Perspectives
Oli Hutt, Head of Analytics
Friday 19 July 2024
WATCH DISPATCHES EPISODE HERE: Britain's Unsolved Crimewave
SEE INTERACTIVE MAP HERE: Top 10 areas by police force: proportion of crime unresolved
Context
Police recorded violent crime has increased substantially over the past decade from around 12 offences per 1000 people in 2012-13 to 35 offences per 1000 people in 2022-23. Policing understandably prioritises these crimes over less harmful offending such as bicycle theft, and vehicle crime. In late 2023, Channel 4’s Dispatches asked us to look at what they termed ‘community crimes’ - bicycle theft, burglary, robbery, theft from a person, and vehicle crime - and whether there were areas of the UK where neighbourhoods might feel forgotten by the police because of the lack of resources to deal with these crimes.
We wanted to look at neighbourhoods where crimes went unresolved, where people might feel let down by the police because of the high proportion of these crime types that went unresolved. After considering a number of definitions for ‘unresolved’ crimes we opted for a simple approach of only counting crimes as unresolved where no suspect was identified. To do this, we used data from police.uk which includes individual-level data for each crime including the month it happened, the approximate location (both the Lower Super Output Area - which we'll call a neighbourhood - and the location as coordinates, though these aren't exact to preserve anonymity for victims), the crime type, and (in most cases) the outcome of each crime.
The main crime data covered the calendar years of 2021, 2022, and 2023. Outcomes data were available up to February 2024, so outcomes recorded in early 2024 are included in our analysis for crimes that happened within our analysis period. The first thing we needed to do was check which police forces had reliable data. We already knew about a number of data quality issues that would exclude some police forces from our analysis, but we looked at each force to see if there were any gaps or abnormal patterns in their crime and outcome data. The excluded forces are detailed in the table below with the reason they were left out of the analysis. This left us with 35 police forces across England and Wales.
Police Force | Reason for exclusion |
Avon and Somerset Constabulary | Over 75% of offences do not have an outcome: From Police.UK: Due to a change in IT systems there are known issues with crime and outcome data since October 2015 as latitude and longitude information is missing from approximately 2000 crimes each month. |
British Transport Police | Outcomes data are not provided to Police.UK |
City of London Police | The unique position of the City of London within the wider Metropolitan Police Service makes it an unusual case within policing. It also has a consistent high proportion of unknown outcomes (10-20% each month) |
Devon and Cornwall Police | Data is unavailable for a large part of the time period being analysed and outcomes data are not available for approximately half of the offences that are included. From Police.UK: Due to a range of issues, including the implementation of a new record management system in November 2022, outcomes data is unreliable. |
Greater Manchester Police | Recording and reporting issues have been widely reported on and data have not been available for several years. From Police.UK: Due to a change in IT systems no crime, outcome or stop and search data is available from July 2019 onwards. |
Gwent Police | Excluded for having a consistent relatively high proportion of offences with unknown outcomes (10-20% each month). |
Lancashire Constabulary | Excluded for having a consistent relatively high proportion of offences with unknown outcomes (approx. 20% each month). |
Northumbria Police | Excluded for having a consistent relatively high proportion of offences with unknown outcomes (approx. 20% each month). |
Police Service of Northern Ireland | PSNI does not supply outcomes data to Police.UK |
National trends in neighbourhood crime
Before we start looking at neighbourhoods it's worth considering the general picture. Across the five crime types we’ve looked at, the majority of crimes go unresolved, but this does vary by crime type and police force. Overall, around 90% of bicycle theft, vehicle crime, and theft from the person goes unresolved. For burglary it's 75% and for robbery 65% go unresolved. Although a lot has been said about the increase in violent crimes, over the past three years the number of robberies and thefts from a person have also increased significantly.
Bicycle theft shows some significant seasonal effects. Unsurprisingly, bicycles get used more when the weather is better and this provides would-be thieves with greater opportunity to steal a bicycle because there are just more of them around.
The proportion of these community crimes that get resolved varies by police force as well. The Metropolitan Police Service (MPS) is a substantial outlier, with only 3% of community crimes in London being resolved compared to 15% in Surrey which has the next lowest proportion among the forces we could look at and 45% Norfolk which had the highest proportion of resolved cases
London also has significantly more crime than any other force, with over 800,000 community crimes recorded over our three year period, of which 24,508 were resolved. Four police forces (West Midlands, West Yorkshire, Thames Valley, and Kent) resolved more community crimes than the Met despite having significantly lower volumes
Neighbourhoods with high volumes of unresolved crime
To help Channel 4’s Dispatches focus on specific neighbourhoods, we looked at both the volumes of unresolved crime and proportions of crime that went unresolved to create ‘top 10’ neighbourhoods for each police force in our sample. To avoid our Top 10s by proportion from being filled with neighbourhoods that had one or two crimes in it that both went unresolved (Thus making crime in that area 100% unresolved) we only looked at neighbourhoods that had at least 50 unresolved crimes over the three years of data we were using. Because of this cut-off, Dyfed-Powys only had a top 5 - i.e. only five LSOAs experienced more than 50 unresolved crimes over the three year period. 32 forces could have had a Top 20 as well - only Cumbria and North Wales (and Dyfed-Powys again) had fewer than 20 LSOAs with under 50 unresolved crimes.
