Incarceration as a labor market outcome

by John Quiggin on May 30, 2009

I wasn’t all that surprised that Bryan Caplan didn’t like my interpretation of our bet on EU and US unemployment rates, which was that the combined rates of unemployment and incarceration in the US would exceed those in the EU over the next ten years. I was, however, surprised by the vehemence with which libertarian-inclined* commenters here and at my blog objected to this interpretation.

A string of them echoed Caplan’s argument that

From a labor market perspective, though, Quiggin’s incarceration adjustment would only make sense if you thought that most or all of the people in jail would be unemployed if they were released.

Caplan has missed my main point. I’m not suggesting that incarceration is disguised unemployment (though obviously it reduces measured unemployment). Rather, I’m saying that, like unemployment, incarceration should be regarded as a (bad) labor market outcome. If you want to evaluate the performance of the labor market, you need to look at both.

There’s nothing radical or leftist about this viewpoint: it’s one that is at least implicit in all economic models of the labor market of which I’m aware, and is most particularly explicit in that of the Chicago School*. Most of the crimes for which people are imprisoned in the US can be understood as reflecting economic choices which in turn are determined primarily by the labor market in which those choices are made. This is obviously true of property crime and drug dealing, and it’s true, directly or indirectly, of lots of violent crime as well. As Gary Becker put it (quoting from memory here) “a burglar is a burglar for the same reasons as I am a professor”. (You don’t have to buy Becker’s assumption that criminality is a “rational” choice, to agree that it is a choice and that choices reflect the attractiveness of the available options).

There’s plenty of statistical evidence from scholars like Glenn Loury to show that criminals, and particularly those who end up incarcerated, are drawn disproportionately from groups with bad labor market prospects: poor, disproportionately black, facing low wages and high risk of unemployment. But well-done case studies are often more convincing, so I’ll point to the Venkatesh study of Chicago drug dealers reported in Steve Levitt’s Freakonomics. Venkatesh found that most street dealers were making less than minimum wages, and were motivated by the very low probability of surviving to attain the only high-paying job realistically available to them, that of the local kingpin. Even more striking was the observation that, when gang members learned Venkatesh was a university professor, they approached him in the hope that he would be able to wangle them a job as a janitor – otherwise an ambitious, and probably unattainable aspiration.

The Chicago theory on which the case for flexible labor markets is based predicts that the lower is the return associated with the “outside options” of employment or reliance on social insurance, the higher will be the incentive to engage in crime as a way of making a living. The only way to offset this is to make crime still less attractive, or less feasible, through high rates of imprisonment and long prison terms. That is, other things equal, low wages and weak or non-existent unemployment benefit systems can be expected to lead to higher crime rates, higher rates of imprisonment of both. So, any consistent advocate of the Chicago theory should treat both incarceration rates and unemployment rates as labor market outcomes.

Unfortunately, as has been shown by the current debate, there’s not a lot of willingness to explore the logical implications of the Chicago line to a position that might undermine its policy conclusions. Loury has noted the destructive effects of imprisonment (in Chicago terms, it causes rapid depreciation of human capital). There’s no good reason a priori to suppose that a labor market in which wages are low and unemployed are treated badly will do better, when both unemployment and incarceration are taken into account than one with higher minimum wages and more generous social welfare.

So, I would argue, my interpretation of my bet with Bryan Caplan is the more relevant one in terms of policy evaluation. The proportion of bad labor market outcomes is better measured by the sum of unemployment** and incarceration (expressed as a proportion of the labour force) than by unemployment alone.

  • Or maybe shmibertarian: as we saw during the Bush era lots of alleged libertarians are quite comfortable with extreme use of state power as long is doesn’t touch their bank balances. On the other side of the coin, I should note that the Cato Institute has done some very good work on this subject, including publishing this Glenn Loury piece.

** I’m leaving aside issues about the best definition of unemployment, underemployment and so on, which have been canvassed extensively in earlier discussion.

{ 100 comments }

1

David 05.31.09 at 12:11 am

The reasons for incarceration are many and intertwined, but it should occur to anyone other than a libertarian that quite a few people end up incarcerated because of poor employment opportunities. In other words, one of the many less than desirable outcomes of a whole slew factors involved in poverty. Poor health care, housing, nutrition and so on. I’m not at all surprised the ls are up in arms about the suggestion.

2

Nick 05.31.09 at 12:12 am

Speaking ‘as a libertarian’, I think you might well be right on this point. One can’t easily split up economic choices along moralised lines and attempting to do so is usually what get conservatives (and conservative leaning libertarians) in rather a muddle: “Rich people need incentives, poor people are just lazy!”

3

John Protevi 05.31.09 at 12:12 am

Thank you for the link to the very powerful essay by Loury.

4

Kieran Healy 05.31.09 at 1:14 am

Bruce Western is the go-to guy on this. This paper with Katherine Beckett — How Unregulated Is the U.S. Labor Market? The Penal System as a Labor Market Institution — is an early statement of the view. I’ve written here in the past about some of his subsequent work with Becky Pettit on the role of the penal system in the lives of unskilled (especially unskilled, black) young men.

5

Omega Centauri 05.31.09 at 1:55 am

While I agree that some component of the incarcerated population are economic criminals, I am under the impression that a good bit of our excess incarceration rate is due to a draconian approach to popular recreational drugs. I wouldn’t consider a person in prison for possession for personal use to be there because of a labour market failure.

6

John Quiggin 05.31.09 at 2:27 am

Thanks Kieran. Bruce’s name had come to mind, but I didn’t turn up the right reference.

OC, I don’t think people imprisoned for nothing more than personal use (as opposed to plea-bargained possession charges for those suspected of dealing) make up a large proportion of the US prison population. And I think its quite unusual for users with steady middle class jobs to be prosecuted at all, let alone imprisoned. Does anyone have good evidence on this?

7

T. Paine 05.31.09 at 2:33 am

I’m obviously under-informed here, but aren’t incarcerated individuals unemployed by definition? They’re not working, regardless of the reason(s) WHY they’re not working…

8

John Protevi 05.31.09 at 2:33 am

I wouldn’t consider a person in prison for possession for personal use to be there because of a labour market failure.

I’m not sure, but I think we’d have to distinguish people in prison (as opposed to jail) for simple possession versus those there for petty dealing, which would seem more in line with JQ’s thesis. Anybody know the numbers here?

9

StevenAttewell 05.31.09 at 3:08 am

Even without getting into the labor market outcome issue, wouldn’t it be more accurate to include them if we’re talking in broad terms about what percentage of the working-age population is working and which isn’t? The U.S has an abnormally high number of people in prison – and they’re not counted as unemployed, even though they don’t have jobs and are working age.

However, John Quiggan, this does raise the question of how economists deal with prison labor – after all, these people are performing labor, are creating goods and services, they get some form of wages, but they’re not exactly what we would call free labor. How do we quantify prison laborers – in the work force or out of it?

10

Matthias Wasser 05.31.09 at 3:23 am

I’m obviously under-informed here, but aren’t incarcerated individuals unemployed by definition? They’re not working, regardless of the reason(s) WHY they’re not working…

Sure, but that’s not enough to show that it’s relevant for discussing the strength of US labor market regulations versus European labor market regulations. The fact that it’s a causal consequence of those regimes is.

11

StevenAttewell 05.31.09 at 4:59 am

Matthias – yes and no. Normally, we’d think of incerated people as not working, but one of the weird things about the U.S prison system is that we have a massive and growing system of prison labor, especially in privatized, for-profit prisons.

