Reducing Coronavirus Fatalities: A Cost-Benefit Analysis
Q: From an economic perspective, what’s it worth spending to curb the pandemic? A: A lot.
So far as I know, nobody has really worried about doing a cost-benefit analysis of the epidemic. That’s probably a good thing. But it’s worth asking what such an analysis would look like. In other words, what would be worth spending to reduce the death rate based on cold-hearted economics?
To get a rough idea of the amount, we need to start with some picture of what the world would look like without such an intervention. According to the CDC, its modeling shows expected mortality between 200,000 to 1.7 million people. The lower end is about 2.5 times the number of deaths from one of the worst flu years in the recent past. The upper end is about the same as the 1918 flu pandemic, adjusting for the tripling of U.S. population in the meantime.
That’s assuming there are no major government interventions to halt spread. But that assumption is useful if we want to know the possible benefits of government intervention. We don’t know the right figure within this range, but an average between 200,000 and 1.7 million seems like a plausible estimate. That’s 850, 000 deaths.
Now let’s assume we have an intervention that would cut the death rate in half, which would mean saving 425,000 lives in our middle-of-the-road scenario. What is it worth to save that number of lives? There’s a well-developed economic methodology for answering questions like that.
To do cost-benefit analysis, regulatory agencies like EPA have to convert the benefits of life-saving regulations into a monetary equivalent. In effect, people act on average as if they valued their lives at $9 million. This number is extrapolated from studies of how people respond economically to smaller risks. Economists call this the value of a statistical life (VSL). Since people may not be fully informed about risks, or may not have free choices about whether to take risks, the $9 million figure is arguably too low. But it’s the number EPA economists use to attach a monetary value to the lives saved by a regulation.
Using the $9 million figure, we can calculate the value of saving 425,000 lives. If you do a little arithmetic, it turns out that the benefit of cutting the number of deaths in half is $3,825,000,000,000 or about $3.8 trillion, which is 18% of 2019 U.S. GDP. So in economic terms, it would be worth spending up to $3.8 trillion to save those lives.
Obviously, this analysis could be refined in various ways, including using more realistic measures of the benefits of various interventions. Calling this a “back of the envelope” calculation is an injustice to envelopes. It probably only suggests about the right order of magnitude.
Still, the bottom line is clear. If we can hold back the death rate by throwing more money at testing, medical treatment, and behavioral change, even the most steely-eyed accountant would say that’s a good deal.
Reader Comments
6 Replies to “Reducing Coronavirus Fatalities: A Cost-Benefit Analysis”
Comments are closed.
About Dan
Dan Farber has written and taught on environmental and constitutional law as well as about contracts, jurisprudence and legislation. Currently at Berkeley Law, he has al…
READ more
This is a case where the difference between using VSL and VSLY is important, given that most of the people likely to die from COVID-19 are elderly. In this article, I argue that we should be using VSLY more broadly: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3424332
This doesn’t necessarily mean we’re that close the margin at which the costs of pandemic interventions outweigh the benefits, but it does reduce the expected benefits of those interventions.
Obviously, this blog post is a very rough cut. There’s a lot of disagreement about the VSLY issue, but it would be worth doing the analysis that way as an alternative. There are also some serious costs omitted from the analysis. A substantial number of people will get sick enough to be on respirators for extended periods. They presumably have some willingness to pay to avoid that. There are also the direct costs of the medical care, plus deaths of people from unrelated causes because the system is overloaded. No doubt other things on both sides of the balance sheet. Very complicated!
Super helpful start, Dan, thank you. I think the point that we are not anywhere near investing “too much” as a society in addressing this pandemic is a great takeaway. Two additional thoughts: 1. As you mention in your comment, sickness matters too. I think, even more than that, that this may be the rare analysis where—despite a significant mortality rate—the morbidity impacts are similarly significant, given % of serious complications and longevity of those complications. So worth including a morbidity estimate on the back of the envelope too. 2. This is a helpful exercise for telling us how many total societal resources might be reasonably spent on addressing this # of risks (if they were faced independently), but—and I’m not suggesting you don’t know this—doesn’t tell us about the marginal benefit of any particular policy, or even that any particular policy is cost-justified. Or in other words it tells us that we should be willing to invest in accessing high-hanging as well as low-hanging fruit…but it doesn’t tell us if getting more fruit will actually reduce coronavirus risks. So identifying the strategies with the biggest leverage on marginal benefit seems like a separate key question, and questions about the cost justification of specific policies will still have to be dealt with individually by policy (or at least, would be so dealt with in an ideal world that we seem not to live in)…
Arden – As I think you anticipated, I completely agree about the marginal benefits issue.
Dan, as a physician, I’m glad someone is finally looking at this. Kudos for taking up this difficult subject. I also agree that these are gross ballpark estimates. However, I come to the opposite conclusion because I believe you are missing several important factors. These include 1) Herd immunity. Once a large portion of the population is infected, they will almost certainly be immune, providing herd immunity to much of the population, so N is not equal to the US population. The CDC is remiss in not factoring this. 2) This devastated economy will result in increased homelessness in the US and other 1st world nations, which in turn, results in a definite, though small, uptick in deaths in all age ranges. 3) 3rd world economic devastation will result in starvation, armed conflicts, and deaths due to other diseases that go unresolved due to lack of resources. 4) Most importantly, you are counting only money SPENT, not the extreme multiplier represented by economic losses, tax losses, permanently closed businesses, etc. These will be associated with ongoing huge economic losses for years, if not distinct expenditures.
In addition, now that we have set this precedent, what will we do next time we have a 4x average flu season, a new SARS or MERS, a new coronavirus? Spend another $3.8 trillion? How about if we encounter such a disease every 3 years? Unsupportable insanity to do so.
Unfortunately, the lack of cost/benefit analysis has come from the idea that we can afford anything, since that’s been relatively true on earth for many years. It’s not true now, and there comes a time for triage, both medical and financial.
I’m a conservative, retired Orthopedic Surgeon. I’m not concerned for myself medically or financially, but I am concerned for my kids and grandkids, and the people of the world.
Dr. Johnston — That’s a fair point, and I would be the last to downplay the economic impact of some of the measures certainly being taken to control the spread of the virus. The current projections don’t show that they would be big enough to outweigh the benefits of saved lives, but that could change. And as you say, there are multiplier effects. Of course, we don’t know what economic effects an uncontrolled virus would have either. There are just a lot of imponderables.