CPS charging decisions - examining demographic disparities in the outcomes of our decision making
- Introduction
- The context for this work
- How we make our charging decisions
- The research methodology
- The research findings
- Table: Charge Rates by suspect’s ethnicity
- The limitations of this research
- What we’re doing next
- External scrutiny
- Conclusion
Introduction
"A fair and functioning justice system is vital for any democratic society. The decisions we make at the CPS are at the heart of criminal justice in England and Wales and we know that every decision to charge - or not - has a potentially profound impact on suspects, defendants, victims and the wider public.
"As DPP, I have a personal responsibility to ensure that all our decisions are taken in a way that is transparent, consistent and fair. It’s a responsibility I do not take lightly and one I have put at the heart of my priorities during my tenure. Fairness and transparency have been the driving force behind our commitment to undertake this research and publish at this early stage - ensuring our decision making is proactively, independently and robustly scrutinised.
"This research has found evidence of disproportionality in relation to ethnicity in the outcomes of our decision-making. These findings are troubling. While we cannot yet identify what is driving these disparities, it is clear we must do further work to establish this as a matter of urgency. I am personally committed to ensuring we take whatever action is needed to reduce disproportionality in our Service."
Max Hill KC DPP
"We committed to undertake this research as part of our 2025 Inclusion and Community Engagement strategy. While previous reviews had reflected favourably on the CPS, we understand how important constant check and challenge is in relation to issues of disproportionality and inclusion.
"The findings of this report are concerning and we know that this feeling will be shared by our colleagues, communities and stakeholders as well as those involved in the cases that we handle.
"Understanding and tackling the drivers of disproportionality is a top priority for our senior leadership team. While issues of disproportionality cut across the whole of the criminal justice system, and wider society, there are no excuses. If there are actions we can take to reduce disproportionality then we will do so and we’ll continue to work closely with our partners to ensure that the justice system as a whole is transparent, fair and inclusive."
Baljit Ubhey, Director of Strategy and Policy
Grace Ononiwu, Director of Legal Services and Inclusion Champion
The context for this work
It has long been established that there is an over-representation of ethnic minorities within the criminal justice system.
The 2017 Lammy Review noted that men and women from ethnic minority backgrounds were vastly overrepresented in the prison population – making up 25% of prisoners despite making up only 14% of the overall population.1
The Ministry of Justice’s latest (December 2021) publication on statistics on race and the criminal justice system found that minority ethnic groups appear to be over-represented at many stages throughout the criminal justice system compared with the White ethnic group. The greatest disparities they noted appeared at the point of stop and search, custodial remands and the prison population.2,
As part of its overarching review of disparities across the criminal justice system the Lammy Review looked at CPS charging decisions from 2014 and 2015. It concluded that CPS charging decisions were broadly proportionate – the CPS prosecuted the same proportion of cases across all ethnic groups. In 2014/15 irrespective of a defendant’s ethnicity we took the decision to prosecute in approximately 70 to 72% of cases.
However, given the scale of the problem that exists throughout the criminal justice system, the CPS undertook a commitment, as part of our latest Inclusion and Community Engagement Strategy, to examine our own work more closely.
Ensuring defendants are treated fairly is central to upholding the rule of law - any evidence that defendants from a minority ethnic background have a different, or indeed worse, experience of the criminal justice system undermines that foundation.
We know that our role is central to the functioning of justice and the decisions we make are life-changing so we’re committed to holding ourselves rigorously to account.
How we make our charging decisions
At the Crown Prosecution Service we share responsibility for making charging decisions with the police.
In less serious cases, which account for around two thirds of all criminal offences, the police make the decision about whether a suspect can be charged with an offence.
In more serious cases, the charging decision is made by the CPS. In these cases, the police will conduct an investigation and send a file to the CPS only when they think the case against the suspect is strong enough that it has the potential to pass the CPS charging test.
We take the decision to charge the suspect in the vast majority of cases which are sent to us by the police. In the most recent quarter for which data is available (July to September 2022) we charged the suspect in 79.4% of cases which were sent to us.3,
We make all our decisions by following a legal test which is set out in our Code for Crown Prosecutors. Prosecutors must only start or continue a prosecution when the case has passed both stages of the Full Code Test. The exception is when the Threshold Test may be applied (see below.)