12

Shawn Crowley 05.31.09 at 7:26 am

Re: John’s question @6

I can’t give you hard data, only personal observations. Although “dealer” and “user” might seem to be distinct categories, and US law treats them as such, reality is much fuzzier. Consider a crack addict lacking crack and any money to purchase crack. He is approached by a would-be user asking if there’s any crack to be had. Our addict says yes and steers the other user to a seller in return for the promise of a hit off the pipe. Our addict, having facilitated a sale is legally culpable to the same degree as the actual seller. Determinate sentencing sends our “cluck” off to prison for a long time.

In Washington state, where I practice law, the sale of “bunk” (fake drugs) is punished just as if actual drugs were sold. The sale of bunk is another tactic used by poor addicts to get drugs. Once again, “user” and “dealer” are conflated.

Several years ago the prosecutor in our county finally agreed to allow clucks to plead to a reduced charges in some circumstances thus avoiding harsh determinate sentences. I’m convinced the policy change came about only due to county budget problems and not any sense of fairness. This change was too late for a client of mine who served 8.5 years for steering an undercover cop to a single $20 rock. The client had two prior sales of bunk so he was treated as a recidivist dealer. When sentenced he had gotten clean and was employed as a store clerk. Didn’t matter.

The US prisons are full of people doing long sentences because they are screw ups who keep screwing up. The level of criminality might not escalate but the sentences do, being based on the number of prior offenses. “Three strike” laws have most frequently been used against unarmed robbers rather than more violent criminals. And unarmed robbery can be as little as a shoplifter pushing past a clerk to get away.

The only good thing to come out of the current recession is a greater willingness to discuss incarceration as the solution to social ills. Elected officials can characterize sentence reform as economic necessity without being tarred as soft on crime.

13

jsalvati 05.31.09 at 7:38 am

Omega Centauri:

I am not under the impression that there is a lot of imprisonment for simple possession. However, draconian US drug laws do make crime a lot more lucrative (since people really like doing drugs and are willing to pay for them), and as Quiggin points out, this encourages people to be criminals. So, I think there’s big component of the incarceration rate that can’t really be explained as a labor market outcome.

John Quiggin:

Why didn’t you discuss this when you were talking about the bet with Caplan? I certainly didn’t quite get what your point was.

14

Henri Vieuxtemps 05.31.09 at 8:22 am

It’s a good analysis, but I think there’s more to it. Low wages and weak or non-existent unemployment benefit systems can be expected to lead not only to higher crime rates, but also to greater social unrest, riots, political instability – thus rending “flexible labor market” model unfeasible. This makes high incarceration rate not a choice but necessity, and it has little to do with crimes. The purpose is not to reduce or punish criminals, the main purpose is to reduce volatility in the system, neutralize and remove potential trouble-makers.

15

David Wright 05.31.09 at 9:25 am

Why isn’t it a “bad labor market outcome” that I’m not making $1M/year? Well, you say, probably you don’t have skills that the market could put to use creating $1M/year of value for others. Why can’t the same explanation apply to the ciminals? Why are you so sure that it’s not socially optimal that they be that they be incarcerated?

In any case the policies John proposes to make non-criminal employment more attractive, namely high minimum wages and unemployment benefits, would hardly seem to do what John wants them to. If no employer can get enough value out of the drug dealers sketched in Freakonomics in hire them as minimum-wage janitors, then wouldn’t raising the minimum wage just cast yet a larger fraction of the population into that unemployable caste? Yes, I am less likely to resort to crime to fee myself if I am getting an unemployment check, but the unemployment check makes me more likely to languish longer in that socially disconnected state. (John and Bryan can argue about the noise in cross-country comparisons of unemployment rates, but one one signal that rises far above the noise is that unemployment in European social democracies tends to be much longer than in the U.S.) Anyway, unemployment benefits in most countries, including the European social democracies with which I’m familiar, acrue to those who loose long-term, full-time jobs, not the sort of marginal workers who flit in and out of short-term, part-time jobs.

16

novakant 05.31.09 at 10:02 am

Venkatesh found that most street dealers were making less than minimum wages, and were motivated by the very low probability of surviving to attain the only high-paying job realistically available to them, that of the local kingpin. Even more striking was the observation that, when gang members learned Venkatesh was a university professor, they approached him in the hope that he would be able to wangle them a job as a janitor – otherwise an ambitious, and probably unattainable aspiration.

Hmm, is that really true? Granted, I get my sociological insights on crime from The Wire, The Sopranos and suchlike, but it seems rather implausible to me that these guys are all making less than minimum wage and would dream of a job as a janitor.

17

JoB 05.31.09 at 10:44 am

Henri-14, spot on! The case isn’t against unemployment. The same that cry ‘jealous!’ if one makes a comment on high wages in the private sector cry ‘lazy!’ if people can live a reasonably healthy life whilst unemployed. These people are themselves just jealous.

The case to be made is the case against the need for employment to lead a decent life. I believe the labour market will be plentiful if those willing to work get those jobs that in current days are of necessity occupied by those really not willing to work.

Crime would not go away though. Not as long as people strive to have power & money, far in excess of the median. Only social unrest would go away if unfairness, and forced labour, are taken out of the equation.

Really, what purpose is served by ‘everybody pulling her/his own weight?’ And really, why would there be a correlation between unemployment & lack of contribution? The best thought and art has come from those willfully unemployed.

18

Phil 05.31.09 at 11:38 am

Loury has noted the destructive effects of imprisonment (in Chicago terms, it causes rapid depreciation of human capital).

And not just among those imprisoned – if n% of the male population of an area consists of people who are either about to be sent down or have just been released, the social fabric of the entire area is going to severely stressed.

19

F. Connell 05.31.09 at 2:10 pm

This seems like a bizarrely economistic understanding of crime, particularly for a site like CT. Americans and Europeans have dramatically different attitudes toward violence, dramatically different access to guns, and dramatically different drug cultures (and drug laws). The idea that the difference in prison rates is, at the core, about differences in the two regions’ economies seems ill-considered and reductionist in, ironically, the worst Chicago-School way. (There’s something frankly odd about John citing Gary Becker here, simply because it suits his broader argumentative purposes.)

20

William U. 05.31.09 at 3:47 pm

T. Paine wins this thread for his lulzy moniker.

Anyone read this NY Times article from yesterday?

http://www.nytimes.com/2009/05/31/us/31border.html?em

“Investigators determined that the two immigrants, Jose Manuel Cazeras-Contreras, 30, and Victor Delgadillo Parra, 23, began distributing heroin when they were unable to find jobs. […]

Mr. Contreras, the dispatcher in the case, told federal authorities that he had crossed the border illegally and lived in Oregon for several years before moving to Columbus in 2007 on the promise of a job as an auto mechanic. But that job never materialized. In a letter to The New York Times, he said he had worked a variety of other jobs but had hit an unemployment streak that left him without a car or a house for his wife and two young children.

Desperate for work, he said he found an acquaintance in Columbus who promised him easy money for distributing heroin.”

21

Barry 05.31.09 at 4:27 pm

novakant 05.31.09 at 10:02 am
(re: Freaknomics and drug rings).

“is that really true? Granted, I get my sociological insights on crime from The Wire, The Sopranos and suchlike, but it seems rather implausible to me that these guys are all making less than minimum wage and would dream of a job as a janitor.”

Daniel Davies, of this blog, has a rather scathing five-part take-down of the whole ‘Freaknomics’ book on his blog. I’ll let him link, because the posts are
hard to find.

22

dsquared 05.31.09 at 6:05 pm

Yes, Venkatesh found what he found, but I am pretty sceptical of the claim that drug dealers were making minimum wage. Other studies of crack gangs have found significantly higher wages.

