Monday, 27 July 2009

Was Michael Howard Right?

by Richard Nabavi

Crime can easily become a vote-loser for any government, almost irrespective of what they actually do. The reaction by politicians and the media to the recent publication of Crime in England and Wales 2008/09 was typical. The government hailed the figures as showing that crime has fallen substantially; “Oh no it hasn’t”, said many commentators, “you’ve been fiddling the figures, it’s all about reporting rates and definitions”, often adducing anecdotal evidence to make their point. (And of course, there is nothing more irritating than being told how wonderful it is that crime is falling, when you have just had your car torched or your house burgled.) Few seem to have tried to see what lessons can be learned from the data.

This is a pity, because there are some very interesting lessons for policy-making in all those tables and graphs. But first let’s cut through the party-political point scoring, and look at the overall trends. And these are not in doubt. The truth is, both sides are partly right. Crime - certainly most categories of serious and violent crime – is higher than it was in 1997, and much higher than it was if you go back further. But it is substantially lower than it was three or four years ago.

To see this, forget about ‘total recorded offences’; this is a completely meaningless figure, in which the theft of a bicycle is given the same weight as a murder. Since there are more bicycle thefts than murders, a hypothetical fall in this total figure might simply reflect that bicycle thefts were down 10% even if murders had doubled. Would that be a fall in crime? The Home Office should be ashamed of themselves for publicising such a nonsensical figure.

In addition, to get a clear picture, it is important to concentrate on those figures which are most likely to be reliable. The statistics for rape, for example, are notoriously unreliable, so underlying trends are very hard to discern. And changes in recording methods make many other figures hard to interpret.

Instead, let’s concentrate on offences which are both serious and likely to be reliably recorded. In this graph, I have shown the yearly figures for three such categories: homicide and attempted murder, serious wounding, and robbery. In each case, I have taken the figures directly from Crime in England and Wales 2008/09 and re-based them to a starting index of 100 for 1997-8. (To get a clear picture of trends, I have adjusted the homicide/attempted murder figures to exclude the victims of the 7 July London bombings, and the 173 Harold Shipman murders recorded in 2002/3 but which happened many years earlier.)

(Click on chart to see higher resolution version)

As you can see, the figures for these categories – which are hard to fiddle or misrecord - show a substantial rise initially. But then something quite extraordinary happened. Having peaked in around 2002-2004 (depending on the category), these serious crimes have fallen quite dramatically - by 25% or more over four years. This is truly astonishing. If you look at many other serious crimes, in both the BCS survey and the recorded crime statistics, the overall picture is the same. The bottom line is: Crime kept rising until four or five years ago. (This may be why so many people, from their own experience, think that serious crime has risen a lot in recent years – they are right, looking at the longer term.) But then it started falling, surprisingly rapidly.

Understanding why this might be so is key to designing policy. We all know that the underlying long-term trends are heavily influenced by changes in our society: demographic changes, family breakdown, poor education, drug-taking, fractured social structures, and so on – the ‘broken society’. But those factors haven’t changed much in the last three to five years when crime fell. The drugs problems has, if anything, got worse. Family breakdown hasn’t reduced. Economic factors and unemployment were fairly stable over the period under review . Alcohol-related hospital admissions are up (although total consumption is down slightly). The broken society hasn’t been mended since 2003 – certainly not to a degree which might come close to explaining a fall of 25% or more in many serious crimes.

What else could possibly have changed to cause such a big effect?

Could it be that this chart , showing the total England and Wales prison population over the same period might hold some clues?

(Click on chart to see higher resolution version)

Is it a coincidence that the peak in the crime figures occurred just after a dip in the prison population in 2000, and then as the prison population increased the crime figures started falling? And that this is particularly true for ‘professional’ crimes such as robbery, but less so for homicides, which are often domestic?

Back in 1993, Michael Howard, then Home Secretary, was much criticised when he made his famous speech saying ‘Prison works’. Yet, curiously, it has been under a Labour government that we have seen that statement tested in earnest. And it appears he may have been right.

Of course, even if this is true, it doesn’t follow that all of those who are in prison should be there. Most people who are familiar with the justice system think that prison is quite inappropriate in many cases. But since it is believed that a large proportion of serious crime is carried out by a relatively small number of repeat offenders, maybe crime has dropped simply because more of those hardened criminals are in prison now than was the case in 2001.

In political terms, the issue of crime has been relatively subdued in recent years, perhaps because, after a shaky start, Labour has ended up presiding over a justice system which is more draconian than that of Michael Howard’s day. The recent debate has been more about civil liberties, and the failure to supply enough prison places, than about being ‘tough on crime’.

But if the pundits are right, and the economic crisis leads to an increase in crime, that could begin to change. Crime may well become an increasingly important issue for voters over the next few months. There are good electoral reasons why the political parties should consider very carefully both how they will address this issue in policy terms, and how they will present it to the public. The traditional divide of “Labour is soft on criminals, Tories want to lock ‘em up and throw away the key” seems to have broken down. It will be interesting to see whether that changes.