High volumes of unresolved crime were apparent in exactly the places you might expect - areas that had a high number of targets for offenders. The highlighted neighbourhoods were almost always in dense urban areas. It’s no surprise that bicycle thefts happen where there are a lot of bicycles; theft from the person and robbery happen where there are lots of people. These are well known in the crime of places literature and are called crime generators. Places like the West End or Stratford in London, where huge numbers of people go every day to shop, eat, work, and meet friends. Places like Cambridge train station where a huge number of bicycles are kept every day and provide a target-rich environment for bicycle thieves. You can see what Channel 4 thought of these areas by watching Britain's Unsolved Crimewave: Dispatches here.
Reducing crime in these locations can be challenging because it is exactly the legitimate activities happening in them which makes them attractive to offenders. Thankfully there is a rich literature on situational crime prevention and problem-oriented policing. It’s been more than 20 years since Cornish and Clarke [1] (2003) produced their 25-techniques of situational crime prevention. We’ve worked with a number of clients to help them identify and respond to crime. Last year, we completed a problem profile of Night-Time Economy violence across Cleveland to help partners reach an agreed definition and understanding of NTE violence to ensure they deliver a joined-up response across agencies. We’ve also completed a number of Strategic Needs Assessments for clients to help them understand the scale and nature of offending in their area.
Neighbourhoods with high proportions of unresolved crime
Perhaps the more interesting or unexpected results came from looking at where the proportion of community crimes being resolved was lowest in each police force area. We have purposefully avoided talking about ‘the worst’ places for three reasons: we haven't been able to include every police force in England and Wales; there will be some differences in recording practices across forces; and there are always a number of crimes that don't have recorded outcomes. So absolutist comparisons should be avoided, but there are still things we can learn from looking at the neighbourhoods with the highest proportions of unresolved crime in each force. As part of our analysis we also created this interactive map which allowed us and Channel 4 to get a better understanding of the areas, what was nearby, and how they were distributed across each force area.
For some of the smaller, less populated areas the Top 10s still included areas that had a relatively high proportion of resolved crimes. For most forces however, the Top 10 LSOAs represented areas where less than 10% of community crimes were resolved - that is, led to a suspect being identified. In some areas, substantially less.
We discovered that 167 neighbourhoods have a 100% unresolved rate. i.e. over the past three years, they've had at least 50 crimes happen and in none of these cases was a suspect identified. All but one of these neighbourhoods are in London. We found this quite surprising, even considering the unique geographic and demographic characteristics of London when compared to other police forces. LSOAs can vary in size quite substantially as they are defined by population not area, but in London they tend to be quite small. One area of Richmond-upon-Thames (shown below) experienced 141 community crimes over three years, none of which had a suspect identified. Of all the neighbourhoods where none of the crimes were resolved, this area had the highest volume. This wasn’t driven by any one crime type - it was mostly a mix of vehicle crime (n=68), bicycle thefts (n=36), and burglary (n=28).
The distribution of these ‘100% unresolved’ areas is very different to the volume areas. Rather than being focussed around city centres and areas with large numbers of potential targets they seem almost randomly distributed. The map below shows just how spread out these neighbourhoods are and you can explore all the top 10 LSOAs by police force on our interactive map here.
Conclusion
Despite the police-uplift programme, the number of police personnel is substantially below what existed ten years ago. When considering population growth, an additional 32,000 officers would be needed to return to 2010 levels of 4.42 police personnel per 1000 population. In particular, police community support officers (PCSOs) who were primarily focused on neighbourhood policing have seen their numbers more than halve over the same time period.
Labour’s campaign manifesto promised to ‘restore neighbourhood policing with thousands of extra officers’ (and you can read more about what we thought of all the manifestos here) which will be a welcome boost to police forces that have seen substantial resourcing pressures over the past decade. This doesn’t remove the challenge that they have in deciding how to prioritise different issues. We’ve written previously about what the future holds for serious violence policy, the policing response to violence against women and girls (VAWG), and it was nearly a year ago that we highlighted how police forces have been criticised by His Majesty’s Inspectorate of Constabulary and Fire and Rescue Services (HMICFRS) for basic errors in their response to more ‘traditional’ crimes such as burglary and vehicle crime.
The police uplift programme had already brought extra scrutiny by policy makers, budget holders, and other interested parties keen to see how policing utilised the additional resource. And don’t forget you can see what Channel 4 discovered when they visited some of these areas by watching the episode Britain’s Unsolved Crimewave: Dispatches. The question of what a ‘productive’ police force looks like will no doubt get substantial attention over the coming months and years and we’re currently conducting our own research in this area as well. Alongside that work, we’ll continue to work with police forces and other organisations across the criminal justice system to help them understand and respond to the challenges they face and ultimately help make communities safer.
Footnotes
[1] Cornish DB and Clarke RV. (2003). ‘Opportunities, precipitators, and criminal decisions: A reply to Wortley’s critique of situational crime prevention’. Crime Prevention Studies, 16, pp 41–96.
Comentários