23

Shawn Crowley 05.31.09 at 6:48 pm

Whatever the problems with the Freaknomics analysis (I have only read summaries) the conclusion rings true to me. The vast bulk of street level drug dealers can’t make even nominal bail amounts (often less than $5000) and lacking any collateral (title to a car) can’t utilize bail bond agents. This observation is not personal sampling error. For almost two years I handled all of the felony arraignments for unrepresented defendants at the King County Courthouse in Seattle. I was also present to observe defendants who had managed to hire private counsel. Only a very few upper-level dealers could bail out. Exaggeration is common to street criminals about how much they make and how well they live.

Psychic gratification is an overlooked factor. Being a drug dealer has positive social status in the circles dealers move in. You are connected. People come to you. Higher quality dope means higher social status. I think that many dealers would keep selling even if they didn’t make any money just to maintain this identity and social relationship.

24

Bloix 05.31.09 at 8:22 pm

Henri #14 – riots? social unrest? political instability? is this supposed to be a description of the US over last two decades? Because we’ve had none of those things. So either our high incarceration rates (in particular, of non-violent petty drug offenders) have been notably successful in maintaining social peace, or they’ve been unnecessary, wasteful and cruel. Are you arguing in favor of the first alternative?

25

Henri Vieuxtemps 05.31.09 at 9:02 pm

Yes, notably successful, IMO. And it’s three decades, not two. Incarceration rates took off mid-1970s and by mid-1980s, despite all the ruinous effects of reaganomics, political and social stability had notably improved, occasional workplace massacre notwithstanding.

2 million desperate young men taken off the streets at any given time – that’s not a fluke, it’s a strategy.

26

sophie 05.31.09 at 10:30 pm

if you have certain apparel labels in your closet, you are wearing clothes manufactured by incarcerated labor. Are they ‘unemployed’ if they work but don’t get paid? ditto for a lot of customer service jobs, etc. In other words, you are arguing that slave labor and unemployed numbers should be combined, so better count the third category as well: unpaid workers in the US.

but, i did argue a portion of your idea long ago with a local politician and he came unglued as well. and i was equally baffled by the reaction. seemed pretty darned obvious to me.

27

Crystal 05.31.09 at 11:20 pm

It’s not just young men actually in prison – once they get out they’re far less likely to receive job offers, if they’re African-American:

http://www.nytimes.com/2005/06/17/nyregion/17felons.html

Of course, these unemployable ex-offenders are likely to go right back to prison for another crime, after all, a guy’s gotta eat…

28

dan 06.01.09 at 2:58 am

“Henri #14 – riots? social unrest? political instability? is this supposed to be a description of the US over last two decades? Because we’ve had none of those things. So either our high incarceration rates (in particular, of non-violent petty drug offenders) have been notably successful in maintaining social peace, or they’ve been unnecessary, wasteful and cruel. Are you arguing in favor of the first alternative?”

He’s saying that we’re solving the problem of social unrest by throwing millions of marginalized people into jail when we should be dealing with the problem of social unrest by not marginalizing millions of people.

29

T. Paine 06.01.09 at 3:14 am

Thank you, William U. I was using this name long before the artist-known-as appeared on the scene.

Matthias @ 10: Sure, but the causal effects of imprisonment do have labor market outcomes, that, as alluded to elsewhere on the thread, make formed felons well-nigh unemployable. Why distinguish between them when they are in prison and when they get out? Not to mention that it sounds like you’re asking that we apply some sort of motive evaluation to prison policies (“Politicians aren’t purposefully seeking to make prisoners unemployed, therefore they don’t count as such.”). That seems to be setting the bar a little high for evaluating the outcomes of any policy at all.

30

Henri Vieuxtemps 06.01.09 at 8:56 am

…when we should be dealing with the problem of social unrest by not marginalizing millions of people.

Well, I wouldn’t put it exactly that way. People will be marginalized anyway; they are marginalized in France as we saw a few years ago, and in Sweden too, I’m sure. It’s only a matter of degree, it’s an equilibrium: a more liberal labor market requires more control, more indoctrination, more repressive policies.

Which is pretty much what John is saying, only in my view it’s not because people choose to break some stupid law, it’s more like they reject the whole model and it’s legitimacy, and that is dangerous and unacceptable to the Establishment.

31

mpowell 06.01.09 at 10:28 am

Yeah, I think there is a good reason that Caplan is annoyed with your interpretation. Only a fool would argue that social policy as a whole doesn’t determine both unemployment and imprisonment rates, but the measure you recommed is actually quite terrible. I would much rather be unemployed than incarcerated and so by those measures the US society fails dramatically in comparison to France, for example. Now maybe we have a history that makes our government’s job a little harder, but no need to argue that the outcome is worse. But that’s a little different from the much more limited debate about what our government’s policy should be regarding labor and employment. Generally to discuss the higher incarceration rates in the United States as a labor outcome in the context of this debate just doesn’t make any sense. They are a result of social policies; it is not our labor market that is at fault. You could convert our entire labor system to a French one and our incarceration rate would barely budge. I think inner city blacks who are committing crimes would actually have an even harder time getting an honest job in a French style labor market (would you really challenge this claim?). The unemployment leading to crime in American inner cities has everything to do with social policies that creates a large class of virtually unemployable Americans. They are unemployable in spite of labor policy that provides degrading jobs to unskilled and disadvantaged workers, not because of it. The result is 99% social policy and 1% labor policy. Caplan is right to not want to let a measure of social policy get involved in the debate.

32

Henri Vieuxtemps 06.01.09 at 11:57 am

a large class of virtually unemployable Americans

Isn’t this how Mortimer lost one dollar to Randolph?

33

Danila 06.01.09 at 12:16 pm

Another aspect is that imprisonment costs a lot. One number that I found here by a quick search is 19300 $/year, which is more than a European-style unemployment benefit would be (15000$? – I am not very aware of US prices). Even if we consider the revenue generated by the work of the prisoners for outside firms, I still think it will not make a big difference for the government expenses. So, economically, the situation where a number of people are on long term unemployment and the situation where they are in prison are more or less equal. Socially it would be a big improvement.

Of course, this rests on the premise that they are a large number of people that would have not resorted to criminal activities had they been guaranteed a decent income, which I think is true. What proportion of the prison population enters in this category is, I suppose, the main point of this debate.

34

mpowell 06.01.09 at 12:28 pm

32: I’m not sure what you’re referring to. I’m just pointing out that raising kids in a crime ridden environment with fathers in prison, terrible schools and poor nutrition has more to do with them lacking the ability to find a job in the legal economy than is determined by union regulations and unemployment benefits.

35

Nick 06.01.09 at 12:39 pm

“It’s a good analysis, but I think there’s more to it. Low wages and weak or non-existent unemployment benefit systems can be expected to lead not only to higher crime rates, but also to greater social unrest, riots, political instability – thus rending “flexible labor market” model unfeasible.”

Really? Then how come it is France with all the burnt-out cars?

36

Henri Vieuxtemps 06.01.09 at 12:58 pm

Oh, I was referring to this old movie.

Well, I don’t know anybody from projects, but judging by popular culture people there certainly show more common sense than, say, virtually all 100% of the pundits in the mainstream media. Even if they are indeed unemployable in the positions available for then in the actually existing economy, it certainly doesn’t mean they can’t do useful work.

37

Henri Vieuxtemps 06.01.09 at 1:00 pm

Then how come it is France with all the burnt-out cars?

Why, I suppose it’s because their rate of incarceration isn’t high enough.

38

Tim Wilkinson 06.01.09 at 1:07 pm

A 1st-year-PPE-dropout turned London ‘philosophiser’ writes:

So are professional criminals of various kinds regarded as employed or not?

