Monday, March 03, 2008

More on Valuing Lives

The discussion of valuing lives in the comment thread below is great. In case anyone is still struggling with the approach used by economists, Steven Landsburg (an economics professor at Rochester who also writes a column for Slate) provides more discussion and several examples in this column. He also cites some evidence that shows that lives today are more valuable than lives in the past (because we are richer):

So, how do we find out how much a life is really worth? One of the best ways is to measure how much extra you have to pay someone to take a dangerous job. If lion tamers and elephant tamers have comparable skills and comparable working conditions, but lion tamers earn $20,000 a year more than elephant tamers, it's probably because that's what it takes to compensate someone for the risk of being eaten by a lion. And if that risk amounts to, say, an extra half-percent probability of dying on the job, then you figure that the value of a life must be $20,000 per half-percent, or $40,000 per percentage point, or $4 million.

So, once you carry out that experiment, how much does a typical life turn out to be worth? Professors Dora Costa of MIT and Matthew Kahn of Tufts point out that it depends on exactly when you asked the question. As incomes have risen, so has the value of life. The increase is more than proportional: A 10 percent rise in income is generally associated with about a 15 percent rise in the value of a life. Between 1940 and 1980, according to Costa and Kahn, the value of a life increased from about $1 million 1990 dollars to between $4 million and $5 million 1990 dollars.

(Other researchers, notably Harvard's Kip Viscusi, have found higher numbers. Viscusi estimates that the value of a life in 1970 might already have been as high as $8 million 1990 dollars.)

If lives tomorrow may be worth significantly more than lives today, I think this poses some additional difficulties for figuring out what to do about long term environmental problems.

Don't you have to assume a utility function where the marginal benefit of each dollar paid for risk decreases as risk increases? With a higher risk of death, the chance one will be around tomorrow to enjoy one's wealth lessens, so money is worth less. It doesn't seem like you can just multiply a half-percent by 200 and have a good estimate of the value of a human life.
Clearly the valuation of lives is a necessary policy decision that needs to be made by someone, sometime and somehow. However the problem i have with this concept is the way that Landsburg (and economists in general) go about valuing a life.... If i accept $x more per year to work a job with a 1/10,000 higher chance of death than the one i work now, then the value of my life is (10,000)($x), right? well what if i don't take the job, then in my risk-free world there is no value to my life. Even as an uninformed irrational individual, i would still put a positive value on my life, even though i am not engaged in a risky job....
Given this, it seems that lost (potential) wages is a more accurate way of measuring a life, even though it does put extra weight on the value of millionaires to our society. Perhaps an answer to Summer's 3rd world pollution idea would be to shift all our dangerous jobs and trash and nuclear waste to the third world, let these workers accept higher wages (for the risk) to clean up our mess and work in our dangerous factories, and as the value of their life rises (faster than the rate of growth of their income, as mentioned in the article), we will be able to re-evaluate the cost of our messy habits and consumption, and produce at an internationally efficient level, because the opportunity cost of third world lives is now more comparable to our own. ...perhaps this is completely incorrect theorizing, but it's an interesting idea!
Post a Comment

Subscribe to Post Comments [Atom]

<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to Posts [Atom]