Before I gave evidence to the Justice Committee the other day, I did a bit of swotting up on what the MPs have said previously on prisons. I noted that Philip Davies MP argues that prison sentences are not long enough (1). Foreseeing a possibly line that doubling the prison population since the mid-90s was the cause of the dramatic halving in crime over the same period, and therefore worth every penny, I sought for a proper study of the connection between the two. To my surprise, I could find none (2) and **. (Come on, criminologists!)
Here, therefore, is my home-made analaysis. Let me know if you can do better!
Trends in crime and the prison population, E and W, since 1991
Despite the pleasing symmetry of this graph, it seems extremely improbable that the doubling of prison numbers has made more than a marginal contribution to the halving of crime, for the following reasons:
1, Big reductions in crime rates have occurred across the world, including countries where the prison population did not increase (3). Clearly, it is a global phenomenon, not the consequence of criminal justice policies in particular countries, so much so that universal explanations, such as the ending of lead in petrol, have been proposed.
2, Taking the argument of incapacitation (if they're locked up, they can't be committing crimes, at last not against the public), to get from a 40,000 increase in prison numbers to a reduction of 12, 000, 000 crimes a year since 1995 (4), you have to make assumptions about the rate of potential ooffending by those locked up, about their contribution to the overall rate of offending by all offenders, and the likelihood of others 'filling the gap' while they are inside which look completely implausible. [Seealso Addenda].
3. Taking the argument of deterrence (offenders are so frightened of jail that they give up crime), we know that offenders are very poor at estimating and acting on assessment of long term risk and we also know that what they do pay some attention to is the risk of being caught. But the risk of being caught has barely changed until the mid 2000s, by which time most of the increase in prison numbers had already happened (5) (and there is so much doubt about police crime statistics that they are no longer pubished by ONS (6)). The idea that an offender will reflect and think: well, my chances of being caught haven't changed - but if I am caught, I'm likely to do an extra x months inside, so I won't take the risk, is implausible.
4. International studies that map changes in crime levels again change in prison numbers, especially in the many jurisdictions of the States, have failed to find any conclusive link, but generally indicate that the effect of increased incarceration on crime is likely to be marginal (7). [But see Comment posted below]
5. Activities such as binge drinking by young people. which are not in themselves criminal, so not affected by changes in sentencing, but which are linked to risk of offending, have also reduced over the same period, suggesting again that this is part of a general change in social behaviour (8)
6. The reverse case: the rise in crime since WW2 up to the mid-90s was not accompanied by a falling prison population.
Both the remorseless rise in crime up to 1995, and the remorseless fall since, seem very unlikely to have much to do with the activities or policies of the criminal justice system. Society changes: we speculated what drove crime up: we now speculate what drove it down: but we don't really know.
Or, as a Tory of another age put it:
“How small, of all that human hearts endure,
That part which laws or kings can cause or cure.”
Which leaves open the question of why we have twice as many people in jail as 25 years ago, at vast and, it now turns out, insupportable cost.
** I have since come across a study which sought to model the impact on crime rates of increasing the prison population in the UK, R. Tarling 'Analysing offending: data, models and interpretation' (1993). Tarling's elaborate modelling estimated that based on incapacitation alone, a 25% increase in the prison population would achieve a 1% fall in crime, though noting that better results would be achieved in the criminal justice system managed to target high frequency offender (which it never has). Tarling's figures were an underestimate because he did not consider the impact through deterrence. The conclusion is that the estimated impact of increased imprisonment on crime rates must vary considerably depending on the assumptions used, but is likely to be pretty small.
(1) There is of course a reasonable case for increasing prison numbers - if we somehow succeeded in apprehending and convicting a larger proportion of serious offenders. But that seems vanishingly unlikely to happen, and certainly isn't the cause of prison numbers rising so much to date.
(2) The Carter report, 'Managing offenders, reducing crime' (2003) suggested that about one sixth of the fall in crime could be attributed to increased use of imprisonment - but gave no source or evidence for this assertion, and the Home Office have been unable to shed any light on it. [see Addenda]
(3) 'Exploring the international decline in crime rates', A. Tseloni et al., European Jo. of Criminology, vol 7, issue 5, 2010
(4) 'Crime in England and Wales: year ending June 2016' ONS, 2016.
(5) 'Crimes detected in England and Wales 2011-12', P. Taylor and S. Bond, Home Office Statistical Bulletin 08/12, 2012.
(7) 'The growth of incarceration in the Unites States: exploring causes and consequences', J. Travis et al., Nation Academies Press, 2016.
(8) 'Adult drinking habits in Great Britain: 2013' ONS 2015
I was formerly Finance Director of the Prison Service and then Director of the National Offender Management Service responsible for competition. I also worked in the NHS and an IT company. I later worked for two outsourcing companies.
Click below to receive regular updates