Is drug prohibition (and theft!) a market failure/defect of some type (labour or ‘other’, if those can be disentangled)?

It seems clear that this must have some relevance to the discussion, though don’t ask me what. I’m asking you lot.

FWIW I think economics is a lot more baffling than it’s given credit for. ‘Schtop – this science isn’t ready yet’. (allusion: UK beer ad commercial.)

39

Barry 06.01.09 at 1:07 pm

Nick 06.01.09 at 12:39 pm

“Really? Then how come it is France with all the burnt-out cars?”

They have burnt-out cars; in the US some car-burning counts as a college football victory party; it’s not a *real* US riot until some people get shot.

It’s interesting that people will point in shock and horror to things in other countries which (to my eyes) seem to be far less violent than in the US.

40

Tim Wilkinson 06.01.09 at 1:11 pm

#38 – I think it’s just that cars have equal status with human beings in the US.

41

magistra 06.01.09 at 1:57 pm

This is from a recent NY Review of Books article:

“The inmates at San Bruno were typical of prison and jail populations across America. Over half were black, although blacks make up only 6 percent of San Francisco’s population. Approximately 75 percent were high school dropouts, and most had reading skills below the seventh-grade level; 65 percent had been relegated to special education programs before dropping out of school, and 90 percent had never held a legal job. “

Don’t you think most of those would be unemployed if not in prison?

42

Preachy Preach 06.01.09 at 2:53 pm

Out of curiosity, how do US incarceration rates compare the percentage of the young male population in continental Europe doing their national service?

(Although both may take significant numbers of young men out of mainstream society, it strikes me that the latter is slightly less detrimental to their future lives…)

43

ralph 06.01.09 at 11:01 pm

Cripes, you people. I assert that people make the choices they make from their assessment of the best outcome for them from among the options they can see at that moment. Many of these are “rational”, many are “not.” But by defintion — I should think — “illegal” activities are also behaviors that someone wants to use their human freedom to pursue. The “getting into prison” part of it is a political issue, not whether there was an exercise in freedom; there was.
But is it a “failure” to incarcerate all these people? Heck, that’s not an economic question, I shouldn’t think — I’d think the economic issue was whether, once they were in prison, they are “unemployed in some formal sense.” I think John’s right in that there are two interpretations here:

1. That some proportion of the prison population must be considered unemployed to perform statistical comparisons of this sort.
2. That they are not unemployed for such comparisons, either because they “couldn’t find jobs anyway” or simply on principle that crime isn’t economic activity at all.

Frankly, I tend toward 1, although I cannot see how you can conclude that all released prisoners would have jobs that will register with the statisticians. First, their will always be criminal economic activity of some sort, and second, it’s not clear how many jobs there could be for these people of the sort that will help. But I DO believe that it’s reasonable to apply some heuristics here and choose some reasonable number or range of numbers and run with it. The problem seems to be that each system circumscribes in different ways what kinds of statistically detectable employment can exist. If so, how to balance that for a good comparison — if not a good bet!

44

ralph 06.01.09 at 11:03 pm

Doh. “First, there will be…” Dang it.

45

ralph 06.01.09 at 11:14 pm

Re: David Wright — a fine person, no doubt:

I can’t see how the fact that a single person might or might not benefit says anything about the labor market as a whole, although I understand the point you’re making. You’re right: It **is** a labor market failure that you’re not raking it in more than you are. :-)

I’m not clear on your second graph: If you’re trying to point out that some social-democratic policies result in less employment of a legal type, then I agree with you. But these policies, like the incarceration rate, are the socially constructed political externalities, by definition, I should think. It’s all a balance; there are costs for each of the choices we make as a society. If there were a silver bullet, we’d have found it, and then we’d happily escape the Malthusian trap, right?

46

Tim Wilkinson 06.01.09 at 11:28 pm

Why isn’t crime economic activity? Especially, say, drug dealing?

47

TGGP 06.02.09 at 3:56 am

This is the kind of discussion I’d like to see more of. I’ve got a post about the topic at my own blog.

48

lemuel pitkin 06.02.09 at 6:53 am

So your general argument is of the form if people aren’t working due to X, and there is a convincing link from labor markets to X, then X should be added to the unemployment rate?

Not convincing.

1. People leave the labor force for lots of other reasons — disability, care of family members, early retirement, premature death, etc. — that are at least as directly related to labor institutions as incarceration. Adding only the prison population seems ad-hoc, it’s special pleading. If young men had better job opportunities, perhaps fewer would be in prison (but see below), but the same goes — actually it goes farther — for young women who might choose not to be stay-at-home parents, for older workers who might choose to delay retirement, for disabled workers who could have found less physically demanding work, etc.

2. Labor markets are one driver of prison populations, but not the dominant, let alone the only one. The incarceration rate in the US is far higher than can be explained by labor market factors, as is the gap between black and white rates. Your argument comes dangerously close to an assertion that the US has a big population of prisoners because it has a big population of criminals, which is absolutely not the case — for the non-violent crimes that account for the bulk of the prison population, American rates are comparable to other industrialized countries. Or to look at it the other way, reform the drug laws and you could sharply decrease the prison population. Would you really want to call that a labor market change?

3. Unemployment plays a specific role in macroeconomics, especially since Keynes, as a measure of the slack capacity in the economy and of the cost of job loss. Incarceration simply isn’t relevant there (or at least you haven’t shown it is, or even really tried to). Given the importance of measuring unemployment as it’s generally understood, it’s just muddying the water to try to turn it into a catchall for bad labor market outcomes in general.

I understand where you’re coming from here. You think the superiority of the US labor model is overstated, and you have an intuition that there is a deep connection between US labor institutions and the high incarceration rate here. Personally, I’m totally with you on the first point and halfway with you on the second (altho not to the extent of seeming to deny, as you do, that racism, the fearfulness and punitiveness of American culture, and other factors play an independent role in incarceration rates, totally apart from labor market institutions.) But you’re going up a blind alley here.

49

lemuel pitkin 06.02.09 at 7:03 am

Or to put it another way, should we imagine this exchange?

X: France should adopt American labor market rules, like flexible workweeks and employment at will. Then their unemployment rate would fall to American levels.
JQ: Probably so, but their prison population would also rise to American levels.

If that’s not what you would say, then these posts are irrelevant to the larger debate, which is precisely about whether moving toward flexible labor markets can lower unemployment. But if it *is* what you would say, without France also adopting the American criminal justice system, then that’s a rather extraordinary claim that would take some extraordinary evidence to back it up.

50

John Quiggin 06.02.09 at 7:20 am

LP, point 1 is a desperate stretch, in my view. The idea that decisions on child-raising are, in general, analogous to poverty-driven decisions to engage in crime is not one I can engage with.

On point 2 and your next comments, I claim that incarceration is effective in reducing crime, certainly through incapacitation and probably also through deterrence. So, my claim is that, if France adopted US labor market policies, but not the US criminal justice system, it would experience higher crime. If it added US criminal justice, this would translate into higher incarceration.

On point 3, as you say, unemployment measures have a range of purposes. If you want to measure labor market slack, then (unless prisoners are forced to do productive work, and available to compete with non-prisoners) incarceration doesn’t count. But, I doubt that any unemployment or employment measure will be of much use in international comparisons of labor market slack. And on the other hand, if you are measuring over time within one country, the precise choice of measure is unimportant. That’s why, with all its faults as a measure of labor market performance, the standard measure (did no work last week, actively looking, ready to start immediately) is a pretty good guide for this purpose.

