Fact Check: has violent crime gone up?

In the last 12 months, reports of violent crime have gone up very substantially. It is a very significant increase in violent crime reported in the last 12 months.

Yvette Cooper, Labour shadow home secretary, speaking on the Daily Politics Home Affairs debate.

This is pure scaremongering. The fact is that crime has gone down, it’s now at record low levels … both under police recorded crime and the independent crime survey which started in 1981.

Norman Baker, Liberal Democrat, former minister for crime prevention in the same debate.

Labour also came under fire in this Daily Politics debate for a claim in its manifesto that “violent crimes have gone up”. To check whether violent crime is going up or down, it’s important to note that we have two means of measuring crime.

First, police-recorded crime statistics: these include incidents that come to the attention of the police and are recorded by them as “crimes”. Second, the Crime Survey for England and Wales (CSEW): this is a very large, rolling annual survey of sizeable sample of the population. Both sources have their problems, but it is generally the crime survey that is considered the more reliable indicator of the two.

The data from the crime survey is very clear. Violent crime has been in almost continual decline since the mid-1990s and current estimates suggest it is at its lowest level since the survey was instituted at the beginning of the 1980s.

Trends in violence 1981 – 2014Crime Survey England & Wales

As with all sources of statistics, the CSEW comes with various “health warnings”. It is a household survey and therefore misses a lot of people who live in institutional settings such as prisons or student halls of residence. Until very recently it excluded people below the age of 16. It relies on victims’ reports and therefore cannot deal with so-called “victimless crimes” but, with the exception of homicide, this doesn’t affect violence.

In reality, the care and consistency with which the CSEW is undertaken undoubtedly makes it a reasonable medium to long-term indicator of crime trends. Any remaining doubts about what is happening to violent crime can be dispelled by turning to two other sources of data, both of which lend credence to the picture painted by the crime survey.

National Health Service data on assault admissions to hospitals in England suggested a 5% drop over the past year. The impressive survey work in hospital emergency departments and walk-in centres, undertaken by the Cardiff University Violence Research Group, found a 10% drop between 2013 and 2014 in serious violence-related attendances. This research from A&E also supports the longer-term trend indicated by the CSEW, suggesting that with the exception of 2008, there had been decreases in serious violence every year since 2001.

At this point we should return to Yvette Cooper’s statement for it is important to note that what she actually claimed was that reports of violent crime have gone up very substantially. To check this we need to look at the latest police-recorded crime statistics. Additionally taking Yvette Cooper’s much shorter time period – the last 12 months only – they do support the idea that violent crime generally, and sexual offences in particular, have increased. Selected victim-based recorded crime categories – percentage change 2013-14ONS

Why might this be the case and how might we explain the difference between the two data sources? The answer, in short, is that these statistics tell us more about recording trends, especially in the short-term, than they do about crime trends.

It’s likely that they are illustrating increased compliance with the rules that govern how the police record crime – so they probably reflect improved recording. These improvements come in the wake of the decision in early 2014 by the UK Statistics Authority to cease to designate recorded crime statistics as “national statistics” given their unreliability.

However, it seems likely that the increased emphasis that has been placed on improving police activity in the area of domestic violence has also contributed to increased reporting and recording in that area in particular.

Verdict

Although the data is reasonably clear, the verdict is complicated. There is some evidence, albeit limited, of increased reports of violent crime, certainly domestic violence, over the past year. However, there is little evidence to support the claim more generally of reports of violent crime going up very substantially. Much more likely is that they are being recorded more accurately by the police.

The real problem, however, lies in using police-recorded crime statistics, especially over very short time periods, to make claims about crime trends. In fact, more reliable measures like the CSEW show violent crime to have fallen over a very long time period, and by a substantial amount – just as Norman Baker indicated in his reply to Yvette Cooper, though he then ruined it by overstepping the mark with his claim that recorded crime was also at an all time low (it is not).

So, Yvette Cooper’s claim is technically correct, though she is on very thin ice. What she could not easily defend is the claim in the Labour Party election manifesto which states quite baldly that: “Violent crimes have gone up”.

Review

This is a balanced assessment of Yvette Cooper’s claim. It is generous in describing this as technically correct.

It is clear that once the Audit Commission stopped auditing police recording practice in 2007 the overall recording rate fell. What caused the police to improve their recording rate in 2013-14? The UK Statistics Authority removal of the “National Statistics” designation was clearly one factor. But there was also the highly critical inquiry of the House of Commons Public Administration Select Committee, and a series of critical reports by HM Inspectorate of Constabulary (HMIC).

So long as HMIC maintains pressure on the police to record crime more fully, we can expect to see a divergence between the crime survey trend and police statistics. – Mike Hough

Tim Newburn has received funding from a number of government departments, the Economic and Social Research Council, the Joseph Rowntree Foundation, Nuffield Foundation and Open Society Foundations. He is a trustee of the Howard League for Penal Reform. The views expressed in this article are his own.

Mike Hough was involved in designing the British Crime Survey at the Home Office. He has worked for and advised the Home Office and other organisations including HM Inspectorate of Constabulary, the College of Policing, the Prison Reform Trust and the Howard League for Penal Reform. The views expressed in this article are his own.