How Private Insurers Regulate Public Police

Policing is on a lot of minds these days. Liability insurance is probably not. But insurance bears on policing far more than one might think. Municipalities nationwide purchase insurance to indemnify themselves against liability for the acts of their law enforcement officers. The policies are broad, often covering punitive damages and intentional acts like assault and battery and discrimination. This arrangement creates a potentially serious moral hazard problem, but it also positions insurers, in the name of “loss prevention,” to act as “private regulators” of police activity. In my latest article, “How Private Insurers Regulate Public Police,” I present original qualitative empirical research describing how insurers do just that.

Here are some examples; the paper has many more details. First, as part of their loss-prevention programs, insurers work closely with the police to promulgate and update departmental policies. (Do you want to know how to conduct a strip search without violating the Constitution? Travelers Insurance has a pamphlet on that.) Depending on the degree of expertise among insurance personnel, they sometimes bring in outside consultants to do this—usually retired officers—or subsidize a turnkey policy-writing service.

Second, insurers support and influence police training efforts. They provide video libraries and online training systems, and even do some classroom instruction. They also subsidize the use of otherwise prohibitively expensive use-of-force virtual reality simulators. Early academic research has found that these simulators increase the number of preventative actions officers take and decrease the number of unjustified shootings.

Third, insurers audit the agencies they insure. They review internal documentation and do site visits and ride-alongs. My favorite anecdote was from an insurer out west who said that she sends insurance representatives to visit “cop bars” incognito to listen to the local gossip.

Fourth, insurers create incentives for police departments to seek accreditation from a recognized accreditation agency. This is a way of outsourcing the policy review and auditing functions I just mentioned, as the accreditors perform those functions as part of the accreditation (or re-accreditation) process.

Finally, some insurers put pressure on agencies to “correct” or even terminate officers they view as disproportionately heightening liability exposure. There are examples of police departments terminating officers from the beat all the way up to the chief in response to insurer demands.

The carrots and sticks that drive municipalities to cooperate in these loss-prevention initiatives are the availability and pricing of coverage, which both affect the public fisc directly and educate covered agencies about the likelihood they’ll be hit with embarrassing and politically problematic lawsuits. In extreme cases, municipalities have shut down their police forces after their insurers pulled coverage.

One implication of my findings is to suggest an unconventional form of police regulation. Specifically, once we understand that insurers regulate the police, we can see that the state might regulate the police by regulating insurers. In other words, we can use insurance regulation to shape the terms (and prevalence) of police liability policies, which, in turn, should influence police behavior. What might this look like? I should caution that mine is the very first attempt in legal scholarship to survey and assess the market for police liability insurance. It’d be wise to have more data before making significant changes to insurance regulation. So, for now, take these as illustrative examples rather than concrete reform proposals. With that in mind, here are three ideas:

First, my research suggests that insurers struggle to regulate effectively their most diminutive customers—the so-called “commodity clients.” These small municipalities are abundant. And they pose a number of challenges for insurers’ loss-prevention programs. Because the premiums these municipalities pay are relatively small, it is often infeasible for insurers to discount rates enough to compensate for the expenses of loss prevention. Nor is it cost-effective for insurers to individualize loss prevention or engage in the monitoring necessary to link premiums to care. If claims are infrequent, moreover, there may be no substantial base of losses from which insurers can experience-rate their policies in a statistically valid way. Insurance regulators might do two things to help. First, they could require small municipalities to pool their risks and resources before purchasing insurance on the commercial market. (Many small municipalities, but certainly not all, do this already.) This would make some additional loss-prevention measures cost-effective, even if individualization may remain challenging. Second, regulators might require small municipalities to carry a deductible or self-insured retention, which forces them to share in all losses. At present, some insurers write “first-dollar” police liability policies for small municipalities. This practice undermines the capacity of insurance to promote loss prevention; it’s also a bad use of premium dollars from the perspective of the municipality (i.e., taxpayers) because the main function of insurance is to spread the risk of losses that policyholders cannot bear well themselves.

Second, as I discuss more in a separate work in progress, certain types of police misconduct are difficult for insurers to regulate, even when they have the financial incentive to do so. The best example is police misconduct that leads to wrongful convictions, like using suggestive eyewitness identification procedures, faulty forensic evidence, false informant testimony, and false or coerced confessions. Wrongful conviction claims can be tremendously costly, but they’ve got two tricky features that make them a bear for insurers. First, they have a long tail—a large temporal gap between the occurrence of misconduct and the filing of a lawsuit. This creates uncertainty that makes the risk difficult for insurers to price and manage. Second, they’re relatively infrequent—they’re what’s called a “high-impact, low-probability” risk. Research shows that insurers misunderstand how to predict these events. Sometimes they seem to discount the risk altogether, while other times they fixate on the magnitude of recent losses and exaggerate the risk of the next major loss.

So how might insurance regulation help? Because wrongful convictions are relatively uncommon, insurers lack the experience base necessary to do experience rating well. As far as I can tell, however, they are not feature rating based on the risk characteristics known to affect the likelihood of wrongful convictions. For example, insurers could—but, to my knowledge, do not—raise rates for agencies that fail to videotape interrogations (to help avoid false or coerced confessions) or use double-blind lineup procedures (to reduce the danger of erroneous eyewitness identifications). Insurance regulators could work with state attorneys general or other law enforcement experts to devise a list of risk-related features that underwriters should (or must) consider when setting rates. It may be wise as well to institute “sub-limits” that cap the amount of coverage available for wrongful conviction claims, which signals the insured municipalities about the gravity of the perceived risk.

Insurance regulators might also insist that insurers write coverage for wrongful convictions on a claims-made basis. Police liability coverage traditionally has been written on an occurrence basis, meaning it covers liability for acts taken during the policy period, regardless when a suit that stems from those acts is filed. Policyholders tend to like this feature, but it makes pricing very difficult, especially for long-tail risks. Pricing occurrence coverage requires insurers to estimate today all the eventual results of this year’s activities. Claims-made policies, in contrast, cover liability for claims filed during the police year for injuries caused in the past. The insurer need not predict long-term claim exposure, which means that it can price its coverage more confidently. Better pricing, in turn, usually means better control of risk.

My third and final example is the most sweeping but also the simplest to state. As best I can tell, most municipalities do purchase insurance. But the very biggest cities do not—they typically “self-insure.” This may partly explain why we hear so little about police liability insurance—because both academics and the media tend to focus attention on policing in places like New York and Chicago. But note that the problems that small municipalities pose for insurers—a limited experience base and low-dollar premiums that can’t support much loss prevention—are absent for the major cities. If subsequent research finds that, in the aggregate, police liability insurance reduces police misconduct, maybe insurance regulators should mandate insurance for all police operations? Or issue a “soft mandate” that requires the purchase of insurance or proof of an adequate in-house loss-prevention program? A mandate wouldn’t be without drawbacks—or controversy—but if our policing problems continue at today’s pace, it seems like something that should be on the table.

The preceding post comes to us from John Rappaport, Assistant Professor of Law at the University of Chicago Law School. The post is based on his paper, which is entitled “How Private Insurers Regulate Public Police” and available here.