The first stage of the Full Code Test is the Evidential Stage. This asks whether the evidence is sufficient to provide a realistic prospect of conviction against each suspect on each charge.
In other words, does our prosecutor believe that, on the basis of this evidence, a judge, a panel of magistrates or a jury would be more likely than not to convict the defendant?
To make this determination our prosecutors will examine all the evidence the police have provided. They will consider whether the evidence the police have collected is credible and reliable. They will also consider whether there is any evidence which undermines the case put forward by the police or which would support any defence the suspect may be likely to rely on.
If a case passes the first stage of the test the prosecutor will then consider the second stage.
This asks whether it is in the public interest to prosecute this person for this offence.
To make this determination our prosecutor will consider factors such as the seriousness of the offence, the harm caused to the victim and the suspect’s age and level of maturity at the time of the incident.
If a case is being considered under part two of the test a prosecution will usually take place unless our prosecutor is satisfied that the public interest factors tending against prosecution outweigh those tending in favour.
If a case passes both stages of the Full Code Test then we will charge the suspect with a criminal offence. If a case doesn’t pass either part of the test then we will not charge the suspect. The case may be sent back to the police for further investigation if there is a possibility that they could uncover more evidence to strengthen the case.
In limited circumstances, where the Full Code Test is not met, the Threshold Test may be applied to charge a suspect. The threshold test is used when a full file of evidence is not yet available but the seriousness of the case justifies an immediate decision and there are reasonable grounds to believe that continuing the investigation will provide evidence that would give a realistic prospect of conviction.
The research methodology
In 2021 we commissioned researchers from the University of Leeds to interrogate the outcomes of CPS charging decisions for evidence of disparity.
The study we asked the University of Leeds to carry out used a different methodology to that used within the Lammy Review, and previous studies which have looked at the same issue, including Mhlanga (1999) and the Gus John Partnership (2003).
The research carried out for the Lammy Review calculated the ‘Relative Rate Index’ at regular intervals in the criminal justice process to isolate the effect of decision making at each stage, essentially asking whether there were specific stages or decision points within the criminal justice system where disproportionality emerged or changed. This study found that CPS decision making had little to no effect on disproportionality within the criminal justice system.
The University of Leeds research goes further by using regression analysis to allow for closer scrutiny of the outcomes of our decision making. This technique is recognised as industry standard and allowed the researchers to control for different variables to isolate the impact other variables had on the relative outcomes of our charging decisions.
For example, the researchers controlled for variables such as age, sex4 and crime type to isolate ethnicity as a variable in order to understand whether this produced any disparities in relation to the outcomes of our charging decisions.
This research looked at decisions made between January 2018 and December 2021 that led to a charge, a caution or no further action. This amounted to almost 195,000 cases.
The research findings
The University of Leeds research has found that there is evidence of disproportionality in the outcomes of our legal decision making.
The research found disproportionality relating to ethnicity with statistically significant differences in the outcomes of our charging decisions when ethnicity was isolated as a variable.
White British suspects had the lowest charge rate compared to all other ethnicities with 69.9% of cases resulting in a charge. By contrast Mixed Heritage suspects had a charge rate of between 77.3% and 81.3%.
This pattern was relatively consistent across England and Wales.
Table: Charge Rates by suspect’s ethnicity
| Suspect’s ethnicity | % Charge |
|---|---|
| A1 – Indian | 71.8 |
| A2 – Pakistani | 73.5 |
| A3 – Bangladeshi | 73.3 |
| A4/O1 – Chinese | 73.3 |
| A9 – Any other Asian | 73.4 |
| B1 – Caribbean | 77.5 |
| B2 – African | 74.7 |
| B9 – Any other Black | 76.5 |
| M1 – White and Black Caribbean | 81.3 |
| M2 – White and Black African | 79.5 |
| M3 – White and Asian | 78.4 |
| M9 – Any other Mixed | 77.3 |
| O9 – Any other | 77.9 |
| W1 – White British | 69.6 |
| W2 – White Irish | 75.3 |
| W9 – Any other White | 75.5 |
| Not provided | 52.8 |
| Not stated | 84.9 |
When the data was included in regression analysis and variables such as sex, age, and crime type were controlled for, it showed that ethnic minority defendants are significantly more likely to be charged for a comparable offence than White British defendants.
