Skip to Main Content

As governments and communities struggle to figure out the best way to deal with the novel coronavirus, a major concern is how to balance the risk to lives versus the risk to livelihoods. In short, how do we balance the public health versus economic consequences of this pandemic?

The answer comes from a widely used economic tool, and it is telling us that extreme measures are warranted.

It is difficult to get a sense of the direct and indirect economic impact of a human life — and it may seem callous to do so. But economists use a concept called “the value of a statistical life” (VSL) to think about the trade-off between the risk of dying and money. For example, how much more are you willing to pay for upgraded safety features on a car?


Estimates of the value of a statistical life typically come from analyzing how much more must be paid to achieve incremental reductions in the risk of death. How much more, for example, do you have to pay workers to take a job with a given risk of death? Or how much more would you be willing to pay for a safety feature, say a new and improved airbag, that reduces your risk dying in a car crash?

VSL estimates are used widely in the U.S. and other countries to consider the costs and benefits of health, safety, and environmental regulations. Putting a dollar value on human life, however distasteful, can help examine the cost-benefit trade-offs of important decisions such as how much we should be willing to damage the economy to prevent the spread of Covid-19.


To get a handle on this, we considered data from the Centers for Disease Control and Prevention. According to the New York Times, the CDC projected that if no actions are taken to prevent the spread of Covid-19, between 160 million and 214 million American will be infected, with between 200,000 to 1.7 million people dying from it. Given what we know so far, the majority of those deaths will be among people over age 50. This is important because the value of a statistical life decreases with age.

If the average age of those killed by Covid-19 is 60, value of statistical life estimates put the cost of each death at approximately $5 million. Using the CDC’s most dire estimate, the economic cost of lives lost if we do nothing is $8.5 trillion, or just under half of U.S. gross domestic product. This does not account for the costs of suffering for those who get sick but do not die, lost productivity due to illness, or direct medical expenditures.

To be sure, we aren’t doing nothing. The economic costs of combating the Covid-19 epidemic are already large and mounting. Small businesses, restaurants, hospitality, travel and many more face large losses amid the shutdowns and lockdowns. The stock market has plummeted. Schools in the largest districts around the nation are closing their doors and kids, despite their parents’ best efforts, are missing out on valuable learning and socialization. The economic consequences of school shutdowns are difficult to gauge but could be long-lasting if we cannot make up for lost time.

But, as made clear by the value of a statistical life estimate, these costs must be weighed against the exceedingly high economic costs of not containing the virus. We should all be prepared to make significant economic sacrifices now to minimize the harm of this evolving crisis. Furthermore, we should provide assistance to help those who are disproportionally affected by such measures weather the current storm and get back on their feet once the epidemic has passed.

Now is the time to use the U.S.’s immense resources to contain Covid-19 and mitigate its health and economic consequences for all Americans.

Mireille Jacobson is a senior fellow at the USC Schaeffer Center for Health Policy & Economics, and an associate professor at the USC Davis School of Gerontology. Tom Y. Chang is an associate professor at the Marshall School of Business at USC.

  • Let’s do a quick back of the envelope with more realistic worst case numbers. Average age of death = 80, $2M economic value, 10% or roughly 35M people infected in US, burdened mortality rate 0.3% (confirmed case mortality rate currently 1.5%, assume 5x more people will have virus that are never tested as they are asymptomatic or minor symptoms, this is 3x seasonal flu) gives us 100k dead or $200B in value. As mentioned in other comments this should be looked at as market value not GDP. There is no economic justification for shutting down the economy for Covid-19 outbreak.

  • The article states, “Using the CDC’s most dire estimate, the economic cost of lives lost if we do nothing is $8.5 trillion”.

    Point #1: Read through the other comments on why this is an overestimate by orders of magnitude (assuming the most dire estimate instead of a realistic estimate, assuming age 60 instead of a realistic age, etc.).

    Point #2: We’ve already destroyed more than $8.5 trillion in U.S. assets in response to coronavirus, and we’re on track to destroy far more. The total U.S. stock market began the year at ~$38 trillion, and is down ~26%, so there’s ~$10 trillion destroyed. Add in debt markets, housing prices, job losses and everything else to that and it’s far more.

    Oops. The overreaction to coronavirus has already caused far more human misery than coronavirus itself ever will.

  • So let me get this straight.

