Stumbling and Mumbling

Police & the law

Three items:
1. Two PCSOs do nothing to prevent a boy drowning in a pond in Wigan. Senior officers say they acted correctly, in line with health and safety rules - though such rules are applied less fastidiously, it seems, when it comes to people dying in custody.
2. The Cambridgeshire police force that claims it can't afford to fight the rising crime caused by immigrants spent £500,000 on luxury cars for senior officers.
3. Home Secretary Jacqui Smith is refusing to cooperate with a US inquiry into corruption at BAe Systems.
There's a link here.
Let's say your Bayesian priors were Marxist-cum-public choice ones. The rule of law is a fiction which advances the interests of the powerful - hence the failure to investigate allegations against BAe. The notion that the police are public servants is also a lie, which serves to legitimate laws which repress ordinary people whilst allowing wrong-doing in boardrooms. In truth, the police are agents of the ruling class who aim to split the working class by stirring up racism whilst feathering their own nests.
Don't these stories strengthen these priors? What contrary evidence weakens them?

Crime: Polly's right

It's standard practice to attack Polly Toynbee for her ignorance of economic principles. The reactionary snob, then, deserves congratulating for taking the opposite line. He lays into her for  understanding economics.
Polly says:

We liberal commentators point as usual to overwhelming evidence that crime and violence thrive most in the most unequal societies. If 30,000 left school this summer with no GCSEs and 1.2 million unemployable youths have gone missing altogether from education or work, what do you expect?

The RS asks whatever happened to personal responsibility? We shouldn't, he says, excuse criminals "by saying 'well, he didn't have any GCSEs, he didn't get to go to Center Parcs, half his classes were taught in portakabins'. It is his responsibility."
True. Crime is the responsibility of the criminal. But this is entirely consistent with Polly's claim that inequality contributes to crime, which in turn is consistent with the standard economic view of crime, as described by David Friedman:

Economists approach the analysis of crime with one simple assumption—that criminals are rational people. A mugger is a mugger for the same reason I am an economist—because it is the most attractive alternative available to him. The decision to commit a crime, like any other economic decision, can be analyzed as a choice among alternative combinations of costs and benefits.

Now, if people are leaving school with no skills, crime pays for them - it's a rational choice because the benefits outweigh the costs.
In this sense, unequal societies are likely to be crime-ridden - because they have some people for whom crime is an attractive career options, and others who have stuff worth nicking. The "overwhelming evidence" Polly refers to can be found here  here (pdf) here (pdf) here (pdf) here (pdf) and here (pdf) , among other places, but is questioned here (pdf).
To assert a link between inequality and crime does not mean we deny any role for personal responsibility. Exactly the opposite. The link exists precisely because people can choose their own actions - this is what it means to respond to incentives.
Nor does such a link  rule out tough sentences. Quite the opposite. because people respond to incentives, it's possible that long sentences will cut crime effectively.
Indeed, it's perfectly consistent to argue that although inequality causes crime, it's more cost-effective to impose tough sentences rather than reduce inequality. Whether this is actually true, however, is an empirical question requiring research. It can't be answered by bleating from either liberals or the right.

Does punishment deter?

Blair's review of crime policy proposes more non-custodial sentences but not, generally, tougher punishments. It's probably an accident, but this has some economic justification, because tougher punishments might not deter.
The reason for this is simple. It's expensive to impose tough sentences; it costs around £40,000 a year to keep someone in stir, and £100,000 to build a prison cell.
Because it's costly to impose sentences, criminals might believe the government won't enforce them. And this expectation of less enforcement might raise crime, thus offsetting the tendency for tougher penalties to  deter it. This point was originally made by George Tsebelis, but it's expanded in this recent paper (pdf) by Rimawan Pradiptyo.
It's easy to see how there's a trade-off between punishment and enforcement. When sentences are long, prison overcrowding leads judges to let criminals off. In the 18th century when people could be hanged for trivial offences, juries were reluctant to convict. In Islamic law, serious  punishments are only inflicted after very high standards of proof are met.  By contrast, lighter punishments can be more rigorously enforced, because it's cost-effective for the enforcer to do so - look at speeding fines.
There's another way in which tough sentences can raise crime. Sending drug-dealers to prison for a long time raises the monopoly profits of other dealers, thus increasing incentives to deal.
These considerations suggest longer prison sentences might not be cost-effective, so the government is right to consider cheaper options which increase the probability of enforcement.
This doesn't, however, mean the review is entirely valid. For one thing, one form of cheap punishment - corporal punishment - is not even considered, despite its theoretical appeal. And for another, the spread of non-custodial sentences, use of special units for mentally ill prisoners, asset seizing and attempting to identify criminals in advance represents an extension of the illiberal managerialist-therapeutic state.

