The Demand For Murder

 From Steven Landsburg's Price Theory and Applications book.

--------------------------------

Example: The Demand for Murder

Many economists have applied the successful techniques of econometrics to the study of

demand curves for a variety of interesting “goods” that were previously viewed as outside
the realm of economic analysis. Consider, for example, the demand curve for murder.

Murder is an activity that some people choose to engage in for a variety of reasons. We
can view murder as a “good” for these people, and the commission of murder as the act
of consuming that good. The price of consuming the good is paid in many forms. One of
these forms is the risk of capital punishment.

This means that we can draw a demand curve for murder, plotting the probability of
capital punishment on the vertical axis and the quantity of murders committed on the horizontal
axis. We can ask how steep this demand curve is, which is the same thing as asking
whether a small increase in the probability of capital punishment will lead to a small or a
large decrease in the number of murders committed. In other words, measuring the slope of
this demand curve is the same thing as measuring the deterrent effect of capital punishment.

Now, on the one hand, the deterrent effect of capital punishment is something about
which there is much discussion and much interest. On the other hand, the slope of a
demand curve is something that economists know how to measure.

Over the past 25 years, Professor Isaac Ehrlich has repeatedly measured the slope of
the demand curve for murder, using essentially the same techniques that economists use
to measure the slope of the demand curves for shoes, coffee, and other consumer goods.
His results have been striking. The demand curve for murder appears to be remarkably
flat; that is, a small increase in the price of murder leads to a large decrease in the quantity
of murders committed. In fact, Ehrlich estimates that over the period 1935–1969 (a period
in which executions were more common than they are today, making the statistical tests
more reliable), one additional execution in the United States would have prevented, on
average, about eight murders per year.3

This is a remarkable example of an application of economics to a positive question:
“What is the deterrent effect of capital punishment?” It is emphatically not an answer to
the related normative question: “Is capital punishment a good thing?” It is entirely possible
to believe Ehrlich’s results and still oppose capital punishment on ethical grounds;
in fact, Ehrlich himself opposes capital punishment. However, knowing the answer to the
positive question is undoubtedly helpful in thinking about the normative one. The size of
the deterrent effect of the death penalty will certainly affect our assessment of its desirability,
even though our assessment depends on many other things as well.