Fraser Nelson recently posted a chart on X showing several types of crime trending downward. The chart was based on data from the Crime Survey for England and Wales (CSEW). In an article for The Critic, Adam Wren describes the posting of such charts as “the latest evolution in gaslighting”. And he suggests that “definitive conclusions about whether crime rates are rising, stable, or declining are challenging to draw from this or any other crime survey”. I don’t find his arguments very persuasive.
The CSEW is just a victimisation survey — an attempt to gauge levels of crime by asking victims about their experiences. It is roughly equivalent to the National Crime Victimization Survey in the US, which is widely considered a valuable resource and has been used by right-wing activists to dispel the myth that police are biased against black Americans.
There is therefore nothing suspect about using a survey to measure crime. The Office for National Statistics explicitly states that the CSEW is “better for tracking long-term trends than police recorded crime because it is unaffected by changes in reporting or recording practices”. For example, a major report in 2014 concluded that “the police are failing to record a large proportion of the crimes reported to them”, which prompted changes in official practices.
To support the charge of “gaslighting”, Wren makes several criticisms of the CSEW’s methodology. He notes that the sample is not disproportionately drawn from high-crime areas, it excludes certain types of dwelling, the questionnaire doesn’t ask about certain types of crime, and repeated victimisations of the same respondent are capped.
To begin with, the sample is designed to be representative of the population, which is a strength not a weakness, and the capping of repeated victimisations is done to reduce year-to-year variability. The crucial point, however, is that none of these features of the CSEW’s methodology can account for the dramatic fall in crime since the late 1990s. After all, the sampling scheme and questionnaire have remained roughly constant over time.
Besides, the CSEW isn’t the only piece of evidence that crime has fallen. The homicide rate is lower now than it was in the early 2000s, and the rate at which people are admitted to hospital with violent injuries has trended downwards. Evidence that crime has dropped since the late 1990s shouldn’t actually be surprising, since the same thing has happened in several Western countries. There is even a Wikipedia page about it.
Another reason we shouldn’t be surprised is that various changes have taken place in British society since the mid 1990s that would lead us to expect a drop in crime: population ageing, more obesity, less alcohol consumption, enhanced personal security, enhanced surveillance and forensics and, most importantly, a massive rise in incarceration.
To his credit, Wren mentions some of these changes in his article. However, he doesn’t draw what I regard as the obvious conclusion — that the trends in the CSEW data are likely real. Instead, he says our expectation that crime should have fallen is “hard to reconcile” with “imprisonment rates reaching historic highs”.
But this doesn’t make sense. High rates of imprisonment are not an indication that crime is high. They are the result of measures taken to reduce crime, or to prevent it from rising. Which is why the early release of inmates due to prison overcrowding portends an uptick of crime in the future.
Wren’s position is far from unique. Many on the right are extremely sceptical that levels of crime could be lower now than they were twenty years ago. But portable video cameras were almost non-existent back then, so few of us had a front-row seat to gang turf wars. Insofar as the crime drop followed a massive rise in incarceration, it should be seen as a victory for law-and-order politics.
Sceptics say claims of falling crime are at odds with how people feel. However, it is well known that survey respondents almost invariably say crime is rising, regardless of actual trends. For example, crime in the US plummeted in the late 90s and 2000s, yet the percentage of Americans saying “there is more crime than there was a year ago” went up rather than down.
The totality of evidence suggests that crime, especially violent crime, has fallen
As Steven Pinker has noted, this is most likely because of the nature of news. We hear about muggings, burglaries and stabbings — not about all the people who didn’t experience crime. In a country of tens of millions, there are always enough incidents to report on. As a matter of fact, sentiment analyses show that news has gotten more negative over time, which helps to explain why public perceptions are skewed.
Scrutiny of government figures is certainly warranted and, to be fair, Wren’s article is reasonably circumspect. However, the totality of evidence suggests that crime, especially violent crime, has fallen.