Economics of crime: a clarification

Two commenters have objected to my claim that people commit crime because the benefits exceed the costs. Crime has risen, they say, because the cultural norm against crime has weakened.
In fact, we're all right on this. One cost of committing crime - the biggest for many of us - is that our conscience will sting us if we do so. I don't dispute that one reason why crime is higher now than in the 30s is that, for some people, conscience - what Adam Smith called the impartial spectator - stings less.
However, it doesn't necessarily follow that the best way to cut crime is to strengthen people's consciences. The belief that solutions to problems must resemble causes is a version of the representativeness bias. If a man's been run over by a bus, you don't restore him to health by merely reversing the bus.
It might be impractical to change culture, especially in anything except the very long run. Instead, the economic approach suggests we can cut crime by tweaking other costs and benefits. Increased punishment, increased probability of punishment, more options to earn a legitimate living are all options to think about. It's an empirical question.
On this point, I think I might be a little unusual on the left. The left is sometimes accused of having an optimistic view of the potential for human nature to improve.
Not me. I'm a pessimist. One reason why I want radical institutional change is that human nature, left to itself, won't improve.   
More: Here's an introduction to economists' thinking about crime. Here are two pdfs that give more detail. Here's a load of papers on the economics of crime.

Lone parents & gun crime

Ross Clark says something that is easy to misinterpret. It's:

Those involved with gun crime tend to have grown up fatherless and in the absence of good male role models have gravitated towards gangs.

It would, however, be wrong to infer from this that growing up fatherless is a cause of gun crime.
Let's put some numbers on this. The Home Office reckon (ch 2 of this pdf) that firearms were used in 11084 crimes in 2005-06. Let's assume this means there are 11084 gun criminals, and let's assume all these were 16-24 year-old men.
Now, in 2005 there were roughly 3.5 million such men. How many come from single-mother households? According to table 2.6 of this big pdf, the proportion of children in single-mother households has risen from 6% in 1972 to 16% in 1992 to 21% in 2005. Taking the 1992 figure, this means there were roughly 560,000 16-24 year-old men who grew up in lone-mother households.
Even if we assume that all gun criminals come from such households, it's still the case that only 2% - 11084 out of 565,000 - of men from single-mother families become gun criminals. To get the figure this high, I've had to make some heroic assumptions.
So, at least 98% of men from single-mother households don't become gun criminals. It would be odd to say that x causes y when over 98% of xs don't cause y.
Indeed, there's a good theoretical reason why single mother households, on their own, shouldn't produce criminals. It's that the alternative to a single-mother family is not, in many cases, a family where the father is a good role model. It's one where the father is a bad role model. Removing a bad role model might be good for a child's development. Had I grown up in a two-parent household and used my dad as a role model, I'd have become a criminal.

Optimum crime

How much burglary should there be?
Intuition says: "none." And intuition's wrong, as this nice paper (pdf) by Michael Kosfeld and Steffen Huck shows.
The thinking here is that one key to catching burglars is the effectiveness of neighbourhood watch schemes. But if there are no burglaries, such schemes will wither away. In the long-run, that'll encourage more burglars, because the lack of neighbourly vigilance increases the chances of them getting away with it. Zero burglary in the short-term can therefore lead to high crime in the longer-run.
This has two nicely counter-intuitive implications.
1. Tougher punishment might lead to more crime.  If the government punishes burglars severely, the fall in crime in the short-run will cause a decline of neighbourhood watch schemes and hence more crime in the long-run, as criminals judge that the low probability of being caught offsets the nasty punishment. The authors say:

Full deterrence in the short-run leads to zero deterrence in the long run...Optimal punishment policy in this situation is to set punishment at an intermediate level that does not deter burglaries completely but maintains a certain crime rate in order to keep participation in neighbourhood watch active.

It is, however, an open question whether current punishments are tougher or weaker than this optimum.
2. There's a possible social benefit to crime. If it brings neighbours together in neighbourhood watch schemes, it can build social capital, and networks of friends. And friends, as Mark Vernon argues, are important.
Another thing. There's another sense in which the optimum amount of crime is positive. It's that the cost of catching the last criminal might outweight the benefits of taking him off the streets. But this is obvious.

