Does stop and search work?
In announcing government plans to expand stop and search powers, the new Home Secretary Priti Patel claimed that “stop and search works.”
But what does that actually mean?
The discourse around stop and search’s effectiveness overwhelmingly focuses on the extent to which it deters crime.
But that’s an oversimplification of the role stop and search is meant to play. Police guidelines suggest that stop and search is primarily intended to be used as a tool to investigate crime, rather than to stop it in the first place.
Looking at how stop and search has been used over time and across the country, there is no single idea of how stop and search should be used. This makes measuring its “success” difficult.
Data suggests that stop and search is not particularly effective at reducing crime by deterring potential criminals.
The evidence is scanter on stop and search’s effectiveness as an investigatory power, but what evidence there is suggests a stronger case that stop and search works in this regard (for example police using the powers to search someone and subsequently making an arrest or giving out a warning or penalty notice).
That said, it’s important to consider any “success” of stop and search alongside possible negative effects, including the disproportionate targeting of BAME individuals and the mistrust in the police that can stem from this.
What is stop and search?
A police officer has powers to stop and search someone if they have ‘reasonable grounds’ to suspect they are carrying drugs, weapons, stolen property or something which could be used to commit a crime.
Stop and search allows the police to check their suspicions about an individual without having to arrest them.
The use of stop and search peaked in 2008/09, when over 1.5 million stop and searches were carried out. It has fallen dramatically since, over which time the likelihood of a stop and search leading to arrest has increased. You can read more about this here.
How is stop and search supposed to “work”?
Public discussion on the effectiveness of stop and search overwhelmingly focuses on the extent to which stop and search reduces crime, suggesting that its success can be measured by its impact as a deterrent.
Using in-house technology we looked at mentions of stop and search between 1 January 2018 and 12 August 2019. During that period stop and search was mentioned almost 1,300 times in the press and parliament in connection with its deterrent effects. But only around 100 times in connection with its use as an investigative power.*
For example in Boris Johnson’s first statement to the House of Commons, speaking of his track record as Mayor of London, he said: “We reduced knife crime in London with a very active policy of stop and search.”
This could, in theory, refer to stop and search leading to the prevention of imminent crimes (i.e. knives being taken off the street before an attack was committed) as opposed to a more medium or long term crime deterrence.
However we couldn’t find evidence looking at the immediate preventative qualities of stop and search, though appreciate it may be a factor.
Regardless, the College of Policing publicly puts more weight on the use of stop and search to detect rather than deter (or prevent) crime, writing: “As stop and search is an investigative power, it may be better to see crime reduction as a useful by-product rather than its main aim.”
Professor of Global Policing at University College London Ben Bradford told us: “The success of a stop and search should be measured at the point at which a crime is or isn’t uncovered, not at a potential future deterrent effect.”
This reflects the fact that the majority of stop and searches are conducted under Section 1 of the Police and Criminal Evidence Act (PACE) 1984 where the legal basis is that they are used to investigate crime.
However, from the start, stop and search under PACE was advocated for its supposed deterrent effects alongside its investigatory use.
The PACE legislation grew out of the Philips Report, an inquiry into criminal procedure, and in a debate on the report then-Home Secretary William Whitelaw said: “Police powers of stop and search have generated particular controversy.
“They are available only in certain parts of the country, yet the police there find them an essential tool in preventing and detecting crime.”
In more recent stop and search initiatives, deterrence has also been a consideration. For example Operation BLUNT 2, a Metropolitan Police initiative, aimed to reduce knife crime by, among other things, increasing searches for weapons.
An official report described the purpose of the initiative as “enhanced stop and search to deter the carriage of lethal weapons.”
So it’s important then when evaluating the effectiveness of stop and search to consider both aspects, deterrent and investigatory benefits.
Does stop and search work as a deterrent?
The evidence on stop and search suggests it is not a particularly good deterrent.
A recent paper by Professor Bradford and colleagues Dr Matteo Tiratelli and Dr Paul Quinton says:
“We used ten years of police, crime and other data from London to investigate the potential effect of stop and search on crime.
