From the front page of The Times...
Shock drop in violent crime
Serious violent crime has halved in a decade as Britons cut down on binge-drinking and move towards a less macho culture, research has found.
This echoes a very similar story from the same newspaper last year...
Violent crime has fallen to its lowest level for 30 years and firearms offences are down for the eighth year in a row.
Indeed, the same story could have been told every year for the best part of two decades. The Office for National Statistics notes that violent crime peaked in 1995 and has been falling ever since. A similar trend has been seen in many developed countries and no one knows quite why.
Many theories have been put forward to explain why murder and violent crime started falling dramatically in the early 1990s. Some attribute it to the the rise of CCTV. Others say it is the result of higher incarceration rates. The Freakanomics authors link it to legalised abortion. Earlier this week, the BBC revived the intriguing idea that it is due to the decline of leaded petrol (although I suspect this is a classic correlation without causation). The Spirit Level authors try to link it to inequality (despite the fact that there is not even a correlation in that instance). And so on.
Today, the claim is that it is due to the decline in alcohol consumption...
Jonathan Shepherd, lead author of the study by the University of Cardiff’s Violence and Society Research Group, said: “Binge drinking has become less frequent, and the proportion of youths who don’t drink alcohol at all has risen sharply. Also, after decades in which alcohol has become more affordable, since 2008 it has become less affordable. For people most prone to involvement in violence, those aged 18 to 30, falls in disposable income are probably an important factor.”
It's always nice to see the press report the decline in alcohol consumption, which now amounts to an 18 per cent fall since 2004. The problem with the boozing/violence theory is that the ONS's figures show that the biggest fall in violent crime preceded the fall in drinking by ten years. Since then, the decline has been less impressive...
Between the 1995 and the 2001/02 surveys, the number of violent crime incidents fell, from 4.2 million in 1995 to 2.7 million in 2001/02. Since then there has been a general trend where the CSEW [Crime Survey for England and Wales] has seen a period of modest annual decreases (though often not large enough to be statistically significant year on year).
The ONS shows the rise and fall of violent crime in this graph...
It's also refreshing to hear - contrary to the usual hand-wringing - that alcohol has been getting less affordable in recent years, not that it will stop talk of 'pocket money prices' or silence those who want even higher taxes on booze. Changes in affordability are primarily the result of changes in incomes (alcohol is still more expensive in real terms), and incomes have fallen back to the levels of 2004 during the economic crisis.
Alcohol affordability has therefore also returned to the level of circa 2004. In the great scheme of things, this change is not huge. After a dip during the recession, affordability seems to have ticked upwards again more recently. As temperance campaigners would happily point out in any other context, alcohol is more affordable than it was when violent crime was much more common.
None of this really tallies with the alcohol consumption data, which show that per capita consumption started rising in the mid-1990s just as violent crime was starting to fall and started falling while alcohol was still becoming more affordable.
It wouldn't be surprising if alcohol consumption and violent crime were linked; a lot of violent crime is committed by drunks. And it is certainly true that recent years have seen a decline in both, but the correlation does not stand up if we go a few years further back.
The bigger picture shows that the decline in violence - which has been seen in other countries which have very different drinking patterns - preceded the decline in alcohol consumption by many years and was initially accompanied by a significant increase in alcohol consumption. The decline in violence was also accompanied by an increase in the affordability of alcohol until quite recently.
So whilst we cannot rule out a 'recession effect' on heavy drinking amongst some groups which might have had an impact on the prevalence of violent crime, it does not seem to have been - indeed, cannot have been - the cause of the long-term decline.