Human life is highly valuable. Basic economic reasoning therefore suggests that, given the risks of COVID-19 to vulnerable populations, we should be willing to withstand large economic costs to prevent the risk of substantial numbers of deaths. This is particularly true if most of those economic costs are temporary.
They highlight estimates suggesting that the estimated value of a statistical life (commonly around $9.3 million) and a quality-adjusted-life year ($129,000) tend to be high. They show that modeling from the CDC and Imperial College suggests a high loss of U.S. life if no action is taken to stop the spread of COVID-19. They work out the value of additional lost lives if no government action is taken to avoid the worst-case scenarios where hospitals are over-capacity, with shortages of ICU beds. They conclude that we should be willing to incur trillions, and potentially tens of trillions, of lost economic output to ensure these lives are saved.
Such simple analysis, though, largely tells us one thing: that inaction could be highly costly. Yet nobody is credibly suggesting today that we “do nothing.” What usually makes economists worth their salt is their ability to think on the margin, and to judge alternative realistic scenarios. Economists need to start being economists again.
Michael Greenstone and Vishan Nigam’s work, for example, estimates that social distancing measures alone (7 day isolation for those with symptoms, 14 day isolation for their family, and infrequent social contact with over 70-year olds for 4 months) could save up to 1.7 million U.S. lives over the next six months relative to doing nothing, providing $8 trillion in benefits.
That figure is sensitive to assumptions about the virus's spread and death rates taken from the Imperial paper. It would be much lower at $3.6 trillion if the peak of daily deaths was 60 percent lower. But, still, even this suggests we should be willing to tolerate medium-term losses of up to 16 percent of today’s GDP to save those lives. Though economists do expect GDP to collapse significantly in Q2, most expect it to rebound strongly whenever the public health issue ends. So, taking this social distancing action looks cost-effective in the medium-term.
Now, one can quibble with the assumptions of such analysis. My real problem though is that too few economists are then asking follow-up questions about the wisdom of additional policy:
- If this is broadly right, what are the net impacts of “nonessential business” closures or shelter-at-home orders? Are these more suppressive measures, in isolation, cost-effective?
- Are they even optimal from a public health perspective, given the larger economic pain and the likelihood of an infection peak if they are released abruptly?
- What if even stronger constraints on economic activity like this risk a financial crisis or widespread business failures that simple social distancing or contact tracing would not?
- After what length of time would such suppressive measures be cost ineffective?
- How does the potential for a vaccine change any of these considerations?
- Would any of these actions be bettered by throwing tons of resources simply into repeat COVID-19 and antibody testing for the whole population, combined with social distancing for those infected?
- Are there some industries caught up in the nonessential business closures we are seeing where the output costs of closures are high and the benefits in terms of reduced risk of infection low? Why aren't these excluded? Are their big behavioral benefits to be seen to be all in it together?
I don’t know all the answers. What I do know is that we seem to be locked in an inane debate about whether what we are doing is better than nothing. We should instead be focused on which policies minimize the combined long-term health and economic costs of this pandemic. Just because lives are valuable, and so action to save them is better than inaction, doesn’t mean specific lockdowns or shelter-at-home orders are optimal policy. We want to find the most cost-effective way of dealing with the public health crisis to enable economic activity to return to normal.
In fact, big errors of thinking abound on all sides in this debate, from the “whatever it takes” crowd right through to the “cure is worse than the disease” view. So here’s some questions you should ask when you read commentators or economists comparing economic costs and healthcare benefits of certain actions:
- Are they comparing realistic alternative scenarios, or just “a particular action” vs. “complete inaction”?
- What assumption for death rates are they using, given this parameter is highly uncertain? (Hopefully greater testing will improve our knowledge in the very near future, and could help avert prolonged economic mistakes).
- How much do they value a human life? Does their assumption vary with age, rather than just presuming all human lives are of equal value?
- Are they seeking to account for a fairly comprehensive measure of costs and benefits for different scenarios? For example, potential lung damage for survivors of COVID-19, mental and physical health costs of shutdowns, risks of economic contagion, and the economic consequences of widespread bankruptcies?
- Do they correctly recognize that it’s likely that many producers and consumers would still restrict their activity to avoid the virus before an effective treatment, vaccine or herd immunity is developed, even if government policy changed?
- Over what time period are they comparing the value of any lives saved against GDP lost? It would be wrong, for example, to compare the value of lives lost against the decline in GDP for this quarter, given economists fully expect GDP to bounce back strongly as things begin to normalize. In fact, some historical analyses find public health policies with bigger short-term costs can produce better long-term economic performance.
- Are they comparing apples with apples? Lots of economists seem to be comparing the value of lives saved against GDP lost. But the value of human life is a welfare measure - it incorporates the benefits an individual expects to derive from his or her own life, including leisure time, friendships, and consumption. GDP, in contrast, just measures lost output. Shouldn’t we also account for the lost value of our liberties? Casey Mulligan has valiantly tried to assess broader losses to our economic welfare from lockdowns, such as not being able to participate in our preferred leisure activities, or experiencing worse schooling. These are real costs too.
Yes, all this makes analysis more complicated. But policymakers are currently making huge decisions fundamentally altering the health and economic well-being of all of us. Economic cost-benefit analysis is made for trying to compare the cost-effectiveness of different policy options. It's time to move the debate on from straw man discussions of "action" vs. "inaction" to assess what’s next.