“We find that S&S has only a very weak and inconsistent association with crime. While there is some correlation, most notably in relation to drug offences, we conclude that the deterrent effect of S&S is likely to be small, at best…
“We struggled to find evidence of an effect of S&S on violent crime.”
They found that after a 10% increase in stop and search during a given month, recorded drug offences would be 1.85% lower during the next month, while non-domestic violent crime was 0.01% lower.
Another study into stop and search’s deterrent effects was the Home Office’s analysis of the outcomes of Operation BLUNT 2.
It said: “Overall, analysis shows that there was no discernible crime-reducing effects from a large surge in stop and search activity at the borough level during the operation.”
However the paper does note that “it does not necessarily follow that stop and search activity does not reduce crime.
“It is possible that there are localised crime-reducing effects of stop and search activity that are masked when analysing data on such a large geographic area.”
Slightly more positive findings come from New York City, where an evaluation of Operation Impact, an initiative where more police were deployed to high crime areas (“impact zones”) and encouraged to conduct investigative stops, found:
“Impact zones were also significantly associated with increases in investigative stops for suspected crimes, but only the increase in stops made based on probable cause indicators of criminal behaviors were associated with crime reductions.
“However, probable cause-related stops were a relatively small fraction of the total number of investigative stops, suggesting that there were excess stops that had little crime suppression benefits.”
Does it work as an investigative power?
An investigative power could be seen as effective in two broad ways. One is that the power allows the police to detect a crime that has not yet been detected (for example possession of drugs or weapons). The other is the police find something that helps them solve a crime they have already detected.
As an investigative power, there is evidence that stop and search “works”, at least in the sense that stop and search leads to the police uncovering more crimes, particularly drug offences.
Across England and Wales in 2017/18 around 30% of stops and searches led to what’s called a “positive outcome.” This refers to cases where action is taken against people who were stopped and searched. This includes arrest cases but also covers other resolutions like warnings and Penalty Notices.
In London 60% of positive outcomes were related to drug offences, 11% to theft, fraud and counterfeit offences, and 9% were for “weapons, points and blades”.
There is an open question as to whether stop and search is better than other investigatory powers.
Professor Bradford told us, in relation to crimes of drug possession: “Without stop and search the police would be arresting more people to investigate these crimes. It is more efficient to search them.”
Dr Tiratelli on the other hand told us: “Stop and search is used very differently by different police forces, indicated by the level of use.
“The Met see it as an essential investigatory tool, while other police forces, for example Essex, find other powers which they use to investigate crime.”
So stop and search may be an efficient way to detect certain types of crimes but it doesn’t appear to be universally embraced by police forces as an essential investigatory tool.
What is the opposition to stop and search?
Whether or not stop and search is effective either as a deterrent or investigatory power, this needs to be considered alongside its possible negative effects.
The College of Policing says: “There are substantial risks associated with stop and search being used incorrectly or inappropriately.”
“Disproportionate use of stop and search against particular social groups—most notably black and minority ethnic groups and young people—may increase their perception that they are being targeted unfairly.
Black people are nearly ten times as likely to be stopped and searched as white people. Three in every 1,000 white people were stopped and searched in 2017/18, compared to 29 in every 1,000 black people.
The College of Policing continues: “Perceptions of the police making unfair decisions and being disrespectful are linked with lower levels of police legitimacy.
“This in turn reduces the public’s willingness to not break the law and cooperate with the police, eg, by not reporting crime, suspicious activity or providing information.
“This is likely to make the police’s job harder in the long run.”
Polling by Yougov on behalf of the Criminal Justice Alliance in 2017 found a significant proportion (36%) of young BAME people have less trust in the police because of what they know about stop and search.
* Full Fact did the following searches to collect this data
(prevent OR preventions OR reduce OR reduction OR lower OR “bring down”) AND stop AND search = 1,282 hits across national media and parliamentary mentions.
(investigate OR investigative OR investigatory OR solve) AND stop AND search = 111 hits across national media and parliamentary mentions.