In 1980 Ronald Howard wrote a paper entitled On Making Life and Death Decisions, in which he introduced the term micromort (mmt), which he used to mean a one in a million chance of death. In the paper, which originally appeared in a book published by General Motors, Ronald Howard points out that the average American runs maybe a 270 mmt/year risk of being killed in a car crash, a 28 mmt/year risk of drowning, and a 7.5 mmt/year risk of dying in a plane crash. People's perceptions of those risks are not consistent with the real risks, and as a result their behavior is sub-optimal -- they should be buying more automobile airbags and less airplane trip insurance.
Like buying airplane trip insurance instead of airbags, acts of thoughtless compassion fall far short of reaching rational objectives. Suppose you love little children and you hate to see them die slow, unhappy deaths. You've recently heard that little Jessica McClure is trapped in a well. When asked by a neighbor to contribute $35 to the local firefighters' fund you readily agree to pitch in to cover the cost of the rescue. After all, what else is there to do? We can't just leave her to die, can we?
But as a result of donating that $35 you now have $35 less to give to Planned Parenthood International. If your objective was to prevent little children from dying slow, unhappy deaths, you would have come closer to reaching your goal by donating it to Planned Parenthood, which would have used the money hundreds of times more efficiently. By helping Jessica you left another 36,000 to die that day. We can't just leave them to die, can we? We can, and we do, every day. But we don't need to.
I want to introduce the term megatragedy, by which I mean the collective experience of a million tragedies -- unhappy, painful events. Let's arbitrarily say that spending a year of your life slowly starving is a tragedy. That means that the human population experiences perhaps 40 megatragedies/year due to starvation. Alcoholism accounts for maybe another 10 megatragedies/year, car crash fatalities about 0.3 megatragedies/year.
These grand totals imply some assumptions. For example, to calculate the total for car crash induced tragedies we need to estimate the total annual incidence of car crash fatalities and then multiply it by the average tragedy value per fatality. But to come up with a tragedy value per car crash fatality we need to be able to compare the misery and suffering induced by a starvation with the misery and suffering induced by a car crash fatality. Of course we have no way of comparing these miseries objectively. It's crazy to even try. Or is it?
Aren't we actually making this kind of value judgment all the time? Think about making the decision to stay late at work at the cost of walking home alone after dark. As you make that decision you do a simple cost-benefit analysis, often without even thinking about it. You weigh the benefit of staying late against the risk of being accosted and the probable misery if you are. For small decisions it's fine to do the math in your head, even subconsciously. For larger decisions it's worth double-checking by doing some math on paper too, even if it means having to make crude estimates to quantify things that you'd rather not even think about.
"The formulation of the problem is often more essential than its solution."
-- Albert Einstein
The people of the world spend hundreds of billions of dollars per year on philanthropic endeavors, in an effort to prevent or alleviate some of the hundreds of megatragedies the world population endures annually. Are we spending that money efficiently? I don't think so. Many people are doing the math in their heads, guessing about how efficiently their money will be used, and how important it is to solve different problems. Unfortunately, many of their decisions -- though well intentioned -- are not the most effective solutions in attaining their desired goal.
The Bottom Line
Quantitative comparison is an essential tool for making informed
decisions about how to offer the most help.