51

lemuel pitkin 06.02.09 at 1:03 pm

So then, would you accept as a modification as your thesis that differences in incarceration rates should be counted in international unemployment comparisons *insofar as they are explained by differing crime rates*?

52

mpowell 06.03.09 at 9:59 am

I find LP’s argument very persuasive here. I am very curious to know what John means in the second paragraph of 50. LP referred to France adopting American employment practices, but did not mention American social insurance policy. These deserve to be looked at differently. In my view, the only way forward in this argument is for John to argue that the lack of social insurance in the United States leads to lower unemployment and more incarceration. It’s hard to see how an unemployed French worker would start committing more crime because he was fired at will instead of just being unable to find a job.

53

Barry 06.03.09 at 10:27 am

mpowell: “The result is 99% social policy and 1% labor policy. Caplan is right to not want to let a measure of social policy get involved in the debate.”

Well, to many people who’ve not been enlightened by right-wing economics, labor policy and social policy are rather closely related.

54

Henri Vieuxtemps 06.03.09 at 10:43 am

If France were to adopt American employment practices, they would’ve faced widespread civil unrest. Even with the labor practices they have now, it’s apparently not uncommon for employees to kidnap their executive and held him hostage until their demands are met; and there’s no retaliation, no criminal charges, the government doesn’t get involved.

So, if the government there decided to follow through with American employment practices, their incarceration rate would’ve probably exceed the current US level.

55

JoB 06.03.09 at 11:47 am

If pigs could fly …

Luckily the reverse is quite conceivable … if the US were to adopt French employment practice. In fact, it seems unconceivable that in the long run the US were not to adopt real social security.

I don’t know whether unemployment would drop but happiness would increase. There may not be social unrest in the US recognizable as such (but the paradigm examples of social unrest are, after all, European still) but that doesn’t mean that criminality isn’t in part caused by the lack of social security. A cause need not be direct for it to be a root cause.

56

mpowell 06.03.09 at 12:59 pm


Well, to many people who’ve not been enlightened by right-wing economics, labor policy and social policy are rather closely related.

I’m not sure what you’re trying to insinuate, but issues like at-will employment and corporate-union relations are pretty distinct from policing methods and education in inner-city neighborhoods. They may be tied in a larger conceptual political framework, but we can talk about their impacts separately where possible. The difficulty lower class african americans have in avoiding prison and finding employment have a lot more to do with their education and socializations, not whether the jobs the entry level jobs available to them have union benefits.

55: You’re mixing labor flexibility with social insurance. I can’t see any reason that reducing labor flexibility in the United States would reduce incarceration rates. The debate that Caplan seems to want to have is whether it would increase unemployment. Social insurance may be tied to both incarceration and employment, though.

57

mpowell 06.03.09 at 1:01 pm

55: I’m also quite curious why you think this is inevitable in the United States. In the 90s we limited well fare benefits and there is nothing in the modern political discourse to suggest that people think giving poor people benefits is worth spending tax dollars on. Are you thinking there will be a long term play there as the R party completely melts down and the entire discourse changes?

58

Barry 06.03.09 at 1:02 pm

“They may be tied in a larger conceptual political framework, but we can talk about their impacts separately where possible.”

That’s the whole point. They aren’t really separable from their common causes. Do you think that the fact that race is a rather good predictor of both incarceration and job finding/keeping/salaries is just a spurious correlation?

59

engels 06.03.09 at 1:39 pm

If you’re in prison you have somewhere to sleep, food, medical care. Some of the effects of unemployment are similar to being in prison: social isolation, the contempt of the decent, hard-working majority, becoming a social and moral ‘problem’ for economists, psychologists and philosophers to ‘solve’. So with only some exaggeration it could be turned around: unemployment numbers should be counted towards the size of the prison population.

60

JoB 06.03.09 at 2:10 pm

56-57: you’re right, I was mixing. I don’t think that the labor flexibility is the root cause. You’ll find many flavours of flexibility in Europe and the only common element here really is a social security which has many versions as well but is broadly similar in its output.

I think it is inevitable because it is the superior system (the one with social security) & not only a more humane system. I don’t think social security ever was at the European level in the US, I think (but don’t know) it never ever reached the kind of irreversibility it certainly has here.

Yes, I’m sure there has been regression from 70’s system onwards in the US but mainly because it wasn’t strong enough at that point in time (strength of unions have nothing to do with it imho -it’s just addictive for citizens to know that whatever happens, it’s not ‘lights out’ in one blow).

59: unemployment is not that bad, really, try both and compare!

61

mpowell 06.03.09 at 2:21 pm


That’s the whole point. They aren’t really separable from their common causes. Do you think that the fact that race is a rather good predictor of both incarceration and job finding/keeping/salaries is just a spurious correlation?

No, there are things here that just aren’t related. At will employment and low union strength in the United States has very little to do with incarceration outcomes for poor african americans. You’re just lumping things together incoherently when you talk about labor market outcomes for this class of people and arguing that everything that effects those outcomes is then inherently related to all kinds of labor market policy. There are policies that influence the balance between capital and labor and drive things like the relationship between unemployment and inflation. These things can be separated in their effects from social policies that ultimately influence the ability of poor african americans from finding employment.

62

Tim Wilkinson 06.03.09 at 3:31 pm

mpowell @61: Means-tested transfer payments aren’t part of labour-market institutions. Exclusion of the incarcerated from the job market is, as, on free-market principles, is the prevention of (at least) drug-dealers, blackmailers and prostitutes from plying their trade. Correct?

63

engels 06.03.09 at 6:47 pm

Jo, if my comment had claimed that unemployment and imprisonment were equally bad then your response might have made sense, although it would still have been presumptuous, smug and rather offensive.

64

Matthew Lillard 06.03.09 at 7:14 pm

I don’t understand why there has to be a hard causal link between the policies you are comparing and incarceration rate to agree with John’s approach. All you should have to illustrate is that the prison population is disparate enough to skew any comparison of the employment rate. If there were other areas of the population that would skew the employment rate results, say comparing a country that does not employ women to the US, those should also be looked at and discounted (I know you have discussed this issue). I would think the argument should be how much of the prison population to discount. All vs. a percentage (you could even argue more than 100 percent, that one person in prison takes away multiple jobs as it may hamper the flexibility of a spouse or family member to find or maintain employment and they would then not count in the unemployment rate as they drop out.)

The obvious problem with these statistical comparisons is that there are too many variables and you are trying to isolate certain labor policies and saying those are the determining differences.*

As an architect the only statistics I come close to is from the real estate sector and would approach this more like a bet on differing appraisals vs. final market outcome. You should be able to each try and agree on all other factors that alter the final numbers beyond “the policy’. Look at factors, other than the policy you are betting, that skew the result and provide a discount or surcharge. This is, I guess, John’s point. I just don’t see how prison population is the only other factor.

*I live in Oregon and we are always high on the unemployment rate, I would argue that is partially cultural. People just don’t leave when the market says they should. When I saw the writing on the wall for my job in Dallas (2001) I left the state immediately, nothing else (other than financial reasons) was keeping me there. I might stay and look for a while if that happens here. I would argue that longer periods of unemployment are not in itself negative. If there is greater flexibility to stay unemployed, one might be able to maximize their potential by waiting to find that better job. If one is desperate to take the first that comes along, that may lead to underemployment.