The limitations of this research
While this research provided for closer scrutiny of the outcomes of our decision making it does have some significant limitations.
For example, it was not possible to control for every factor which could legitimately affect the outcomes of charging decisions.
In particular, the researchers were not able to take into account suspects’ previous offending history. We know that ethnic minority defendants are overrepresented in the proportion of people convicted each year so it’s possible that existence of prior criminal records may be a factor in some of the disproportionality that has been uncovered. For this reason, the University of Leeds researchers have advised that the reported disparities should be interpreted with caution.
Moreover, while this research tells us that these disparities exist, its nature as a statistical analysis means that it cannot tell us anything about why the disparities are there.
There may be factors affecting the charge rate which are beyond our control - for example wider socio-economic factors. However, the scale of the disproportionality may also suggest that there are factors affecting our decisions which shouldn’t be.
Our priority now is to understand what factors may be behind the disparity, to determine what action may be needed to address it.
What we’re doing next
The research so far has focused purely on the type of offence and high-level characteristics of the defendant (their sex, age and ethnicity). To better understand why the data gives the results it does we need to look at a sample of these cases in more depth.
We have designed a comprehensive programme of further research which is intended to help identify the factors which may be causing the disproportionality found by the University of Leeds. We have recruited a team of social researchers who will take forward this work.
For the next stage of this research, our team will analyse how our decisions may be impacted by things like previous offending history, the language and presentation of evidence files to the CPS and the processes that we ask our prosecutors to follow when reviewing these files. We will also break the results down by CPS Areas and consider whether smaller or larger disparities can be linked with factors in the Area, for example do local demographics have an impact, or can we see a relationship with the training that we offer?
These are big questions, and we know that there are likely to be a number of different factors which impact the results we’re seeing.
External scrutiny
We want to ensure we approach this work in the most effective way – it is important that our research into the causes of disproportionality is both credible and robust because it has important implications for the CPS, the criminal justice system and the communities we serve.
To support us with this work we have set up an independent Disproportionality Advisory Group. Its membership is made up of academics, legal professionals and third sector partners with experience in issues of racial equality, disproportionality and the justice system.
This group will be chaired by Susie Uppal, formerly the Director of Legal Enforcement for the Equality and Human Rights Commission where she led on the Commission’s Inquiries, investigation and Judicial Review work, holding private and public sector bodies to account. Susie is a solicitor specialising in public and regulatory law, who, in addition to extensive experience in private legal practice, has held several senior roles in regulatory bodies, was an independent member of the Regulatory Affairs Board of the Law Society and is co-editor of the Blackstone’s Guide to the Equality Act 2010. Susie is the Chief Executive Officer of the Press Recognition Panel, an independent oversight body set up after the Leveson Inquiry into the culture, practices and ethics of the press.
The role of the Advisory Group is to scrutinise the scope and findings of our next stage of research – providing constructive challenge and quality assurance for our work. We will also work closely with them to refine and develop any future proposals for actions we could take to mitigate disproportionality.
We intend to complete this next phase of research by September 2023. It is important to note that this work is complex – given that the research will be looking for causes of disproportionality which are likely to be hidden there is a risk that the sources will be difficult to identify. We will continue to work closely with the advisory group to mitigate this risk and ensure our ongoing research is focussed and pragmatic.
Balancing this need for a strong evidence base to support our actions against the risk of delaying any measures we should take to address the drivers of disproportionality will be central to our work with the Advisory Group.
Conclusion
The Lammy Review praised the CPS’ openness to external scrutiny, and we intend to build on that good practice in how we approach this research. As we know more, we will share more and where we find anything that we, or our partners, need to change or improve we will take swift action.
We know that these early findings will be difficult for people both inside and outside the CPS and that isn’t helped by the uncertainty, at this stage, about the cause.
We are proud of our diverse workforce – we aim to place inclusion at the heart of what we do, and we want everyone at the CPS to feel supported, engaged and have full confidence in our work.
For the public, our role is to deliver justice through independent and fair prosecutions. Every victim, witness and defendant should expect the same standard of care, attention and decision-making in their case, and we will keep challenging ourselves to ensure that we are getting that right.