Try wearing a cap

Contraception use can cut crime. In this paper (pdf) presented to this week's American Economic Association conference Nicholas Hill shows that there's a statistically significant relationship between greater use of contraception and decreased crime 17 years later.
The theory here is the same as Levitt and Donohue's view (pdf) - defended here (pdf) and corroborated by Romanian evidence (pdf) - that increased abortion cuts crime.
Where contraception isn't used, there'll be more unwanted births. And unwanted children are more likely to commit crime, through two channels:
1. Parents are likely to invest less in their education - through spending time teaching them or encouraging them to go to school. And people with worse education are more likely to become criminals, because they are less able to make money legitimately.
2. Unwanted children feel fewer social constraints against crime. They're less likely to worry about bringing shame on the family, or about getting a good hiding from dad, especially as he fled the scene long ago.
In reducing the number of unwanted births, contraception therefore reduces the number of criminals.
Personally, I find this intuitively obvious; Hill's skill is that he's found the empirical evidence.
So, it seems as if the advice of the mighty Specials was good. However, Hill found that the pill and sterilization were more effective at cutting crime than the cap or condoms.

Muggers' motives

Muggers, it seems, are motivated by fun rather than money. The Devil’s Kitchen reads this as evidence that Polly is wrong to claim that poverty causes crime.
I’m not sure. Put it this way. I’d find it fun to beat up public schoolboys or Guardian writers in the street. But I don’t do it. This isn’t because I’m a decent person. It’s because the cost of getting caught would be high for me, as I’d lose a well-paid job and prison would entail a significant loss of utility.
By contrast, someone in poorer circumstances might well feel that the fun of the attack outweighs the cost, as he’s got less to lose.
In this sense, poverty causes crime, even if the criminal is motivated only by kicks rather than the desire for his victims’ money.
Here’s a nice discussion of the notion of rational criminals. It was well put by the vastly under-rated Slaid Cleaves, in his song Something to Lose:

Give me something to lose
Give me something to live for
I’ve done my time and I’ll do the crime
Until you give me something to lose.

Strictly speaking, of course, Slaid is wrong. Even if poverty does cause crime, it doesn’t follow that relieving poverty is the best way to reduce crime. Maybe it would be cheaper to put criminals in the pillory.
The case for relieving poverty lies elsewhere.

Guns, Beyonce, and the Prisoner's dilemma

The Dibble are getting into the music business:

A music video director who has worked with stars such as Beyonce and Kellis has created an anti-gun video for the Metropolitan Police....Commander Cressida Dick, who is in overall charge of Operation Trident, said: "With so much of popular culture glamorising guns we wanted to subvert the messages associated with it."

This might not be as addle-brained as you might imagine. It could be a solution to a prisoner's dilemma problem.
Start from the premise that some young black men carry guns not because they are vicious criminals but because other young men do. Each individual figures: "I'd rather no-one carried guns. But seeing as others do so, I better had, else I'll feel unsafe and look uncool."
With plenty of men reasoning like this, we end up in a sub-optimal Nash equilibrium. Each individual avoids the worst possible outcome - in which everyone but them carries a gun. But in doing so, they all fail to get a better outcome, in which no-one carries a gun.
How do we get out of this equilibrium? One possibility is to change the focal point. Right now, the focal point for many young men is that gun ownership is the done thing.  But if there's enough propaganda to suggest gun ownership is unconventional and uncool, the focal point might shift towards the superior equilibrium of no gun ownership.
The Met's use of a music video is not therefore necessarily stupid.
The question is: is this strategy more likely to succeed than one which raises the cost of gun ownership, for example through higher probabilities of being caught and tougher sentences?

Crime: two recent papers

Barbara Reid, the mother of the murdered Manchester schoolboy Jessie James, wants to reintroduce conscription to reduce teenage crime. Is she right?
Coincidentally, two new economics papers bear upon this very question. This one suggests conscription doesn't work. The authors conclude that:

Participation in military service increased the likelihood of developing a criminal record in adulthood.

So what would stop gun murders? This new paper (pdf) by Naci Mocan has a depressing finding - the death penalty. People respond to incentives.
Of course, there's much to be said against the death penalty. But perhaps our opposition to it carries a cost, of murders that would not otherwise happen.