65

engels 06.04.09 at 2:07 am

JoB: ‘offensive’ was over the top, sorry. I usually appreciate your comments, but I don’t really get what you are saying here. All I did was point to some ways in which the two are similar and I don’t think this licenses any inferences about what I have or haven’t personally experienced. As I see it, involuntary unemployment and imprisonment are both forms of forcible exclusion from social life, which invariably involve some of the same serious harms to the large group of people who are so excluded: social isolation, substantial losses of freedom, damage to self-development, moral stigma, dehumanising living conditions and treatment by the state. They are both essential features of a capitalist economy which involve inflicting severe harms on an out-group in order to produce benefits enjoyed by others: protection of property rights (and general deterrence of crime) in the one case and deterrence of inflationary wage claims in the other. Clearly imprisonment is the more extreme but it seems insensitive to me to say that unemployment is ‘not that bad’.

66

lemuel pitkin 06.04.09 at 3:43 am

Here’s the thing: With the exception of homicide, American and European crime rates are essentially the same.

Here are one-year victimization percentages for the main categories of crime reported in the 2004-05 International Crime Victimization Survey [1] for the US and the six largest EU countries:

Country AutoTheft Burglary Theft Robbery Assault
USA 1.1 2.5 4.8 0.6 4.3
England&Wales 1.8 3.5 6.3 1.4 5.8
France 0.6 1.6 3.3 0.8 2.1
Germany 0.2 0.9 3.0 0.4 2.7
Spain 1.0 0.8 2.1 1.3 1.6
Italy 1.0 2.1 2.4 0.3 0.8
Poland 0.7 1.4 3.5 1.3 3.0

As you can see, the US is at the high end of the range but usually not the top, and certainly not an outlier. And even compared with the lower-crime EU countries, Germany and France, the difference in crime rates is generally about two to one. Whereas the difference in incarceration rates is on the order of ten to one.

So insofar as the link between labor market policies and incarceration is supposed to be via crime rates, there is no link. Because essentially none of the differences in incarceration between the US and Europe are explained by differences in crime rates. Labor markets in the US and in Europe lead people to “choose” crime at approximately the same rates; it is the criminal justice system here that then turns equivalent crime rates into much higher incarceration rates.

I think that’s check and mate, as far as the argument of this post goes.

[1] Franklin Zimring and Gordon Hawkins have a good discussion of the reliability of this kind of survey data in Crime is Not the Problem and show you can see essentially the same pattern in official national data.

67

lemuel pitkin 06.04.09 at 3:46 am

Sorry, maybe this will be easier to read:

Country……Auto Theft……Burglary……Theft……Robbery……Assault
USA……1.1……2.5……4.8……0.6……4.3
England&Wales……1.8……3.5……6.3……1.4……5.8
France……0.6……1.6……3.3……0.8……2.1
Germany……0.2……0.9……3.0……0.4……2.7
Spain……1.0……0.8……2.1……1.3……1.6
Italy……1.0……2.1……2.4……0.3……0.8
Poland……0.7……1.4……3.5……1.3……3.0

68

Righteous Bubba 06.04.09 at 3:57 am

The code tag should produce nice regular spaces.

Country Auto Theft Burglary Theft Robbery Assault
USA 1.1 2.5 4.8 0.6 4.3
England&Wales 1.8 3.5 6.3 1.4 5.8
France 0.6 1.6 3.3 0.8 2.1
Germany 0.2 0.9 3.0 0.4 2.7
Spain 1.0 0.8 2.1 1.3 1.6
Italy 1.0 2.1 2.4 0.3 0.8
Poland 0.7 1.4 3.5 1.3 3.0

69

Righteous Bubba 06.04.09 at 3:58 am

Rats. Does not actually honor spaces.

70

JoB 06.04.09 at 8:29 am

engels, First: thanks for your kind first sentence in 65! Second, when I read your first reply last night (too late to respond also because I needed to think it over): I could see how my remark at 60 could be easily construed as a cheap snipe. I apologize. I hope it did not hit on personal stuff you encountered in your context, I really do.

As to the content of it – I take objection with your analysis of unemployment. True – for people willing to work but barred from it by circumstance some of the negatives you associate are apt. But this association is not a necessary one as you seem to make out. Full employment is not the greatest good whilst being stripped from one’s freedom surely is the greatest bad. If capitalism goes astray it certainly doesn’t go astray in maximizing the out-group, and the analysis that the communist alternative through full employment is a solution is one I cannot share. On the one hand because it minimizes the in-group (apparatsjiks is the term). On the other because forced employment ethics is surely worse than living with more unemployment.

My point is that unemployment (broadly speaking, including leisure and people working in arts and research in a fully independent, possibly non-commercial, way) is good but that such good needs to be equitably distributed (as any other good). Equitable can mean that some – willing to work – work more than the average and others – unwilling to work – work less than the average. Not all of us need social context, not all of us need work to get social context if we need it.

I can understand that all of this seems like statements coming from a Martian recently landed. I know I have not thought it through enough. But I also know that communism & capitalism have a common desire to ‘optimize the labour market’ and a hunch that that’s where both fail modern man.

71

mpowell 06.04.09 at 10:13 am

64: No, I don’t agree with your example. If a country systematically excludes some portion of its potential work force that is obviously significant in some sense. But the whole point of this exercise comes from an understanding that we expect the pool of available unemployed workers to have some impact on the rate of wage growth and therefore inflation. A stable economy therefore requires some percentage of unemployment. And the proposition is that we expect different labor rules to lead to a different relationship. You can’t automatically include a systematically excluded group as part of your unemployment measurement then as a test of this hypothesis. So it is necessary as part of this debate to talk about whether it is really appropriate to include one excluded group or another based on what we are trying to measure. I think John might be targetting a slightly different measure than Bryan, and maybe that should be more explicit, but the argument that incarceration should be counted as unemployment for these purposes truly requires an awful lot of work.

72

Henri Vieuxtemps 06.04.09 at 10:31 am

A stable economy requires unemployment, but also, I think, a low-wage/flexible-labor-market economy requires higher rates of incarceration among the unemployed, non-portable medical insurance and other stuff like that.

You have to be incentivized to take an unappealing low-paying job, otherwise facing a not too far-fetched prospect of ending up in jail, which is really (or so I heard) much more unpleasant than just being unemployed.

These are all integral parts of the program, not some random arbitrary pieces.

73

John Quiggin 06.04.09 at 10:33 am

LP your entire analysis assumes that incarceration has no effect on crime, including no incapacitation effect. The correct way to interpret the data is “with massively higher incarceration rates, the US has average crime rates”. Hence, unless incarceration is ineffective, if the US had the same incarceration rate as Europe it would have much higher crime rates.

More generally, given the underlying social institutions, including labour markets, the US faces a choice between higher crime rates, higher incarceration rates or both. Fortunately for our statistical purposes, the difference in incarceration rates is large enough to cancel out effects tending to increase crime, so we can use this difference as a measure of the bad outcomes produced by the institutions.

The only remaining point at issue is the claim that the institutions that produce the problem are neither labour market institutions nor incarceration policies but something else, such as deeply entrenched racial inequality. Black imprisonment rates are exceptionally high, but even the white rate is four times the rate for European countries (not disaggregated by race etc). And, obviously much of the black-white difference is due to poverty. Blacks are three times as likely as whites to be poor, about six times as likely to be imprisoned.

74

Tim Wilkinson 06.04.09 at 11:15 am

FWIW, using ^table^ markup might work, though marking up manually would be a pain:

Country                  Auto Theft   Burglary     Theft           Robbery      Assault
USA1.12.54.80.64.3
England&Wales1.83.56.31.45.8
France0.61.63.30.82.1
Germany0.20.93.00.42.7
Spain1.00.82.11.31.6
Italy1.02.12.40.30.8
Poland0.71.43.51.33.0

the figures measure the incidence of perceived victimhood rather than the patterns of offending, though. And how much crime against businesses, or victimless (complicit anyway) ones like drug-dealing would show up in a poll of householders?

