The Costs and Benefits of Prison:
Several readers have said that prison would be fine if it did not cost so much. They typically quote the figure of $25,000 per year per inmate. Nobody knows if that is the right number, but let us assume that it is. Anne Morrison Piehl, Bert Useem, and others have estimated the costs of crime. Their calculations, presented in the book Prison State, shows that property and assault crimes committed by the median (50th percentile) offender in Wisconsin, New Jersey, New York, Arizona, and New Mexico range from $26,000 to $46,000. It would seem that prison pays: crimes avoided exceed the annual cost per inmate of locking them up.
Pretty weak beer thus far, Prof. Wilson.
If the payors get the savings, there would be lots of new prisons and very little crime. We could put the punks in at 16 and let them out at 26.
If you have ever confronted the persona of victims, then the cost of prisons is cheap, not matter what cost we decide is accurate.
I imagine that if risk was factored into the equation, the costs for every taxpayer would not always cut in favor of imprisonment. (I'm not saying this is morally justifiable, but only that it may be empirically correct.)
Most of your posts provide just enough information that someone who is inclined to think you're right, can say, yeah, that's right. For everyone else, you're posts contain little that would encourage someone to change their mind.
1. assumption that imprisoning one person per year at $25k deters at least one "property and assault" crime per year is wildly unsupported. (mentioned above.)
2. reliance on median (rather than mean) prices, which is irrelevant for cost/benefit analysis.
3. disregards economic output prisoners could have produced had they not been imprisoned. (e.g., taxes on earned income.)
4. disregards court costs necessary for imprisoning, which would be far less were prison "off the table."
At least for some reasonable assumptions (single modality comes to mind) about the distribution over all perpetrators of those costs of crimes.
I'm not sure that first part is correct. Not all property crimes are covered by insurance and to the extent that they are, their costs are passed along to policy-holders in the form of higher premiums overall. Also there are "costs" that come from crime that are above and beyond the costs imposed on the victim of a particular crime. We all pay when our employers or companies we do business with have to hire security guards, install cameras or other security measures, or initiate security procedures that slow down business and limit our ability to access certain locations in order to prevent crimes. Even just the time wasted standing in line at department store while the cashier removes those ink things from a shirt you buy to prevent shoplifting imposes a cost that gets passed along to whoever buys a shirt and everyone else who has to wait in line longer to have it removed.
($10 billion spending, 173,000 inmates)
(I couldn't get links to work but Google "California Prison Budget Up By 79 Percent" )
First of all, property and assault crimes are only a subset of the universe of crimes. The benefits of preventing other crimes need to be considered.
Second, as has been mentioned above, this assumes prison deters 1 property or assault crime per year per offender.
Third, it ignores questions concerning the right level of analysis. Should we consider cost benefits on the aggregate or on an individual basis? If you want to maximize aggregate net benefits, the answer is that you consider these on an individual basis. For some offenders, the benefit of preventing the property or assault crime that would have occurred is much less than the median.
Fourth, it ignores questions concerning whether the median is the right metric. Consider this hypothetical. If the cost of imprisonment is $25,000 and the median benefit is $25,000, this is an indication that, for approximately half of the prison population, the costs of incarceration exceeds the benefits.
Fifth, the conclusion that "prison pays" is either obvious, or very crudely stated with a meaning that is unexpressed. Yes, obviously on the aggregate, nearly everyone is in agreement that the net benefits of having a prison system exceed the costs. The alternative is anarchy, which would have a very high cost indeed. However, the debate does not center around the aggregate benefit of running a prison system. No one is arguing that we eliminate all prisons. Rather, the question is whether there would be a large social benefit if we imprisoned less people. Do there exist at least some people where the social benefits would exceed the social costs if we released them? If yes, there is a potential for improvement by reform. The next question is whether there is a large number of such individuals. The next question is to what degree can they be identified and separated.
Sixth, very obviously, the $25,000 to taxpayers does not accurately state the full social cost of imprisonment. What about the daughter who grows up without a trusted and protective father? What about the son who grows up without a role model? What about children who do not see their mothers?
Now, in some cases, imprisonment obviously benefits the children of prisoners. But in many case, it does not.
Overall, I would say that this quick and dirty analysis by James Q. Wilson borders on the incompetent.
Yes, this is a blog, not an academic paper. But really, Wilson would apply a little more intelligence to something he is going to attach his name to.
