FLASHPOINTS #11: Why is knife crime rising - and is stop-and-search the solution?

A conversation with Gavin Hales

Branded knives seized by police but typical of knives which are widely and legitimately on sale in the UK.

In the FLASHPOINTS series, I interview experts about controversial questions, like free speech, the gender gap in mental health, the ethics of immigration, and AI risk.

In 2021/22 the number of people killed with a knife in England and Wales was the highest for 76 years. Around where I live, in north London, it’s no longer much of a surprise to learn that somebody has been killed or seriously injured in a knife attack. Just last week, for instance, a sixteen-year-old boy was stabbed to death as he left school. What’s behind this awful trend, and what can be done about it? One remedy may be encouraging the police to do more stop and search, but the policy is very controversial.

There’s a lot of noise around this issue in the media and a shortage of even-handed information and analysis, which makes it a perfect topic for FLASHPOINTS. I talked to Gavin Hales, a Senior Research Fellow at the London Metropolitan University and Visiting Fellow at the London School of Economics. Gavin’s Twitter feed is an essential source of information on crime and policing. We talked about the interaction between policing and knife crime; how the drugs economy has changed in recent years; the retail market for knives, and whether stop and search leads to racist discrimination.

This is an edited transcript of a live conversation.

Hi Gavin. Let’s define our terms a little. What do we mean by knife crime?

Essentially, crimes in which a knife was used to facilitate the offences, although much of the focus in recent years has been on knife crime in the context of what is often described as “serious youth violence” - where the victim is under 20.

That kind of knife crime has been rising since around 2016, is that right?

Yes, although it might be helpful to take a slightly longer-term view and to look at it in the context of stop and search. In around 2007 and 2008, there was a surge in the number of teenagers being killed in London, with knives. That triggered a large increase in use of stop and search by the Metropolitan Police under the name Operation Blunt. But then the 2011 riots in London happened. Afterwards, part of the accepted explanation for that event became that it was a reaction to the way that communities had been policed, and especially to the use of stop and search. As a consequence, the Met implemented a large reduction in stop and search - volumes of stop and search were reduced to about a third of the level they had been prior to the riots.

For a couple of years after the riots, various forms of violent crime fell - including the kind of knife crime we're interested in here. So there was this kind of simultaneous fall in the forms of violence that might be targeted by stop and search - and in stop and search itself. But there was an inflection point around 2013 and 2014: knife crime bottomed out and started increasing. Then in 2016, there was a sudden spike in knife crime and gun crime, and in 2017 and 2018 there were further large increases in knife crime, and knife homicide in particular. In the summer of 2018 the Met started to re-increase the amount of Stop and Search they were doing. The pendulum had swung away from “We need to reform/reduce stop and search” and back to “We need to do something about all these young people being killed on the streets.” And after 2018, following that increase, levels of violent crime fell back a bit.

That takes us to the start of the pandemic. Everybody was staying at home, and many forms of violence - and other crime types - fell dramatically as a result of those restrictions. Latterly, we’ve started to see crimes, including knife crime, starting to creep back up again. So that's the longer narrative which I think is important to understand today.

Why did knife crime fall after the riots?

At the time of the riots I was embedded in the police, doing some research on crime and policing in East London, using a desk in an intelligence unit. A lot of the individuals in that part of London who were otherwise involved in the kinds of violent crime we are discussing here were involved in the riots. More particularly, they were involved in the likes of violent robberies during the riots rather than fighting with the police - it was an opportunity to target jewellers and places like that.

So a lot of them were already well-known to the police and quite a number of them were then arrested in the aftermath of the riots. My hypothesis is that, across London, that had a kind of suppressive effect on levels of violence, both because a number of people who were otherwise involved with violence had been sent to prison and given quite lengthy sentences, and also because a number of others were susceptible to being arrested at some point, because they had been involved in some way, even if they weren't actually caught.

So there was both political pressure for a rethink about how stop and search was used and simultaneously a fall in violence, which created political space for the Met to reduce stop and search without difficulty. The Met had also been coming under pressure from the Equalities and Human Rights Commission (EHRC) about their use of Section 60 powers – stops and searches where you don't require “reasonable suspicion” about a particular individual.

So there was the riots, there was the EHRC pressure - and violence was falling. Nationally, Theresa May, as Home Secretary, introduced reforms to the use of stop and search in 2014. At which point you see stop and search reducing right across the country.

You don’t then see a rise in knife violence until 2016 – was that a belated reaction to the decline in stop and search?

That's a possibility, as is a gradual wearing off of the post-riots effect I’ve hypothesised. We know there's quite good evidence that one of the strongest deterrents to crime is people's perception about how likely it is they're going to get caught. That has a much stronger deterrent effect than, for example, sentencing policy. So one hypothesis about the relationship between stop and search and knife crime is precisely that there was a kind of de-policing effect. There was less stop and search - and this was happening at the same time as austerity was being implemented. The number of police officers was falling, police stations were being sold off and closed. There was less of a police presence on the streets. (We don't have conclusive answers about these things, just hypotheses.) 

One of the things that I think is very interesting is that in the crime data for London we started to see an increase in people being caught carrying knives and other pointed or bladed weapons before the Met started to restore stop and search - so around 2016 into 2017. It looks like, in the background, more people were carrying knives and other weapons on the streets, as stop and search continued to bump along at a very low level. That seems to have presaged the sharp increase in violence that started initially in 2016.

People were carrying more knives because the risk of them being stopped and searched was lower?