BTW the US isat the top, if you leave out the anglo-model UK.

Rather than trying to disentangle the causal influences (as one would want to) on the much higher imprisonment rate (e.g. the preventive effect of imprisonment, greater punitiveness v greater seriousness/recidivism/professionalisation), wouldn’t you be better off directly looking at the relationship between long-term unemployment and imprisonment for economic crime?

75

Tim Wilkinson 06.04.09 at 11:26 am

It didn’t like table tags.

Country              Auto Theft  Burglary   Theft      Robbery     Assault    
USA                  1.1         2.5         4.8         0.6         4.3
England&Wales        1.8         3.5         6.3         1.4         5.8
France               0.6         1.6         3.3         0.8         2.1
Germany              0.2         0.9         3.0         0.4         2.7
Spain                1.0         0.8         2.1         1.3         1.6
Italy                1.0         2.1         2.4         0.3         0.8
Poland                0.7         1.4         3.5         1.3         3.0

76

Tim Wilkinson 06.04.09 at 11:27 am

or that. Sorry…looked fine in preview

77

Henri Vieuxtemps 06.04.09 at 12:09 pm

You need to unescape the ampersand.

78

Tim Wilkinson 06.04.09 at 12:34 pm

The server escaped them for me. A combination of code tags to give fixed-width font with spacers such as dots that won’t get messed up by ‘helpful’ server-side transformations would have done it. But I have made such a terrible mess, as well as posting my comments without checking that they hadn’t been superseded in the interim, that I will just slink away shamefacedly and get on with what I am supposed to be doing.

79

lemuel pitkin 06.04.09 at 1:50 pm

your entire analysis assumes that incarceration has no effect on crime, including no incapacitation effect.

The correct way to interpret the data is “with massively higher incarceration rates, the US has average crime rates”. Hence, unless incarceration is ineffective, if the US had the same incarceration rate as Europe it would have much higher crime rates.

So John, let’s say we look at the data and see –a s we do — very wide variations in incarceration rates among countries with roughly comparable crime rates. We can interpret this in two ways:

(1) Crime rates are not a major determinant of incarceration rates.

(2) Each country has hit on exactly the right level of incarceration to reduce very different crime rates to a similar level.

You’re saying we should prefer explanation 2. Why?

80

lemuel pitkin 06.04.09 at 1:59 pm

the figures measure the incidence of perceived victimhood rather than the patterns of offending, though. And how much crime against businesses, or victimless (complicit anyway) ones like drug-dealing would show up in a poll of householders?

The advantage of the victimization survey is that it is a single survey, administered the same way in every country. Other crimes have to be tracked using national statistics which are harder to compare. But for our purposes, leaving those other crimes out doesn’t matter, because the type of crimes covered account for the bulk of the prison population.

BTW the US is at the top, if you leave out the anglo-model UK.

Well, first of all, the conventional view is that UK labor market institutions are intermediate between the US and Europe. Second, again, the question is how much. Yes, the US is near the top (in fact several smaller European countries, such as Ireland, are consistently higher), but not by anywhere near the margin it is for incarceration rates. Again, John’s argument is not just that there is some link from labor market institutions to crime rates, and some link from crime rates to incarceration rates; it requires that those links be the predominant influence on incarceration rates. And the empirical evidence just does not support that.

81

JoB 06.04.09 at 2:05 pm

John,

It seems to me that the gist of your point could be restated as: absence of social security leads to increased need for repression which in turn increases the incarceration rate (Henri’s version). It is then unfair to claim lower unemployment as a sign of superiority of economic system (maybe even lemuel pitkin could agree to this rephrasing if he steps out of his chess metaphor).

This sounds fair enough (although maybe less eye-catching than your original phrasing). But it’s not the same as what I understood to be your original contention that flexible labor markets are the underlying issue. In fact, lack of social security is the root cause.

An alternative analysis of the issue might in fact be that, to be fair, one needs to add ‘hamburger job’ employment (so characteristic for the US model) to the unemployment rate. Indeed – these jobs are like incarceration & none of these jobs would exist in case of decent social security. It’s also a misleading analysis because it tinkers with the base notion of (un-)employment deviating the discussion away from the root (social security) and onto secondary notions (employment).

In fact, I don’t see why it would be impossible to have a highly flexible labor market and a good social security. In theory, a good social security would allow for a more flexible labor market & the combination for good flexibility on the part of the potentially employed i.e. a context for an improvement in labour conditions.

82

lemuel pitkin 06.04.09 at 2:07 pm

I am going to present some more data on this shortly. (It would be interesting to see some data from John supporting his strong labor-crime and crime-incarceration links.)

But one other point. The “left” side in this debate seems to have become identified with John’s view. But politically, there is nothing progressive about his position, since his argument is just as much a defense of the US criminal justice system, as it is a criticism of US labor markets. As someone who thinks that the US prison system is an appalling, vicious institution, I think we should be very wary of assuming — as John, so far without evidence and contrary to most progressive criminologists, is doing — that it is really effective, and even in some sense necessary, in reducing crime.

83

engels 06.04.09 at 2:29 pm

JoB – I think we are talking partly at cross purposes. By unemployment I mean involuntary unemployment. You appear to be referring to what I would call leisure. I don’t think it’s at all crazy and in fact I agree that capitalist economies, and previously existing socialist economies, suffer from a bias towards to consumption (and therefore labour) at the expense of leisure. I am not in favour of forcing anyone into work and I believe that government should support those who do not work (whether through circumstance or choice) as a human right. But I believe that someone who isn’t a criminal, who wishes to work and is prevented from doing so by the nature of our collectively chosen social arrangements suffers from a serious injustice, perhaps not as serious an injustice as wrongful imprisonment but one which bears important points of comparison with it.

84

lemuel pitkin 06.04.09 at 2:35 pm

From Michael Tonry, Why Are U.S. Incarceration Rates So High?:

***

in the United States, 1 resident per 150 is imprisoned; that is 6 to 12 times the rate in other Western countries …
The first explanation for why so many Americans are in prison, that our crime rates are higher or faster rising than other countries’, has virtually no validity. Crime rates in the United States in the 1990s are, for the most part, no higher than in other Western countries. … Gun violence is [higher]; however, fewer than a fourth of those sentenced to prison are convicted of violent crimes of any type, so this cannot be why U.S. prison patterns and penal policies are so different. …

If higher crime rates do not explain American exceptionalism, perhaps crime trends do. Perhaps … that is why the number of people locked up has increased by five times in the past quarter century, from about 300,000 in 1972 to 1,802,496 in mid-1998. … Figure 2 and Figure 3, which show comparable data for Finland and Germany during the same period, indicate that there is no such necessary connection. Other countries could have been used for the comparison, but the United States would still be an exceptional case (Kuhn 1997). Although the homicide and violent crime curves in Finland and Germany rose as steeply as the U.S. curves shown in Figure 1, the imprisonment rate in Germany fell throughout the 1960s and remained roughly level thereafter, and the incarceration rate in Finland fell sharply and steadily throughout the entire period. …American imprisonment rates did not rise simply because crime rates rose. They rose because American politicians wanted them to rise. …

As Figure 4 shows, crime rates for most crimes peaked around 1980… However, the first three-strikes law was enacted in 1993, and the federal truth-in-sentencing law, which authorized $8 billion for state prison construction, was passed in 1994. The meanings of these data are complex, but whatever else they show, they do not show any simple interaction between crime trends and imprisonment patterns.

