We Oppose Nearly Every Government Insurance Program. We Could Support One for Pandemics.
Over the eight years of our existence, the R Street Institute has offered critiques of government-backed entities that provide insurance or reinsurance for flood, crop, windstorm, earthquake, auto, workers’ comp, nuclear energy liability and terrorism. We have warned that such entities almost inevitably misprice risk, foment moral hazard and displace private capital.
Given that, and given recent discussions in public policy circles and on Capitol Hill about the potential to create a federal program for “pandemic insurance,” one might expect we would be leading the opposition.
We are not. In fact, we think the idea has merit.
The outbreak of COVID-19 has left the economy, both domestically and globally, at the precipice of what appears to be sharpest correction in recorded economic history. Already, we have seen 3.28 million first-time unemployment claims for the week of March 21, four times as large as the previous weekly record. James Bullard, president of the Federal Reserve Bank of St. Louis, has projected U.S. gross domestic product could fall by a staggering 50 percent in the second quarter, representing a $2.5 trillion drop in income.
And yet, insurance coverage for business interruption (BI), a core component of every firm’s risk management toolkit, is almost completely unavailable for businesses that have been forced to shut down, often by government order, due to the coronavirus outbreak. Standard commercial property policies require evidence of physical loss—such as from a fire, flood or even terrorism—to trigger a contingent business interruption claim.
Even coverage for closures in response to military or civil authority orders typically require a physical loss trigger. To eliminate any confusion, most policies—particularly those written in the wake of the 2002 to 2004 epidemic known as Severe Acute Respiratory Syndrome (SARS)—also carry an exclusion endorsement explicitly barring coverage for losses arising from communicable disease.
There’s no mystery why insurers would eschew extending BI coverage to pandemic events. It isn’t just the size of the losses, but their extreme correlation that make them nearly impossible to insure. It simply isn’t possible to diversify away from a global pandemic. They hit every geography at once, nearly every sector at once. Reinsurers might be slightly better positioned to manage the risk, but they too, would face extreme correlation of claims across geographies, as well as across product lines that include life, health, disability, various casualty and liability lines and, perhaps most notably, workers’ compensation.
Compounding the problem, the risk isn’t limited to the underwriting side of an insurance company’s balance sheet. The very fact that pandemics cause huge interruption in economic activity means that even insurers’ notably conservative investment portfolios are left decimated. Ratings agencies Standard & Poor’s and Moody’s have begun what are expected to be several rounds of downgrades in response to coronavirus, stripping some issues of AAA status and moving other investment-grade bonds to junk.
To be sure, there may be outlier cases where policies are found to provide business interruption for the COVID-19 pandemic. Some policies lack explicit exclusions. Civil authority clauses will be scrutinized. Some claims will point to contamination of premises as a form of “physical loss.” Already, policyholder lawsuits have been filed by the Oceana Grill restaurant in New Orleans and Oklahoma’s Chickasaw Nation casinos seeking declaratory judgments that their policies cover corona-related business interruption.
But even if courts find some individual claims meritorious, the inevitable response will be to rewrite contracts to make the exclusion clearer. This is a risk the industry does not want to write, does not believe it can write.
So, what then? If pandemic is a risk that private insurance markets cannot absorb, the alternative cannot be simply to allow hundreds or even thousands of businesses to fail, to see millions of employees laid off, commercial spaces shuttered and personal and professional relationships go to rot.
It falls to government to step into the breach.
Obviously, that’s precisely what government is doing right now, most notably with the unprecedented $2 trillion recovery package negotiated in Congress this week. But ad hoc responses are less than ideal. Better still would be a permanent government solution.
The devil, obviously, is in the details. The early efforts on that front are, to be frank, not terribly encouraging. The New Jersey Legislature currently is considering a bill that would force insurers to extend business interruption coverage for COVID-19 closures, even where such claims are explicitly excluded in the terms of the policy. For its part, the bipartisan Problem Solvers Caucus has proposed ordering insurers nationwide to pay business interruption claims related to coronavirus.
Proposals to enforce coverage retroactively on claims for which no premium has ever been collected have prompted swift pushback not only from the industry, but from the National Association of Insurance Commissioners. The regulators issued a statement this week warning that such plans “would create substantial solvency risks for the sector, significantly undermine the ability of insurers to pay other types of claims, and potentially exacerbate the negative financial and economic impacts the country is currently experiencing.”
