[Editor’s note: This story originally was published by Real Clear Policy.]
By James Broughel
Real Clear Policy
COVID-19 is forcing rethinking in a lot of areas. Work, dating, and entertainment have all been upended by the coronavirus. In government, once-trusted federal agencies like the CDC and FDA have lost clout due to their poor handling of the crisis. Economists and other policy analysts are experiencing a shakeup as well. They too are having to do some hard rethinking, as certain old ways of doing things just don’t make sense in the pandemic era.
The tradeoff between reopening the economy and losing lives to the coronavirus has forced the public to think about an uneasy question: How much is a life worth? It’s something economists have long pondered, though usually safely in the pages of academic journals, tucked away from public view. With the coronavirus, however, these practices are receiving an unusual level of scrutiny. As the public gains unfiltered access into the economist’s workshop, there is a lot they find they don’t like.
For example, when economists prepare a cost-benefit analysis that evaluates the good and bad outcomes a policy like an air pollution regulation or a vehicle safety requirement will have, they will often resort to attaching a dollar value to each life expected to be saved. This number can then be compared to the expected public and private costs of the regulation to gain some insight into whether it will benefit society overall.
The most common way of doing this is by using a metric called the value of a statistical life, or VSL, which is based off of how much money people are willing to spend to reduce risk. This metric has many shortcomings.
One example is that the same life can vary dramatically in value depending on the magnitude of the risk involved. If a risk is fairly small, like the chance of dying by electrocution, it might be valued with one number. But if the risk is larger, like some projections of the danger Americans face from coronavirus, that same life might receive a very different value.
Thus, if Policy A will save a single person from being electrocuted in a bathtub, and Policy B will save a single person from coronavirus, the VSL can lead to two very different valuations. This is true even if both policies save the same person, even if the form of death is equally unpleasant, and even if the success of both policies is equally likely.
To be clear, this particular problem with the VSL has nothing to do with the number of people the policy will actually save — it’s simply because in our daily lives, we take some dangers more seriously than others. That affects how much we spend to mitigate them, which is what the VSL measures.
This isn’t just strange — it’s bad economics. In a cost-benefit analysis, economists are supposed to value resources in terms of their social opportunity cost, i.e., the value of their next-best alternative use. If the same resource is being valued differently depending on seemingly irrelevant factors, then that is a sign that the analysis isn’t being conducted properly (it’s also a sign our spending habits may not be optimal from a social point of view).
The problem actually gets worse. Let’s say a policy is expected to save one life in Boston. If the cost-benefit analysis takes a domestic perspective, only counting benefits and costs to people in the United States, then the VSL might value that life at around $10 million (roughly the average of what Americans are willing to pay for it). However, if the analysis is global in scope, then standard VSL practices should put the value closer to $2 million for that same Bostonian’s life (or roughly the average of what everyone on the planet would pay). What, then, should we do with domestic regulations with global implications, like those addressing climate change or national security threats?
It should be obvious that the VSL just seems to defy common sense in many cases. Here’s another example: What if society has to make a tough choice between saving a 70-year old with 10 years of expected life remaining and a 20-year old with 60 years left? This is not a hypothetical, as hospital staff in places like Italy have been forced to make just these kinds of life-for-life decisions during the pandemic.
Most reasonable people would probably agree the young person is the one to save, if we absolutely have to make the choice. Yet, the VSL approach would be more likely to give priority to the elderly person. Why? Because the elderly tend to have more wealth than young adults, and therefore are willing and able to pay more to reduce risk. This is a lot like assorting spots on a lifeboat by selling tickets to the highest bidder.
Academics have not failed to notice these shortcomings of the VSL, and a vigorous debate is playing out as to whether a metric that has been a standard part of cost-benefit analysis for decades should be rethought. In recent months, the University of Pennsylvania Law School’s Regulatory Review blog has hosted a series of back and forth essays on the value of life, highlighting the lack of consensus among academics.
One of the biggest problems with the VSL is that it simply isn’t consistent with economic efficiency, which is purportedly what cost-benefit analysis measures. Using the VSL requires the extreme assumption that the private value of risk reduction to one small segment of society —usually workers whose willingness to accept money for dangerous jobs is measured in VSL studies — corresponds with the value to our whole society.
There is plenty of evidence that this isn’t correct, yet economists proceed with the assumption anyway, despite the fact that reasonable alternatives are readily available. Although no method is perfect, the courts, for example, have developed standards for valuing life in wrongful death suits, and these have stood the test of time. In general, these alternatives are less controversial than the VSL, and also more efficient.
It is a welcome development that a diverse array of economists, law professors, and even biomedical engineers from across the political spectrum are converging around the idea there is something very wrong in the state of policy analysis. Just as COVID-19 is forcing rethinking in so many other aspects of life, it’s unsettling parts of economics too, and for good reason.
James Broughel is a research fellow with the Mercatus Center at George Mason University.
[Editor’s note: This story originally was published by Real Clear Policy.]