The claim:
Interpreted to mean that there would be no net social benefits from reducing the number of individuals in prison is so far from being established by this pathetic "analysis" as to be embarrassing.
Interpreted to mean that having a prison system at all results in a net benefit is so obvious as to not even be worthy of discussion, given that the alternative is anarchy.
However you interpret the conclusion, this post is pathetic.
I think even an idiot (vide supra), comparing the amount of one's economic resources drained off by security concerns in the United States today versus, say, Russia today, or the US 150 years ago, would conclude that the derivative of economic growth with respect to security costs is still positive.
At least, if anyone seriously suggested that less incarceration and fewer criminal prosecutions, fewer police, et cetera would increase economic growth, even in California -- then I would seriously doubt his intelligence, sanity, or honesty.
In the real world, that kind of argument is never made. Instead, exponents of this or that justice methodology argue the relative efficiencies of security investments, e.g. that spending more money on drug treatment or prisons will both contribute to security and to improved economic growth, but you get more "bang for the buck" if you spend it on drug treatment.
You mean, like a couple drive-by trolls just did to Prof. Wilson?
I'd appreciate it if you not exclude half of the readership in that fashion.
Given finite resources, an argument for more money on drug treatment is an argument for less incarceration.
In fact, people do argue for less incarceration. Keep in mind, that one of the meanings of less incarceration is a sentence of 4 years in prison instead of 6 years. It is not clear that long prison terms for drug crimes is always a net social benefit in all cases in which such terms are imposed.
Wilson hasn't even tried to make any kind of full-scale detailed argument. He's just introduced the question and posted the conclusion, pointing to the book in which one might find the detailed argument. Would you suggest he post the contents of the book on the VC? I think Professor Volt might object.
And if he posted any proper subset of the book, you would complain about the key arguments and data being left out. So what can he do to satisfy your demands for a "better analysis?"
I suspect the only answer is "change his conclusions."
By the same token, very obviously "$26,000 to $46,000" does not accurately state the full social benefit of imprisonment.
Saying that there is nothing wrong with something is not an argument to specific points which suggest that there is in fact something wrong. It would be interesting if you made something more than entirely vacuous and conclusory statements.
If his analysis is going to be as pathetic as this post, I would say that he should not bother. That you have written a book about something does not excuse poor argumentation.
I am not suggesting that Wilson republish his book on this blog. I would however suggest that if he is going to present an argument, that he do it some justice and not present the most pathetic version possible. Wilson is, in short, making his own argument into a pathetic straw man.
It is bad enough when someone else makes your argument into a straw man. It is worse when you do that to your own argument.
The bottom-line is this. The data that Wilson presents do not even come close to even partially establishing his conclusion.
Agreed. This is not a very good metric of anything.
Oh don't be silly. The justice system is hardly a zero-sum game. We don't, as a polity, allocate X billion dollars to the system and then argue fiercely with how it should be divvied up. In practise, when people argue for increased drug treatment (say), they could (and probably are) also arguing for more total amount of money spent on criminal justice. The extra money comes out of other economic activity, schools, roads, vacations in Hawaii, whatever.
I've never heard of anyone arguing let's get more drug treatment programs for first offenders going, and to pay for it, let's close a few jails. The usual argument goes: let's get more drug treatment programs going, and then we'll find that we don't need so many jails and we'll be paid back for our investment in treatment programs. That fits with the paradigm of saying drug treatments are more efficient, and once we've got them going, we'll be able to reduce our investment in other forms of criminal justice.
If you can point to a contrary example, where someone suggested closing jails or reducing prosecutions or whatever in order to pay for alternative rehabilitation programs, please do so.
In fact, people do argue for less incarceration
Of course they do. But they do so for one of two reasons:
(1) Long sentences are a priori unjust. Victimless crimes, decriminalize marijuana, whatever.
(2) Shorter sentences would work just as well, sort of a diminishing returns kind of argument.
I'm sympathetic in principle to either argument, and either could be correct. But notice the absence of...
(3) Shorter sentences would work better, and would increase overall security and economic growth.
What would be the logic there, eh? If you keep people in jail for 6 years they get really pissed off, and turn to crime when they get out in revenge? Whereas if you keep them in for only 4 years, they become thoughtful and repentant and turn over a new leaf when they get out? Ha ha.
