Lies, Damned Lies and Charles Murray’s arithmetic

In 1997 the Sunday Times published an influential series of articles by Charles Murray’s entitled ‘Prison Works’. Murray claimed an aversion to prison and its under-use was responsible for a dramatic increase of crime during the second half of the twentieth century. The policy implication was simple. Any Government wanting to significantly reduce crime must build many more prisons. To do any less was to fail to protect the public. Murray claimed his arguments in favour of mass incarceration were “a matter of arithmetic, not ideology.” To prove this he cited selected crime and prison population statistics from the United States and England and Wales. To investigate his claims I looked at the statistics for England and Wales in the early twentieth century, four European countries statistics for the period 1990 to 2000 and United States statistics for the whole twentieth century.

According to Murray increases in crime in England since 1955 are explained by a change “in the elite wisdom about how to deal with crime” that concluded, “prison was retrogressive”. In fact in the mid 1950’s these ideas were increasingly being rejected in favour of a more hard-line approach. The 1958 White Paper “Penal Practice in a Changing Society” and Home Secretary, R. A. Butler’s speeches demonstrate both an increasingly punitive approach and a rejection of the reformative ideas that had dominated the first half of the twentieth century. The elite wisdom Murray claims developed in the mid 1950’s had in fact first emerged at the end of the nineteenth century. It was put into practice early in the twentieth century with the development of Probation, reformative Borstals and the requirement of magistrates to give offenders time to pay fines. All these policies were intended to reduce imprisonment and they did dramatically. English and Welsh prison receptions declined from over 200,000 in 1908 to less than 28,000 by 1918 and were maintained at this level until the 1940s. Reported crime remained relatively stable in this period. The dramatic reduction in the use of imprisonment had no statistically significant impact on crime.

In Europe I looked at Denmark, France, Greece and the Netherlands using reported crime and prison statistics for the years 1990 and 2000. Two of the countries significantly increased their prison populations, Greece by 57% and the Netherlands by a whopping 101%. Demark and France had stable prison populations with increases of 1% and 3% respectively. If Charles Murray’s claims were correct Greece and the Netherlands would have experience a decrease in crime. However both countries recorded 12% increases. In Demark however crime fell by 4% and in France it increased by 8%.

Murray also looks at the relationship between crime rates and imprisonment in America and concludes that the apparent stabilisation in the crime rate in the 1990’s was a direct result of increased imprisonment since 1980. Reviewing the statistics for twentieth century America it was clear that two government policies could explain variations in crime and imprisonment rates far more effective than Murray. The first was Prohibition. In 1907 when Georgia and Oklahoma became the first states to prohibit alcohol the American murder rate was 1 person per 100,000 and it imprisoned 69 persons per 100,000 population. By 1933 when the error of prohibition was conceded and the 18th Amendment repealed the murder rate had increased by over 1000% and America’s prison population had doubled. By 1943, ten years after the end of prohibition the murder rate had halved and America’s prison population had reduced by a third.

The war on drugs had a similar impact. By 1970 homicide had doubled and was again at 1933 levels. The impact on imprisonment was far more dramatic and sustained. By 1980 it had risen to 138 per 100,000 and by 1990 it was 297 per 100,000. This rise has continued to date with the USA now locking up over 2 million people, a rate of over 700 per 100,000. As the end of prohibition showed ending the war on drugs would be the most effective crime reduction strategy. Combined with a general amnesty on incarcerated drug offenders the USA (and indeed Britain and the rest of the world) could benefit simultaneously from reductions in crime and people incarcerated.

A quick glance at Murray’s other writings including “The Emerging British Underclass” (1990) and “The Bell Curve” (1994) shows his preoccupation with the poor and powerless and the threat they pose. In 2002 the Social Exclusion Unit reviewed the characteristics of the prison population. They found that prisoners are 14 times more likely to have been in care as a child, 20 times more likely to have attended a special school, 13 times more likely to be unemployed, 50 times more likely to be suffering from three or more mental disorders, 12 times more likely to have a personality disorder, and 4,700 times more likely to be sleeping rough. Politicians may claim, as former Home Secretary David Blunkett did, that their agenda is about “tougher sentences for violent, dangerous sex crimes” but the reality during his period in office was imprisonment targeted at the weak, vulnerable and powerless. Murray’s mathematical fallacy, followed by Governments on both sides of the Atlantic, has had a dramatic impact. In both the UK and US the homeless, women, children, black people and the mentally ill are all being disproportionately sucked into the ever-expanding penal estate.

Whilst we would all like simple and easy solutions to crime we need to realise that crime covers a multitude of unrelated phenomenon. Imprisoning people has high social costs and is as likely to increase crime, as it is to decrease it. The lessons from crime and prison statistics are clear and the exact opposite of Murray’s claim. The powerful moral and political arguments for prisons abolition are backed up by historical and statistical evidence that strongly suggests closing our prisons will at worse have no impact on crime and could potentially significantly reduce it.

JM Moore