Council House Tory said...

Interesting. Personally I don't think crime can be readily discerned over any timescale less than a decade, so the fact you've got about 10 years there starts to give it weight. Starts.

If prisons weren't full of those in possession of drugs (or supplying smallish amounts) there wouldn't be overcrowding either.

Council House Tory said...

grrrrr. I meant crime trends in the first sentence.

Martin Day said...


An interesting and thought provoking piece Richard!

On the other side of the coin there are repeat offenders that seem to just keep commiting crime. Maybe only once they have been caught do they stop. Certainly some of the people who are in jail should not be there as quite often they have mental illness or personality disorders. It tends to be the real nasty ones who see it as an occupational hazzard or inconvienence. The numbers in Prison may have risen but the length of stay i would argue is failing. In recent years we have seen sentences deminished because there is no room in the prison!

The other thing you have to look at is how many people not in the full jails are on licience through tagging! There must be quite a few and if you add all that lot together then it is not hard to argue society is broken!

In terms of the two crimes commited against me in the last year - Last year my car was broken into in around July and a Sat Nav stolen! This year someone else blew it up! The difference in this years more serious crime being investigated where as last years break in was taken as just intellegience is quite astonishing!

It maybe a local crime hot spot in the area i live, which explains it - on recent stats it was the only place to see rising crime but do i trust government statistics!

MichaelK said...

Interesting article.

It would be interesting to know what type of offender is involved in the increase in the prison population.

Also, I'm also not sure about the argument for excluding 7/7. Shipman, OK. But the 7/7 killings happened during the period, they were unquestionably murders, and there's no reason to think similar crimes won't be attempted in the future.

Plato said...

Great article - will have to mull it over before having anything sensible to say.

bunnco said...

Back in 2005 the Tories under Michael Howard tried to distil their entire manifesto into six simple phrases.

1) More Police, 2) Cleaner Hospitals, 3) Lower Taxes, 4) School Discipline, 5) Controlled Immigration, and 6) Accountability

And they got stuffed by Labour just as badly as in 1997 and 2001. How could that be?

Five years on you get the feeling that the Tories could re-run the election on the same six phrases and hit Labour for six. It's just like that old joke about the Economics Exams at school. The questions stay the same every year... only the answers change.

I've felt for a while that the poor result in 2005 was all to do with timing rather than policy. Certainly, Howard identified the six big issues that would become mainstream during the life of the Parliament that was elected.

Was there anything Howard could have said in 2005 that would have resulted in a victory? Or at that time, was the public's mind simply just 'closed' to the Conservative message just as today it appears to be 'open' and listening?

It's a theme I raised with Michael when he visited Norwich to promote a local charity John Grooms Court during the height of the expenses scandal two months ago. [Incidentally, the luncheon marked one of Ian Gibson's final appearances as an MP in the city. Just a few days later, he became ensnared in the Telegraph's net and the rest is history.]

Howard confirmed to me that indeed timing was an issue and that probably it was just unwinnable given the swing that was required but he laid the foundations for the subsequent revival in Tory fortunes. It was too much to do in just one election.

There's no doubt that events have shown that Michael Howard /was/ right. But then, in politics, being right doesn't necessarily make you the winner. For that you need luck as well.

Ave it 09 said...

Ave it!!!!!!!!!

Barnesian said...

Does anyone know what the breakdown by type of offence is for the 80,000+ prison population?

How many are in jail for murder, robbery, shoplifting, possession of drugs etc. I've often looked for this data but failed to find it.

The trend by type of offence would be interesting. Then we would know what type of offences were driving up the prison population.

Anonymous said...

bunnco - interesting post.

At some point prison numbers must impact on crime rates - if every criminal were to be locked up and never allowed back into society crime would undoubtably fall, and i'll wager it would fall dramatically. The key question is at what point does the number in prison begin to actively affect crime rates? I'd like to see the above graphs in five or ten years time, but from the limited evidence here, maybe we've found it.

Richard Nabavi said...
This comment has been removed by the author.
Richard Nabavi said...

Thanks for the comments. To respond to some of the points raised:

Martin Day: I was thinking of you in my reference to "there is nothing more irritating than being told how wonderful it is that crime is falling, when you have just had your car torched.."! Obviously, the national figures don't necessarily reflect experiences in particular areas, but I think you are right to point to the problem of repeat offenders.

MichaelK: It's always hard to know what adjustments to make. If you include the 7/7 killings, you get a sudden peak in the figures, which doesn't really tell you much about underlying trends. I think it may be more sensible to look at terrorist attacks as a separate category. But however you look at it, the general trend is an increase in crime to around 2002-2004, and then a quite sharp decrease.

Barnesian: The figures you mention are published monthly by the Ministry of Justice, here:

bunnco: Yes, in politics, timing is all. Actually, I think Michael Howard was partially right; prison works in some cases, but not in others. I'd like to see a sharper distinction between confirmed criminals, and those who are drifting into crime through fecklessness.

stjohn said...