Most likely it's a figure approaching zero.
First of all, if we fund ANY drug treatment, we are implicitly allocating less resources to imprisonment (and everything else for that matter -- think opportunity cost). You may argue that it is not a zero sum game in the long run. I am here to tell you that the short run exists, and budgetary choices have to be made.
Yes, the pie we bake in the future may be larger. But the pie we have right now has to be sliced.
I will tell you how someone would make this argument. People who are employed are more economically productive than people who are in prison. When both parents are in the household, teenagers with more guidance commit less dangerous crime. Etc. etc.
In fact, the logic of (3) is fairly strong in some contexts.
If you proposed a 50-year prison sentence for shoplifting $500 worth of goods, I think it would be undeniable that a 2-year prison sentence would be better, and would increase overall security and growth.
What do you think would happen to the children of people imprisoned for 50 years? Would they face greater economic insecurity? Would they have discipline with only one parent? Would they face other consequences? Would these consequences manifest themselves in more crime by these children?
I think it is fairly obvious that if we had 50-year prison sentences for shoplifting, we would plausibly have less security and less economic growth. At least if you think there is any connection whatsoever between the probability that a child will commit crime and being raised in a one parent household.
Obviously, the analysis is not conclusive, because we would have to consider deterrence effects.
You really believe that the taxable income of prisoners released from prison is zero?
Criminal as role model?!?
I would love to see the data on recidivism rates/unemployment rates supporting that claim.
On a more serious note: Prof. Wilson, I appreciate your willingness to post here, but it seems to me that you don't quite understand the format of this particular method of communication. To be specific: the great advantage of a weblog is its exploitation of hypertextuality, its ability to link to supporting evidence and data and provide it in easy and convenient form, and readers have come to expect that sort of backup. In several posts so far, you've made an arguable assertion (most prison terms for drug possession are a result of plea bargaining from more serious crimes, property and assault crimes do, on average, $26 to $46k worth of damage, etc) and, instead of backing that assertion up with facts and figures, directed readers to other people's books and so on. I'd imagine that not one in a hundred of us has access to the book, Prison State, that you refer us to; certainly I'm not about to drive five miles to my college library to check your claims about it... and what results is irritation among the readership, because a scholarly blog post is expected to either provide the data or link to an online source containing the data; otherwise the post is considered mere assertion and goes in the same category as a WSJ editorial.
So, yeah. We'd appreciate it if you'd provide supporting evidence for your claims - quote the books if you have to, I assume you have them handy - so we can argue about that instead of descending into ad hominem due to the scantiness of your claims :)
tl;dr: [[citation required]]
As it is now, virtually all of the prison systems have virtually no education, counseling or rehabiliation as a cost-saving measure -- virtually guaranteeing that the prison will be overcrowded in the future.
Check out this report by the Center for Inquiry concerning the accuracy of Wilson's textbook on Government.
It turns out that in his textbook, Wilson describes the vast majority of scientists who believe in global warming as "activists" and those who do not as "skeptics."
Wow. Talk about ideologically charged and insulting language. To describe the vast majority of scientists as "activists" is to suggest that they are being led by something other than science while to describe global warming deniers as merely "skeptics" suggests that they are applying the scientific method impartially, given that the scientific method has skepticism built in.
My level of trust in Wilson has fallen rather precipitously at this point. Can you say "axe to grind."
As the discussion seems to have centered a bit on treatment in lieu of (or in conjunction with) prison time, I will assume we are talking about drug offenses.
*Most* people in prison for drug crimes tend to be at the low end of the economic scale, earning minimal taxable income. A good number earn their entire income via the black market.
The answer to your question is, yes.
You're right, Jmaie. Anyone who goes to prison is irredeemable trash, and may as well be thrown away. The idea that people could repent their crimes, clean themselves up and become contributing members of society is utterly ludicrous; statistics show that recidivism rates approach 100% among the criminal element; a 'three strikes, you're out' law like California's is far too lenient.
(sarcasm)
My comment was partly snarky, but I would be interested in seeing the tax return data on the median thug who commits property and assault crimes.
Citation required, Jmaie, unless you're just implying the traditional conservative 'drug crime = black male = ghetto welfare trash' equivalency.
People are much more than their mistakes. People who have at one time or another committed a crime are obviously not perfect. But then again, neither are you. We are all sinners.