***

Tonry is Director, Institute of Criminology, Cambridge University, and Sonosky Professor of Law and Public Policy, University of Minnesota. So if he says there’s no link between crime rates and incarceration rates, you don’t have to take his word for it, but you might wnt to adjust your priors a bit in that direction.

85

engels 06.04.09 at 2:40 pm

Hey, what happened to my comment?

86

lemuel pitkin 06.04.09 at 2:58 pm

It is then unfair to claim lower unemployment as a sign of superiority of economic system (maybe even lemuel pitkin could agree to this rephrasing).

I do accept it. I’ve said so repeatedly. One of the things that frustrates me here is that John is making a weak (well, false) argument in support of a correct & important overarching position. Those are the errors we have to be most vigilant for.

87

engels 06.04.09 at 3:01 pm

Okay, it was the dreaded ‘S’ word obviously. But why does it vanish rather than going into moderation?

88

lemuel pitkin 06.04.09 at 3:05 pm

O rather, I accept what I quoted. But is just is not true that that the US incarceration rate is a necessary result of the inadequate safety net here. Nor has John presented any evidence for this — he’s just said that “Chicago school” economists ought to think it (which, since none of us are they, is irrelevant) and he’s asserted an a priori belief that incarceration must reduce crime rates.

89

lemuel pitkin 06.04.09 at 3:15 pm

given the underlying social institutions, including labour markets, the US faces a choice between higher crime rates, higher incarceration rates or both. Fortunately for our statistical purposes, the difference in incarceration rates is large enough to cancel out effects tending to increase crime, so we can use this difference as a measure of the bad outcomes produced by the institutions.

If this tradeoff exists, it ought to be visible in the data, no?

90

lemuel pitkin 06.04.09 at 3:40 pm

The other side of this is that while John’s argument gives too much credit to American prisons, it actually, paradoxically, gives too much credit to American labor institutions as well. Because he is conceding that flexible labor markets and a minimal safety net really do deliver lower unemployment, albeit at a cost of higher crime. Whereas the empirical evidence — I’ll be happy to provide as many cites as you want — show that flexible labor markets don’t work even in that limited sense. All of Europe’s higher unemployment is explained by restrictive monetary and fiscal policy and, in general, slower demand growth; none of it is explained by labor market institutions.

91

Tim Wilkinson 06.04.09 at 7:08 pm

Prison doesn’t incapacitate people from comitting crime. As I understand it, crime in prisons is rife but rarely (officially) reported and, I would guess, not generally included in ‘victim’ surveys.

I still don’t see why the exclusion of people from the workforce by putting them in prison (and not letting them get outside jobs) doesn’t count as a labour market restriction, especially when it’s effectively a policy decision how many are to be banged up.

92

lemuel pitkin 06.04.09 at 7:17 pm

I still don’t see why the exclusion of people from the workforce by putting them in prison (and not letting them get outside jobs) doesn’t count as a labour market restriction

Tom,

Look again at the tile of this post. It’s “Incarceration as a labor market outcome“. Totally different claim.

93

Tim Wilkinson 06.04.09 at 7:17 pm

@87 I don’t think it was that – unless either you’re on some sort of grey-list, or the censor-bot is eccentrically configured. I put probably the ‘worst’ word of all on the malhotra thread and it sailed right through.

94

engels 06.04.09 at 7:41 pm

The ‘S’ word in question isn’t a primal human activity but a final stage of human development.

95

John Quiggin 06.04.09 at 8:39 pm

“Prison doesn’t incapacitate people from comitting crime. As I understand it, crime in prisons is rife but rarely (officially) reported and, I would guess, not generally included in ‘victim’ surveys.”

This is an important point, and one I meant to make myself. Please substitute “reported crime” for “crime” anywhere I’ve written about it above.

LP, as regards authorities for my position, I’ll appeal to Bruce Western above. On your general point, nothing I’ve written excludes the conclusion that the gap between EU and US unemployment in the 1990s is more than fully explained by restrictive macro policy in Europe and unsustainably expansionary policy in the US. I’ve previously asserted that this is at least part of the explanation, which is why I’m willing to take the bet.

96

Tim Wilkinson 06.04.09 at 8:43 pm

@94 I getcha. I just had one moderated myself and don’t know why. I think it might be a posting frequency restriction.

@92 Yes, I had got that far at least. It just seemed somehow relevant to the question whether they should be added to the unemployment figures, but I suppose it’s a fairly separable issue, especially if you don’t include drugdealers, pimps, gangs that enforce rough justice in police no go areas, etc. in the employment figures. I don’t know. Just chucking in a few mutated ‘memes’.

97

lemuel pitkin 06.04.09 at 9:38 pm

John Q.,

I’m a big fan of Bruce Western too. Absolutely. But where does he say that American labor institutions lead to higher crime, that higher crime is the reason for higher incarceration, or that high levels of incarceration reduce crime? Those are the three claims you’ve made in this thread. Western does not make any of them, at least in the article cited here.

It is true he calculates a modified unemployment rate counting the incarcerated as unemployed. He also calculates one doing what I and others here have suggested — adding some fraction of the prison population to the measured unemployment rate based on estimates of their employment prospects if not imprisoned. (The number he uses is 0.36 which obviously produces a number closer to Caplan’s preferred 0 than to your preferred 1.) But yes, he definitely suggests that our seemingly low unemployment rate is misleading because of our high incarceration rate.

But this post is making a much more specific claim — that the American labor model doesn’t just look misleadingly successful because of mass incarceration, but actually causes mass incarceration. And on that claim, Western’s position is close to the opposite of yours: he explicitly denies that mass incarceration is the result of higher crime rates and argues that it exacerbates the problems with the American labor model (by creating a class of ex-felons weakly attached to the labor market) rather than compensates for them.

So no, Western is not an authority for your arguments here. He *is* an authority for the broader view that there is some connection between unemployment and imprisonment, and that the US employment performance is not as good as it looks because of mass imprisonment.But the fact that there is some link between two things doesn’t mean that any particular link exists. And the particular link you’re making gets no support from Western.

98

John Quiggin 06.05.09 at 3:01 am

LP, you’ve mis-stated my position particularly in para 3 of your last. You say “this post is making a much more specific claim—that the American labor model doesn’t just look misleadingly successful because of mass incarceration, but actually causes mass incarceration.”

I’m saying the first and not the second (if I understand what you mean by “cause”), and I agree entirely with Western.

I’ve run out of patience on attempting to clarify my position. Perhaps if you could clearly state your own I would have some idea of what we are disagreeing about.

99

JoB 06.06.09 at 8:26 am

engels@83, OK, yes, cross purpose it was. I don’t take objection to consumption and I don’t see its necessary link with forced labour. Why should more leisure not give rise to more cosumption? Anyway, unemployment in your sense is less of a problem when unemployment in my sense is a real option and it’s a pity that this discussion did not in the main take into account that variable.

Continental European unemployment numbers are to some extent the result of people unwilling to work (or unwilling to work in conventional ways) preying on social system and higher unemployment is not a sign of systemic weakness (as John implies together with the current establishment consensus of job creation at all cost). Similar for pitkin, a worse safety net does lead to higher incarceration and better employment numbers & it is still worse despite being better at employment numbers.

Henri still had imho the most cogent position of the thread.

100

Chris 06.08.09 at 4:03 pm

@engels: I’m not sure what you mean by a final stage of human development, but if it’s a political S word often associated with your alias, you might consider that it contains as a substring the name of a drug often hawked by spammers. Perhaps this site has some variation of the Scun-thorpe problem? (Remove the hyphen to test for another manifestation of the problem, which is named for a small town in England which just happens to have an obscene four-letter word as a substring.)

Comments on this entry are closed.