    According to this article, if we do nothing about the virus and let 1.7 million (less than 1/2 percent of the population) people pass away (many of whom are over 60 and beneficiaries of Medicare or Social Security), then we lose about half our GDP ($8.5 trillion) in outcome immediately? This isn’t correct and you know it isn’t. This is all premised on the brief Aldy article you cite.

  • This is an argument for suppression rather than mitigation. Suppression seeks to stop the virus in its tracks whereas mitigation would likely still produce the loss of life listed above, but would allow businesses and schools to be open while practicing other prevention measures such as social distancing and special management of 70+ population.

    The other commenters’ points about the average age are all valid and good, but the article also misrepresents the capabilities of what is possible at this late juncture. This is not Wuhan and we do not have centralized police/surveillance state that China is. China was able to lockdown Wuhan when they had 400 confirmed cases in a much more draconian way than would be permissible in any democracy. They had effective, top-down party infrastructure by which they could funnel resources towards the problem and a quiescent, collectivist (and repressed) population that accepted this fairly willingly.

    None of these factors are present in the United States. We have a relatively weak central government which controls no top-down healthcare structure. We have 11 million undocumented migrants who are, rather justifiably, going to be terrified of seeking treatment under our current president. And we have a highly individualistic, skeptical (some conspiratorial) rural population that has been effectively dissuaded that this is a serious issues for 2 months, and not just something to derail the president’s election.

    Our federalist structure is ideal for organic experimentation, but a distributed and variable patchwork of health systems means that an effective suppression based locked down is not likely to be viable long-term. This will be readily clear in the coming weeks as the virus spreads rather rapidly from urban to rural areas of our country. Mitigation methods retain value, but we are past the point of effective suppression that can be sustained economically and achieve the desired end. The lack of testing is a further hindrance in this effort, but maybe not even the most consequential one.

    Beyond this, the economic choice presented that frames it in % of GDP is highly misleading. These are lives’ values not their GDP equivalents. Perhaps these lives are worth $8.5 trillion perhaps, but probably not given the average age is likely to be higher than 60 as other commenters have noted. Even if they were, this is not an $8.5 trillion GDP loss, but rather more like losing 8.5 trillion in assets. This is much less consequential and it’s possible that the deadweight loss from suppression efforts actually wind up approaching this figure, which is likely inflated.

    Lots of flaws all around unfortunately.

  • Another flaw: The economic impact is not just that “Small businesses, restaurants, hospitality, travel and many more face large losses amid the shutdowns and lockdowns,” but that many people (some estimate 20%) will be unemployed after such a shutdown. Also, many businesses will have gone out of business and not be able to continue afterwards.
    On medical terms, we are already seeing now that the panic and shutdowns are meaning that people are not willing to come to work at hospitals or even supply businesses, meaning that the companies supplying life-sustaining equipment to hospitals are not able to get them there, even though they are available. A shutdown makes this way, way more difficult as the care necessary for serious cases simply won’t be there.

  • Average age of 60? Seriously? So far, the average age is closer to 80.

    I suggest the authors revise their article by also doing their calculation with a more realistic average age.

  • OK, so if you believe that 1.7 million of the sickest Americans are worth half of GDP, then how much do you believe the other 327.7 million Americans are worth?

    The other half of GDP?

    Or would you assign a value of at least $5 million to each one of them too, in which case you’ve made up a fantasy world where we have more than 96 GDPs to spend.

    • Agreed. These numbers don’t even pass the back of the napkin math test. I hope this is just a reflection on the poor quality of education at USC itself and not an indication of the overall instruction level in this country. Unfortunately, I fear the latter which would go a long way to illuminating how we found ourselves in this current situation.

  • The authors don’t give exact numbers for how much an estimated impact of shutdowns or current measures will cost, which makes it more of a one sided argument.
    Using the “worst case” scenario of 1.7 M, how many deaths occured every year in the US before the COVID-19 hit, for age 60 and up? Taking numbers from statista and CDC, one would get at least above 1 M, yet the US economy seems to have survived so far.

  • In my view, the analysis described in this article is (mathematically) flawed. Even if the average age of those killed by Covid-19 was 60, an assumption that probably (significantly?) underestimates the average age, the average person killed by the disease does not have the average life expectancy for that age (which is over 80) , but rather a much lower one. I wouldn’t be surprised if doing the calculation correctly leads to the $8.5 trillion being replaced by a number smaller by a factor of ten or more.

  • Isn’t the average age likely to be much higher than 60? The most recent information I could find said only one death (at 50) in the US under the age of 60. What happens to the VSL if average age is 70 or 75?

Comments are closed.