Very interesting article Richard. Thanks. Puts me in mind of some of the ideas in Freakonomics.

I think the idea of opening up PB.com2 is great and I will be lobbying for the PB "punters" to make particular use of this "quiet room".

John L said...

Off the top of my head, other factors might be an increase in police and quasi-police (eg pcso) numbers, or cctv. These might cause both a decrease in crime and an increase in imprisonment.

Then there are demographic changes, with much crime being committed by young men. Have their numbers gone down?

Ah, what about universities? Increased participation rates might be a sign of dumbing down but it keeps 'em off street corners.

no longer anonymous said...

The Home Office has its own crime figures. See this from several years ago:

Add up the figures on pages 12 and 13 and you get 62 million crimes - much higher than the BCS or police numbers.

Anonymous said...

Interesting read Another difference, as higlighted at Septicisle, ( policies.html) is the affect of medical advances making crimes 'less bad' then they would have been in the past.

AnneJGP said...

Richard, very interesting article.

"...then as the prison population increased the crime figures started falling? And that this is particularly true for ‘professional’ crimes such as robbery ..."

If true, should it be possible to predict the impact on crime rates from the likely release dates of the present groups of prisoners?

Dave B said...

Memory tells me the first episode of "The Death of Respect"

had a interesting snippet on crime, in that it's increasing in poor areas, and decreasing in wealthier areas.

MrJones said...

1) Very good point about the weighting - drives me nuts when they talk about "crime" going down without taking the weight into account - it's like they're saying to people whose son was stabbed that it's okay because there were two less bikes nicked on the same street so things are getting better.

When the average person says "crime" they mean the weighted version.

2) I think people's mental idea of "crime" in their area is also influenced by the way it's going. If you had two areas with an exactly level amount of (weighted)crime e.g 6/10 but one area it was increasing and the other decreasing then the people in the area where it's increasing will feel it's 7/10 and the other area people will say 5/10.

3) Additional to 2) the speed at which it is increasing or decreasing affects people's perception of crime also.

4) Over the years i've been jumped in the street hundreds of times. It never occurred to me to ever report it to the police. I might if I actually got stabbed maybe but even then not sure.

I've known loads of people who've been burgled or had something nicked but never reported it because they didn't have insurance. Sometimes because of poverty but also partly out of being thick.

Most of my London extended family apart from me have been mugged at least once. Pretty sure none of them ever reported it.

So crime figures in something like the British Crime Survey should *always* be *massively* higher than the police recorded crime figures. If they're not it's being fiddled.

5) Crime is mostly about young men. If the number of young men in a country rapidly increases then the amount of crime will rapidly increase. Because a big part of the increase in crime is the result of ZNL's immigration policy ZNL is trying to cover it up. It's much less of an issue than it should be partly because Pravda is helping to keep it covered up and partly because the people it's mostly happening to have no voice because their traditional voice is the one betraying them.

6) Personally I think crime is still going up. I think locking more people up has helped stem the tide a bit but the only way it's actually going down is if they're still fiddling the figures as well.

Anecdotal obviously.

On the other hand it is possible that it's going down some places and up in others so the total may be down on balance. However the areas I know about are not going anywhere except up towards US ghetto levels.

Also, stepping back a bit maybe the speed has changed. It was slowly getting worse every year up to 97, then very rapidly got worse - maybe it has slowed back down a bit. Hard to be sure once you've gone past a certain level of anger about it all.

7) Overall I'd say Howard was mostly right - crime and prison are a big subject and there are different kinds of criminal. However you have to have enough prison places to keep things down to a certain level otherwise all the kids growing up get led astray.

MichaelK said...

Responding to bunnco's question as to whether Michael Howard could have won in 2005, I think part of the problem was that he was personally very unpopular with non-aligned voters. He was the one reason I didn't seriously consider voting Conservative in that election.

Fwiw, I have a higher opinion of him these days as I can see that he did a good and selfless job for his party. I still wouldn't fancy him as PM though ;)

Putting this another way, I think the change in Conservative mood music in the early Cameron years was absolutely necessary.

Niklas said...

"But since it is believed that a large proportion of serious crime is carried out by a relatively small number of repeat offenders, maybe crime has dropped simply because more of those hardened criminals are in prison now than was the case in 2001."

A very interesting idea in an excellent post. But my concern is that with the awful rates of released prisoners reoffending (70% I believe) we are just storing up trouble for the future. Eventually almost all prisoners (only a small minority have life sentences) will be released, and if our prison system fails to reform them they may well commit more crimes.

Even worse, as Council House Tory points out: "If prisons weren't full of those in possession of drugs (or supplying smallish amounts) there wouldn't be overcrowding either." The trouble is that many of these petty criminals are being turned into more dangerous ones while in prison. I'd much rather fine them or give them community sentences (or in the case of cannibis legalise it) so that they don't become "professional" robbers or burglars on release.

So I'm not sure that the increase in the prison population is such a good thing.