading as fundamental insight into human nature.
Requarth points out that the ideological underpinnings that gave rise to the idea of risk compensation in the first place — worship of laissez-faire economic arrangements as the answer to every possible social problem, and the closely related belief that all those problems should be addressed at the level of individual choice rather than in terms of the social structures in which those choices are made — are still very much with us.
He suggests that one factor in the initial resistance on the part of some public health officials at the beginning of the COVID pandemic to telling people to wear masks was essentially a kind of zombie risk compensation mentality:
We’re seeing the same cultural dynamics play out during the COVID-19 pandemic—even as the people making the arguments are different, and are making them for different reasons. Yes, the infamous hesitation of the CDC and WHO to recommend masks at the outbreak of the pandemic had many causes (including discounting aerosol scientists’ work suggesting that SARS-CoV-2 was transmitted through the air, and protecting the supply chain for health care workers). But a key cause is very simple: The authorities didn’t trust the public. They didn’t trust the public not to use masks as an excuse to leave their house willy-nilly; they didn’t trust the public not to use masks to ditch other protective measures such as hand-washing or physical distancing. In a confusing and fast-moving environment filled with new information, it was all too easy to lean on this flawed model of human psychology. But it probably cost lives as mask recommendations were delayed for precious weeks in the spring of 2020. When empiricism did weigh in, it became clear that masks reduced symptomatic infections.
That’s an intriguing claim, although one that needs more evidence to support it. We’re now seeing similar arguments about rapid COVID tests: are their fairly high error rates going to cancel out their value?
When it comes to rapid tests, experts fearful of risk compensation may be missing this bigger point. Sure, the tests may encourage riskier behavior when it comes to COVID. Sure, people may use the tests incorrectly sometimes. And it’s true that a false negative on a rapid test just before a wedding or before school could result in spread that wouldn’t have happened if everyone had just stayed home. It is well worth considering how to reduce those instances, by educating the public on how to use the swabs and making it easy for us all to access high-quality test kits. But it is too much to ask that the tests eliminate risk. Life is not just about staying safe by avoiding everything. It’s about balancing COVID risk with the very real downsides of staying inside all day. People need to work, to socialize, and kids need to go to school. In a sense, the point of these measures is to allow for a small amount of risk compensation. Masks, vaccines, and rapid tests let two things to be true at once: Individuals can take more risks to do what they like, and society stays safer.
Whatever relevance the (mis)use of the concept of risk compensation may have to the COVID epidemic, the larger point here is an important one: Despite all the screaming about the “woke” university, universities, like all other institutions in our increasingly plutocratic society, are under enormous pressure to tell rich people exactly what they want to hear, and they often oblige this desire, whether innocently or corruptly. (I was surprised that Requarth didn’t mention the closely related idea of moral hazard, so beloved of glibertarians who consider Manchester England in 1850 to be the ideal social setting.)
Anyway, the demand for theories that comfort the comfortable is going to remain as vigorous as that for cocaine.