We don't know conclusively, of course, but that may well be part of the picture. There were also lots of other things going on. I put together what I thought were the main hypotheses in a presentation a few years ago. It is likely, for example, that there were other austerity-related effects such as cuts to youth service and youth work; what are typically described as diversionary activities, positive activities for young people. There were also changes taking place in the criminal economy and in particular around illegal drugs markets, which links to violence.  Starting from the early part of the 2010s there was a sharp increase in the purity of cocaine and crack cocaine in the UK. That reflected changes in the global market for cocaine which meant production and availability increased a lot. There was an increase in the number of people presenting to drug treatment services for cocaine and heroin related addiction problems.

So it’s probably not a coincidence that around the mid-2010s we then saw the emergence of, or certainly a growing recognition of, what's described as “county lines”, centred on crack and heroin dealing. There was an expansion of illegal drug dealing coinciding with this increase in supply of cocaine into the UK and that seems likely to have been a factor in the increase in violence around 2017 and 2018.

You got an expansion of the number of people participating in illegal drugs markets, an expansion of those markets out of their traditional bases into other areas, and an expansion in the number of young people who were being recruited to occupy the most risky positions in those markets. Risky in two senses - in terms of their exposure to law enforcement, because they're carrying the drugs and the cash and conducting the deals, but also risky in terms of their exposure to victimization by other criminals.

Can you say a bit more about the “county lines” model and how it works?

In essence, the county lines model involves groups from urban hubs expanding their drug dealing activities into county locations, typically selling crack and heroin to dependent users. “Lines” refers to telephone lines. If you're a crack and heroin user in a market town you might know about the “John line” – a branded phone line you can use to order your drugs for the day.

It’s worth saying that people commuting out of London or Liverpool or Birmingham to conduct drug deals isn't a particularly new thing. What's newer about county lines is the placing of teenagers in those remote markets. It's a kind of hub and spoke model where the person controlling everything is still based in the metropolitan hub, but they've got somebody doing their dirty work for them in the county market (although we also see the same exploitative model within metropolitan areas as well). And very often the individuals doing those deals are teenage boys. One of the ways this began to be identified was when boys who were missing from care settings in London started to be arrested in other parts of the country. So these are strategies used to expand into new markets, and in the process to push risk onto more vulnerable people, who are are then exploited.

What are the other possible causes of the rise in knife crime?

Social media has grown hugely in significance as a place where reputations can be built and challenged in. public. Conflicts now play out in front of an audience of hundreds or thousands of people. Somebody can accuse someone of something and the person on the receiving end of that feels a greater need to respond because they’ve been challenged in front of this huge audience. What might have once taken place in a local park or on a high street can now happen in front of the equivalent of Wembley Stadium. There's also the immediacy of social media which tends to accelerate the dynamics of conflict.

Why do these online conflicts end up in physical violence?

Physical violence is a key way that some people choose to reassert their status and get revenge for perceived slights. There's a lot of tit-for-tat going on. If you think about it in the context of gang or group dynamics where people have kind of a group identity as well as their individual identity, the reputational status of a group may be called into question as well as the individual’s reputation status. It becomes about collective responsibility and revenge. Somebody may be targeted not because of something they did, but because something one of their peers from their identified group did. The range of people drawn into conflict gets widened. We often see, with the most extreme forms of violence like murder, that it was part of a succession of tit-for-tat stabbings or even shootings.

So the violence becomes slightly detached from the actual business of drugs, right? It is like a culture war for real. People are killing each other not because they've been betrayed on a deal but because they've been insulted on social media.

Yes, absolutely, and the connections to drugs may be blurred or not even a factor. That said, if you're a drug dealer, you don't have recourse to the law - if people are targeting you, you can't call the police and say “someone's just stolen £10,000 worth of drugs from me”, at least not without admitting your own offending. So you are vulnerable to criminal predation and there's a greater pressure to maintain status, as somebody not to be messed with, as a defensive strategy. Violence, or the threat of violence, is one of the ways in which these illicit economies are regulated. If you are vulnerable to predation by other criminals, your reputation is particularly important.

You're therefore likely to be more sensitive to perceived slights and challenges to that reputation from other people because if you show signs of weakness in a social context, for example, other criminals may be more likely to target you. So what you then see is a kind of overspill of conflict to or from things like disputes over girlfriends, things that have been said about a family member or artistic settings, music rivalries. In the context of somebody whose reputation is incredibly important to them to stay safe they may feel the need to respond very robustly to those other challenges.

So these different motivations can become kind of intertwined and blurred. It's not uncommon, in police investigations and criminal trials involving youth violence, where somebody's been killed, to never really understand the exact motive for what took place. They'll be able to establish who did what, to whom, and when, using things like CCTV and phone data. But not always why. There's enough evidence to prove the offence, but the exact motive may not be clear because of these complex interwoven motivations that sit behind it all - and the fact that witnesses and others involved can be very reluctant to provide information to the police.

To some extent I suspect this would always have been true, right? But social media raises the stakes for these reputation battles and generates more of them. So there are more opportunities to fall into these deadly reputational conflicts.

Yes, it amplifies everything, and it also accelerates. There's less time for people to think about how they might react to challenges for example.

They feel an intense pressure to do something about the challenge - and that thing might be going around to that person's house right now, and attacking them…

Yes. It might be getting a bunch of people together and getting in a car and driving to a rival area and looking for somebody from the other group to target.

The rest of the conversation is available to paid subscribers. After the jump: the phenomenon of ‘weapons contagion’; the glamorisation of knives and what we might learn from cigarette packaging - and a truly illuminating conversation about whether stop and search is, in practice, unfair to black people.