Nor can the problem be resolved, as some in Congress clearly are contemplating, simply by replicating the Terrorism Risk Insurance Act and replacing the word “terrorism” with “pandemic.” Under TRIA, insurers are required to offer terrorism coverage to their commercial property, liability and workers compensation policyholders. To compensate them for that requirement, the law established a $100 billion federal reinsurance facility for certified terrorist events that exceed a prescribed trigger and industry deductible, with the payments recouped via post-event assessments, rather than an upfront premium.
To its credit, TRIA’s backstop has worked to reassure the private market that losses from terrorism would be manageable, as underwriters have become more comfortable with the risk. Reauthorizations of the program in 2005, 2007 and 2014 each also made its terms less generous. The effect of inflation, over the longer term, will continue to be to make that $100 billion backstop smaller.
But while terrorism remains a difficult risk to underwrite, it is fundamentally different from pandemics. The terrorist attacks we have seen to date have been localized and have not had lasting macroeconomic consequences. While there are theoretical examples of terrorist events, like cyber-attacks on the energy grid or the financial system, that could pose the same kind of existential risk of ruin that pandemics do, we don’t know that TRIA would handle those particularly well either, because it’s never been used.
What we do know is that there is no evidence the private market has any appetite to underwrite business interruption for pandemics. Is it possible that there is some theoretical backstop sufficiently generous to change that view and get insurers to accept a mandate? It’s possible. But finding that structure would be politically fraught and administratively complex. And to what end? No policy goal is furthered by forcing the industry either to share in the upside profits or downside risk of business it doesn’t want to write.
The best case for conscripting the insurance industry to offer this coverage is that they have experience with how to price it. Indeed, that’s precisely the critique that I have of the structure of the National Flood Insurance Program, whose system of subsidies, grandfathering and inadequate risk segmentation has served to encourage development in areas known be flood-prone.
It would also be a good reason not to extend a pandemic risk backstop to lines of business like general liability and directors and officers, where the private market does offer coverage for exposures related the virus. Firms should have incentive to take proper precautions not to expose third parties to contagion. Subsidizing coverage for those liabilities would invite clear moral hazard.
None of that applies in the context of business interruption caused by a pandemic. In fact, this is a case where the usual incentives of risk-based private insurance markets not only do not serve the public interest, but are directly contrary to the goals of public health.
The principle of insurable interest requires that an insured and insurer must have aligned goals; they must both want to avoid loss. But in the context of a pandemic, a business avoids loss by staying open. An insurer writing business interruption coverage should assess risk-based premiums that are cheaper for firms most likely to stay open and costlier for firms that are most likely to close. That’s what a healthy and functioning private insurance market demands.
The firms that have shut down because of coronavirus or been shut down by government order did nothing wrong. And there’s nothing they could have done to avoid the fate that has befallen them. The sad reality is that the market is punishing them because they did the right thing. They shut their doors.
No company should be forced to choose between endangering the public and going bankrupt. That’s why the federal government should offer financial protection for companies that find themselves in these circumstances. For a small premium, priced as a percentage of the revenues a firm needs to replace, there should be a federal government program to provide business interruption coverage for firms forced to shut down as a result of a public health emergency. It can be called insurance, but its function would be less to transfer risk than to reward upstanding behavior and civic virtue.
Such a program would, no doubt, be subject to adverse selection. Without risk-based pricing, companies in the most exposed industries—transportation, hospitality, retail, food service—overwhelmingly would be the most likely to buy coverage. But unlike the case of flood insurance, that’s precisely what we want to happen. We want the kinds of businesses that are most likely to be vectors of transmission to also be most likely to have a safety net that would allow them to shut down.
Like other government insurance programs, we are certain this one would be flawed. There will be unintended consequences. It could run large deficits. It almost certainly would be subject to gaming, potentially even to fraud. There should be every effort to try to counter these effects as part of a deliberative drafting process. This is not a bill that should be rushed through in the current emergency. It is not for the current pandemic, but for the next.
There is a tremendous amount of social benefit that comes from the way private insurance markets manage, mitigate and transfer risk. But there are limits to what they can do. For some risks, you don’t need an insurance contract. You need the social contract.
Image credit: DimaBerlin