Sir Thomas More: Honorary Economist

Apparently there is nothing new under the sun. George Stigler, one of the University of Chicago’s all-time great economists, introduced the phrase “marginal deterrence” to economists in a 1970 research paper. The idea behind marginal deterrence is that you want to set penalties such that, even if you can’t commit the criminal from carrying out a crime, he still has incentives to shift his criminal behavior toward less socially costly crimes.

Stigler’s idea wasn’t completely new. I believe the Italian criminologist Cesare Beccaria was writing about it back in the 1700’s, around the time of Adam Smith. So Stigler’s contribution was 200 years late, because at least among economists the idea had been lost.

My former student Emily Feldman informs me now that even Beccaria was a few hundred years too late. Back in 1516, Sir Thomas More already understood the concept. He writes in book one of Utopia:

And surely there is no one who doesn’t know how absurd and even dangerous for society it is to punish theft and murder alike. If the thief realizes that theft by itself carries the same peril as murder, that thought alone will encourage him to kill the victim whom otherwise he would only have robbed. Apart from the fact that he is in no greater danger if he is caught, murder is safer, since he conceals both crimes by killing the witness. Thus while we strive to terrify thieves with extreme cruelty, we really urge them to kill the innocent.

Hats off to Emily for this remarkable discovery.


Montesquieu also made a big deal about this in his "The Spirit of the Law". In the section "Of the just Proportion between Punishments and Crimes," he writes:

"It is a great abuse amongst us to condemn to the same punishment a person that only robs on the highway and another who robs and murders. Surely, for the public security, some difference should be made in the punishment.

"In China, those who add murder to robbery are cut in pieces:53 but not so the others; to this difference it is owing that though they rob in that country they never murder.

"In Russia, where the punishment of robbery and murder is the same, they always murder.54 The dead, say they, tell no tales."

I wonder if he got this from More


That Thomas More . . . he was a man for all seasons!


How does this concept of marginal deterrence relate to the opposition of the "Three Strikes" law? Opponents of three strikes claim criminals with 2 strikes are more likely to committ violent crimes because the incentive is shifted to more costly crime for that third strike. I believe there was a series of posts already about it.


Well, yes, this is a book I read in high school English class called Utopias and Distopias.


Ok, nothing to do with the post, but I notice that the NYT has put ads in the feed for this 'blog. I wouldn't mind, except that the feed doesn't even show the whole post! Just a summary! One or the other, guys - ads and full content, or a teaser to get me to come and view the content here with ads. Argh.


There might be some people even before Thomas More that had this same idea, and a little bit of more research is needed to find him.
Incentives are the only way people will change their behavior. In real life, a good example are per-unit taxes; if people are charged a per-unit tax, they are going to reduce the total quantity demanded since the price of the good will increase. As More said, if people are treated the same for murdering or for stealing, more murders are more likely to be occurring. If people kill their victims, they are creating a safety net since the only witness is going to be dead by the time the trial begins. If murders are to be decreased, the penalty for murdering should be notably harsher than that of stealing.

Aaron Brown

"commit the criminal from" --> "prevent the criminal from", I assume.

Avi Rappoport

"even if you can't commit the criminal from carrying out a crime"

I think they meant "...prevent the criminal..."

Nguyen H. Tuan

"even if you can't commit the criminal from carrying out a crime, he still has incentives to shift his criminal behavior toward less socially costly crimes."

It takes him time and money to learn to master other jobs, since crime has been his career for long times, he will agree to do less costly crimes. Learning other stuff has more opportunity cost.

Matthew R.

So is murder a Giffen good?


i think the original law book is greek - by a guy called draco, from which we get the word draconian (and also a character in harry potter).

I think his legal code was pretty severe (hence draconian), however, it was the first time specific crimes led to specific punishment rather than just abitary cruelty for unspecified crimes. So in a way he discovered the concept written about by More, Feldman et al, i.e. the idea that a prospective criminal knows where (s)he stands in front of the law.


what about self-fulifilling punishment like waterboarding?- waterboard a thief and he'll confess to murder (if this doesn't work, repeat process)

C. Larity

So did the year 1516 incarnation internet trolls respond to Sir More by yelling "Slippery slope!" and asking him to cite his source?

Andre Wirjo

Quote: "The idea behind marginal deterrence is that you want to set penalties such that, even if you can't commit the criminal from carrying out a crime, he still has incentives to shift his criminal behavior toward less socially costly crimes."

Shouldn't it be "even if you can't prevent the criminal......." instead of "even if you can't commit the criminal......." ?

jesus alfaro

Nihil novum sub sole. If you search enough you will find a precedent for every economists' idea. Many of them, in books written by lawyers since the study of Law is much older than the study of economics. Take a look to Ihering's Scherz und Ernst in der Jurisprudenz (1885) and how he explains the economic sense of bankruptcy laws, for instance.


It would seem that two problems arise from this notion. As with all economics we assume that criminals behave in rational ways. Two, it assumes that criminals fear death above incarceration...again a rational assumption. It could be put forth that all criminals fear all punishment, ergo it would be nice to see some data to confirm these assumptions. Furthermore, without reading, I would believe that any first year criminal justice student could point out the problems with trying to adjust "incentives" for criminal behavior. If we are to understand the "incentives" for criminal behavior we need to understand how every mind works. I agree with marginal deterrence as a concept...but a very weak one. It is not to be confused with a hard and fast rule, simply an additional consideration when increasing or decreasing the difference in severity of punishment between "more" and "less" severe crimes. Remember, the death penalty (if carried out) works as a 100% guaranteed specific deterrent--even if it has varying degrees of effectiveness as a general deterrent.



I have no quarrel with Sir Thomas, but the concept of marginal deterrence would seem to be a bit different, or at least a bit broader, than the quote suggests.

I would say the idea is that there is to some extent substitution across criminal activities, in addition to the complementarity Sir Thomas appeals to. Thus, if robbery is judged worse than fraud--say because both involve a transfer of money, but the former carries a greater possibility of violence, or tends to be carried out by the "criminal classes"--you want to punish it more severely to incentivize at least some people to try fraud instead.

From an empirical perspective, would it be possible to assume marginal deterrrence and then work backward from actual punishments to an implicit social welfare function? This has been done, for example, with the simpler case of measurements of inequality.