Or is it your position that someone who possesses an unregistered gun in a jurisdiction that requires registration and then uses it in self-defense (i.e. a criminal) can never be a role model?
Yes. We are all rapists, we are all murderers, we all drive drunk, we all...
Oops. Didn't mean to step on a vacuous assertion and break it. Sorry.
First of all -- how do you determine a dollar price to society for the commission of a consensual crime?
Second -- the incarceration of a drug dealer merely provides a job opening for another drug dealer (law of supply and demand).
Third -- what about the cost of the crimes caused by prohibition itself?
If the person imprisoned is actually a criminal, their children and relatives benefit from their imprisonment. After all, a criminal cannot possibly be a role model for his children, or ever hold down a job and provide for his family, etc, etc.
(I'm being sarcastic, but I think some of the posters above actually believe this)
That you are imperfect is perfectly evident from the poor quality of your thinking.
I agree with a previous comment that the benefits of incapacitation are difficult to quantify.
And why do you find such a belief beyond the pale?
As if an outfit like CFI, which is "Committed to Science, Reason, Free Inquiry, Secularism, and Planetary Ethics," does not have an axe to grind. As for their charge, Prof. Wilson's textbook (which I have used in my Intro. to American Government classes) is correct to characterize as "activist" scientists who claim that global warming will be harmful to humans and that governments should act now since modern scientist qua scientist has no business making such "value judgments."
- the costs of policing and prosecution;
- the continued growth of prisons as a politically-favored industry (not dissimilar to the defense industry), with per capita imprisonment far exceeding those in other Western nations; and
- the relationship of imprisonment to the "drug war" and the costs of the drug war (not simply dollars spent, but also costs in terms of civil liberties and contribution to lack of inner city development by encouraging drug-related violence).
Long sentences for armed robbers, serial rapists, mobsters and other violent criminals by definition stop them from praying on society at large for as long as they're locked up (although summarily executing them like the Chinese tend to do, or "transporting" them to Australia would be even more effective, but I guess the Eight Amendment might have something to say about that).
Where this paradigm breaks down, however, is with white collar criminals and others who are unlikely to be much of a recidivism risk. Sentencing Jeffrey Skilling to over 24 years for his role in the Enron fraud is unlikely to make society "safer" in any direct sense, as it is doubtful that he would ever be in a position to carry out similar crimes in the future even if he'd received straight probation. And he's not going to be "more" rehabilitated when he is eventually released as an octogenarian than if he'd served a much shorter sentence [because of the offender-specific nature of the question posed above, I'm deliberately leaving out consideration of separate general deterrence and retribution justifications for sending the likes of Skilling away for a long time].
With armed robbers or rapists with a high recidivism rate, society probably wins from long sentences. Skilling, not so much.
All of which suggests that just looking at "average" figures probably isn't terribly helpful.
And I would assume that some of the book's sillier claims (the doctrine of original sin was a motivating factor in drafting the constitution?) were dropped in later editions?
I don't think the 8th Amendment would have much to say, but I'll bet Australia would.
It's no wonder we keep building prisons - each one is a modern day rotten borough unto itself!
It is indeed widely used. The running joke is that royalties from the textbook helped pay for Wilson's posh Malibu estate. I agree that the section on the Establishment Clause needs some clarification--though any decent professor will provide it. No textbook is perfect. And if you want to talk about political bias, I could cite innumerable examples of liberal bias in introductory textbooks on American government.
Another problem is that even if increasing a prison term lowers societal costs, it increases governmental costs, at least on the state level. So when they increase a penalty, legislators should immediately account for the prison space by appropriating the necessary revenue. But, following typical Republican-style budgeting, they incur the cost and pretend that it either is costless or that the money will magically appear.
Evidence that Prof. Wilson is not taken seriously as a scholar?
No problem.
A quick search of Academic Search Premier for peer-reviewed publications reveals that Prof Wilson has a total of 81 publications since the early 60s. This sounds impressive, except that 27 citations are to Commentary, 15 are to The Public Interest, and 8 are to Current, not really what most would consider peer-reviewed journals. And except that the vast majority are book reviews, not research.
He has one publication in a peer-reviewed journal since 1991, and note that that article does not present research, but calls for more evaluation research. He has a career total of 7 peer-reviewed articles presenting original research (using that latter term very elastically here), 3 in the 60s, 2 in the 70s, 1 in the 80s and the aforementioned 1991 publication.
