INFORMS Open Forum

  • 1.  On returning to work

    Posted 03-27-2020 11:58

    Your Money or Your Life

    Mr. Trump has recently argued that we need to get our economy moving by Easter if we are to avoid an economic disaster.  He argues that the flu and traffic accidents kill a large number of people too and that we just cannot afford to maintain social distancing for a prolonged period of time because of CVID 19.  But, medical professionals tell us that, to minimize the peak number of deaths, we need shelter in place and avoid all but the most needed interactions with other people.  In other words we must choose, "your money or your life."

    For some the choice is clear, Lieutenant Governor Dan Patrick is willing to chance his "survival in exchange for keeping the America that all America loves for his children and grandchildren."   But, some of the rest of us would like a more clear picture of the situation to assess these odds.  How should we as a society decide when, how much, and how to increase economic activity in the presence of COVID 19?  It may seem that that these decisions are unique to today's choices but, in fact, they are common choices that society makes regularly.  When we design and build a traffic interchange we trade off cost, convenience, and probabilities of accidents and death.  We might not make these decisions in a well informed way, but we make them nonetheless.  And when we make these decisions we telegraph the relative importance that we place on life and money.

    Cost – Benefit analysis is sometimes done badly. But, done well, it can deal with decisions that involve uncertain outcomes that include risks of death and even different attitudes about the relative importance of life and economic wellbeing now and in the future.  See our 2006 article in the Annals of Operations Research. Of course these analyses are not easy and they require a careful enumeration of the alternative courses of action. But they can be done and, when decisions may cost millions of lives and trillions of dollars, they must be done.  Indeed, it is crucial that our government subject such important decisions to careful analysis.

    While I do not have the data to do these analyses it is rapidly becoming available.  From what is available several things are clear.  First, the situation is different in different parts of the country. The correct policy for New York City is unlikely to be the correct policy for a low population county in Missouri. Second, the situation in dynamic. As the number of cases and their rate of increase changes the policy should too. Thus the recommendations made by Imperial College of London's COVID-19 Response Team which call for monitoring the rate of infections and adjusting the level of suppression in the area as appropriate.  Third, there are a wide array of policies that should be explored, likely with dramatically different results. The choices are not just do nothing or close down the economy.  For example some news out of New York is that 96% of the deaths from COVID-19 have been to people (young and old) with preexisting conditions.  Perhaps at some stage an approach is to continue to have those with preexisting conditions shelter in place while many others return to work.

    "Benefit-Cost Analysis Using Data Envelopment Analysis." N. K. Womer, M.-L. Bougnol ,J. H. Dula, and D. Retzlaff-Roberts, Annals of Operations Research (2006) 145:229–250.

    Norman Womer
    University of Missouri-St Louis
    St Louis MO