Thursday 18 July 2013

Those crime stat's

There's a lengthy and interesting article in today's FT about the alarming rise in crime figures between the 1960s and the 1990s and the subsequent dramatic fall. Just to whet your appetite...

... criminologists have struggled to explain the decrease, leading to the flourishing of conflicting theories. Some say that advances in policing have tamed criminal behaviour, while others suggest a new generation of youngsters, who drink less, take fewer drugs and apply to university in droves, is increasingly uninterested in criminality.

Alternative hypotheses link the fall to the legalisation of abortion – a controversial idea that fewer unwanted children has led to a drop in crime – or to the declining use of lead in paint and petrol, which are thought to have an adverse effect on the developing brain...

But that argument does not explain why violent crime has decreased significantly over the past 30 years. Steven Pinker, the Harvard psychologist, suggests that the western world's "three-decade binge of crime" starting in the 1960s was itself an aberration in a longer-term trend of society slowly "civilising" over thousands of years.


The online version doesn't seem to have the charts which are in the printed edition. One of them says that the number of crimes committed/reported (in the UK) rose from about ten million to eighteen million per year and then fell back to eight million and that the UK prison population was constant at about 40,000 until the mid-1990s and has doubled to 80,000 since then.

We know that most crimes are committed by a very small number of people, so increasing the prison population is one perfectly plausible explanation for the fall in crime (providing you lock up actual criminals and not people whose only offence is ranting on Twitter or refusing to pay their Council Tax bill), but is it really possible that those extra 40,000 inmates would have committed an average of 250 crimes a year each?

Does anybody know how to calculate or estimate the deterrent effect? Perhaps those extra 40,000 inmates would have committed 100 crimes a year each (just about plausible) and a further 100,000 people were "deterred" from committing another 60 crimes a year each?

10 comments:

PJH said...
This comment has been removed by the author.
Furor Teutonicus said...

XX Does anybody know how to calculate or estimate the deterrent effect? XX

Aye. The larger the calibre, the greater the deterrent.

Bob E said...

Please excuse what might be considered a "knee jerk" response, but the questions that spring to mind are (a) is the data consistent across the piece - how accurately have "crimes" been reported and recorded across the decades covered, and (b) are offences consistently counted on a "like for like" basis or might some previously seen as serious offences been downgraded, and some previously seen as less serious offences upgraded (and these changes perhaps then also being reversed)at any point. As a way of graphically illustrating the questions I would point people at Figure 1, and in particular the "key" at top right here

http://www.ons.gov.uk/ons/rel/crime-stats/crime-statistics/period-ending-sept-2012/stb-crime-in-england-and-wales--year-ending-sept-2012.html

Furor Teutonicus said...

XX (b) are offences consistently counted on a "like for like" basis or might some previously seen as serious offences been downgraded, and some previously seen as less serious offences upgraded XX

Deffinately there have been a few times the methodology has been changed in that time.

Things "de-crimed", and tons of new stuff, particularly under Blair and incapability Brown, that were not even considered offences at ALL, priorly. (Can not think of one off hand, theer were so many).

Mark Wadsworth said...

BobE, FT, yes of course there are different ways of measuring and counting and classifying "crime", but this is not some minor fudge of UK statistics, crime statistics are ALWAYS fudged but probably consistently fudged within a margin of error.

But it has been observed almost globally that there was a huge increase in crime from 1960s to 1990s (i.e. it doubled) and since then it has halved again, i.e. we are now back to 1950s crime levels.

And the question is, why is this so? There are lots of explanations, some valid, some not and nobody really knows.

Bob E said...

MW yes, for reasons I won't bore everyone else with it was of course doubly naughty of me to throw up the "but maybe it's just down to 'statistics'" line ... and to quote from this article from earlier this year - a lot of which the FT article seems to repeat -

http://www.guardian.co.uk/commentisfree/2013/apr/28/crime-is-down-what-a-mystery

"Here is some good news for nearly everyone. Crime is falling. Murder is down. Violent crime generally is down. Property crime is down. In fact, almost every category of criminality that you can think of is declining. Here is the even better news. This is not a blip. The downward trend is now very well established and can be traced back over many years. And here is the funniest bit of this news. No one is really sure why. Those who think of themselves as experts on the causes of crime confess to being nonplussed. The majority of them predicted that a prolonged economic squeeze could only lead to more crime. They are scratching their heads trying to figure out why the opposite is happening."

Tim Almond said...

Crime's pretty complicated because there are a number of incentives.

Consider theft. The incentive is that you get something for free. The disincentives? You might go to jail. The victim might beat the crap out of you. If you have a job, you might lose it.

If you let people have a lot more chances in court, they'll do more crime.

If you have people who are living off the state, they won't care if a criminal record affects their career prospects.

If you put more police on the beat or dedicated to solving crime, or give them crime fighting technology, you'll catch more villains, which will deter crime. e.g. giving the police automatic number plate recognition system has hugely reduced joy-riding. If you drive past a police car with it, a stolen car is highlighted.

If people have more goods, especially easy to steal goods, you'll get more theft. As iPods became popular, muggings went up.

If you lower the price of goods that people steal, they'll steal less of them, as it's not worth it. When DVD players fell to £30 each, burglary fell because you couldn't fence them.

If people invent a way to stop something being stolen, or to kill it after it's been stolen, people won't steal it. So, since they can kill cellphones and SIM cards, cellphone crime has fallen.

And there's even things like the extended licensing hours. Once you stopped the taxi rush at 11pm, nighttime crime fell, because you didn't have people fighting over cabs or birds, because the distribution of cab use spread.

Overall, I think policing is one of the smallest factors in crime. They stop serious crimes, but for most crime, they just can't be around, and the bulk of the millions of crimes are things like theft, not rape.

Mark Wadsworth said...

BobE, exactly, they are scratching their heads.

TS, excellent summary. But as it happens in that Freakonomics book, they say that simply increasing police presence DOES reduce crime. Not hugely, but measurably.

"If you have people who are living off the state, they won't care if a criminal record affects their career prospects."

Ah yes, there must be some correlation between "having criminal record" and "not being able to get a job" and "being unemployed" and "resorting to crime" which is why in my Citizen's Income world, fines will be deducted from your CI.

Not so much as to make people desperate and reckless, but just enough to discourage it.

"Mum, how come Uncle Vern only eats baked beans and never buys me presents?"

"That't because he only gets £40 a week CI instead of £80 because he tried to rob the off-licence last year."

"Right mum, in that case I won't try to rob the off-licence."


Furor Teutonicus said...

XX The majority of them predicted that a prolonged economic squeeze could only lead to more crime. They are scratching their heads trying to figure out why the opposite is happening." XX

Easy. All the good stuff has already been nicked, so saturating the "fence" market, and no one has any bloody mobey to replace it anyway.

Furor Teutonicus said...

Or any money, even.