As far as engaging in substantive debate with Wilson: as someone who has graded a lot of undergraduate papers, there is bad work where you point out the flaws in the arguments in an effort to improve it, and then there is work so bad you just write "NO". Engaging in debate with Wilson simply dignifies his apologetics of repression with the mantle of scholarship. Sorry to go all Godwin on you, but sometimes debate is not appropriate.
Yawn... This is not true. I've worked in several prisons and there always is education and counseling programs. The fact is that these programs, on the whole, don't work (in terms of reducing recidivism) and are often gamed by the inmates in an effort to earn early release.
Had you looked at Wilson's record on google scholar (which includes his books), you'd notice he's been cited close to 10,000 times. Makes we wish that I wasn't taken seriously as a scholar too...
I did; that's how I noticed that his output is almost exclusively books (which are typically not peer-reviewed).
The point is not that he has not published lots of words (he has), it's that he has not established a scholarly publication record commensurate with the position he holds.
Of course he's cited often, because that's his job: to provide "scholarly" backing for harsh penalties toward the underclass; few of those citations are in peer-reviewed scholarly publications, however, because (here's where I came in) he is an ideologue, not taken seriously as a scholar.
I don't know how criminology works, but there's quite a few social science fields where books are the primary research outputs. And it's not like his books were published by his Mum; Princeton and Harvard are amongst the publishers. Also, I had a quick look at where the most recent cites to his bureaucracy book come from: the bulk were peer-reviewed journals. So he's obviously an infuential scholar (although he may be controversial, I have no idea).
Empirical work (the kind he has done little of) generally appears in peer-reviewed journals first, even if it's later recycled into a book chapter.
His Harvard UP publication was 1968; the Princeton one was 1995, but is a non-criminology topic, and thus not relevant to a discussion of criminal justice scholarship; it may well be fine work.
Most of his recent output has been popular press publishers, and, predominantly, the subsidized press of the right (American enterprise Institute, e.g.).
The hidden left-wing agenda in this list is a little too subtle for me to notice.
On the other hand, this list does seem to bear some ideological stamp, wouldn't you agree?
Please use terms with better precision and accuracy. Incapacitation and deterrence are not the same. Conflating the two leads to confusion. Unless confusion is your objective.
Like Prof. Wilson, I haven't done any empirical work in this area.
Unlike Prof. Wilson, I don't earn a living selectively citing the work of others where it can be said that work supports the agenda du jour of the anti-libertarian right, so I won't claim expertise I don't have.
Actually, I think you're missing the precision of the terminology in the post.
Specific deterrence, as some of us learned the term eons ago in our Crim Law courses, is defined as steps society can take to keep the particular offender from initiating or repeating his criminal conduct. A particularly effective method of specific deterrence for someone who has already taken a criminal path is, of course, incapacitation through incarceration.
Do not confuse specific deterrence with general deterrence, which in this context is imposing punishment to set an example that persuades other potential offenders that doing the crime isn't worth it.
Maybe sending Jeffrey Skilling away for north of two decades is a wonderful way of advancing general deterrence -- other corporate CEOs might well sit up and take notice. But IMHO it has little to no value as a specific deterrence technique for either him or other similarly-situated white collar criminals.
As I said, any decent professor will provide clarification (or correction if you prefer), which I do in my lecture on the subject. That's what I'm there for and get paid the big bucks (not really).
"While it clearly pays to incarcerate those at the 80th percentile in all three states, on incapacitation grounds alone, it does not appear to "pay" to incarcerate those below the median. The social costs associated with the offender at the 40th percentile are all below $15,000."
I don't have a cite for you, Tyrant King Porn Dragon. My comment was based on personal experience with two dozen or so individuals I spent time with in my early twenties when I was hanging out.
Insinuations of racism aside, do you have anything substantive to say? Am I incorrect?
Do you have anything substantive to say? Am I wrong?
the OP is not using the benefit of deterrence to calculate the benefit of prision. he is using the benefit of incapacitation.
in order to use his figure of the benefit of prision we need not assume that it will deter one crime per year (or any crime at all) to have a prison sentence for a year...we only need assume that while hes in prison-he wont commit crime.
its the difference between incapacitation and deterrence.