Cooking Britain’s Criminal Statistics

A friend of mine, David Fraser, who is writing a book about our criminal justice system soon to be published, recently pointed out to me a startling, but by no means unusual, discrepancy in the statistics provided by the Orwellianly-renamed Ministry of Justice.


In the Criminal Justice Statistics Quarterly for March 2014, one table tells us that in the year up to that date, 8,300 people were cautioned by the police for acts of violence against the person. In the year up to March 2007, 46,200 people were cautioned for the same reason.


Of course, these figures tell us nothing about the actual change in the numbers of violent offences: but nevertheless the impression is created that, if the number of cautions issued declined by more than 80 per cent, a decline in acts of violence must have played some part.


In another table, we learn that of the 8,300 people cautioned, 5,400 were males and 2,800 were female.  (Here the slight discrepancy is explicable by the fact that the figures are given only to the nearest hundred.) The proportion of males to females, though, is a little puzzling, since males are more than ten times as violent as females. Perhaps this means that cautions are used (as one might hope) only in relatively minor cases, but this hope is somewhat undermined by the Ministry’s own statistics on another web page altogether, showing that at last count only 10 per cent of convicted persons were given non custodial sentences (deliberately misnamed offenders, as if every offender were caught and convicted, the proportion being more like 5 or 10 per cent) were first-timers. By contrast, 35 per cent of the convicted given out of court disposals had 15 or more convictions (how many more the site does not say, but leaves open the possibility, in fact nearly a certainty, that some are given cautions after having committed hundreds of offences).


The puzzle of the sex-ratio of persons cautioned for violent offences is solved when one notices that in one of the tables 47,000 violent offenders of unstated sex were cautioned in that year. So the real total, 55,300 is an increase since 2007, not a decrease, in the numbers thus dealt with, contrary to the impression created in one of the tables.


How can an offender’s sex not be known or recorded? How and why does the Ministry blithely omit 85 per cent of cases from its statistics in its tables? Is this error or fraud? 


First published in Salisbury Review.