Skip to content

Geopolitics & Security

Justice Denied: How the Government Can Change Its Approach to Catching and Convicting Offenders

Paper18th November 2021

Chapter 1

Executive Summary

In recent years, the criminal-justice system in England and Wales has become slower and less effective: the police are taking longer to respond to incidents and detect crimes, fewer offenders are being charged and prosecuted, and victims are having to wait longer than ever to see justice served in the courts. We call this growing problem the “justice gap” and it matters for two key reasons.

Figure 1

National charge rate for offences recorded from March 2015 to March 2021


First, if the justice system fails to offer any deterrence, offenders are likely to be emboldened to commit crimes, putting individuals and communities at risk. The evidence suggests offenders respond more to certainty of sanctions (in other words, the likelihood of being caught) than the severity of punishment (the length of sentences). Yet currently, fewer than 8 per cent of offences lead to charges being brought, and even in early 2020 – before the pandemic had led to court closures – the average time taken to bring an offender to justice stood at 164 days.

Second, in order to function effectively, our justice system relies on the willingness of victims and witnesses to report crimes and give evidence in court, which in turn requires them to trust the system to seek justice on their behalf. Once they begin to lose that trust, the system rapidly loses public consent and grinds to a halt.

In this paper, we examine the reasons for the growing justice gap and identify a number of factors behind the recent deterioration in performance:

  • Resource constraints: Since 2010, both the police and the Crown Prosecution Service (CPS) have been forced to reduce their workforces at a time when demand pressures have grown as a result of a rise in high-harm crimes. Such offences include knife crime and sexual offences, which are time-consuming to investigate. In addition to workforce reductions, the closure of specific infrastructure – such as police-custody suites and courts – is likely to have contributed to lengthier delays.

  • Growth in complexity (and failure to keep pace with technological change): Policing leaders have stressed that one of the biggest practical barriers to charging offenders is that they lack the capacity to process all the material downloaded from suspects’ phones, while lacking the technological tools to analyse this material efficiently and accurately.

  • Skill shortages: Both the police and the CPS are facing a shortage of trained, experienced investigators. Her Majesty’s Inspectorate of Constabulary and Fire and Rescue Services (HMICFRS) has described this issue as a “national urgency” which is hampering the quality and speed of investigations.

  • Structural problems: Bringing offenders to justice requires a number of criminal-justice agencies to work together, with multiple handover points along the journey from offence to conviction, yet multi-agency coordination remains poor. In particular, the relationship between the police and the CPS has become characterised by buck-passing rather than collaboration.

  • Lack of national focus and poorly designed policy: There has been a failure of leadership from the top of government around charge rates and prosecutions. Indeed, the government has inadvertently made the problem worse by introducing policies that were not well thought through, such as the changes to bail introduced in 2017, which subsequently led to a huge increase in the number of suspects “released under investigation”, with no time limits or conditions attached.

The government has signalled in its latest Spending Review that the resourcing problem will improve over the next three years. But it is now critical that new spending is accompanied by reform. In this paper, we will also set out a number of practical recommendations:

  • New minimum standards on dealing with crime: Government should publish new minimum expectations – specifically focused on key metrics including response times, charge rates and the time taken to bring offenders to justice.

  • A national taskforce to examine the role of technology in improving investigations: Government should appoint a taskforce to explore ways to improve investigative capability within policing – for example, by driving better collaboration between government, technology companies and universities in designing technological solutions to reduce the time taken to download digital evidence, and improve the recruitment and retention of skilled, trained investigators.

  • Joint prosecution targets for the police and the CPS: Government should incentivise the police and the CPS to work jointly in order to streamline and improve the charging process, rather than accept the culture of buck-passing.

The government clearly understands the critical importance of law-and-order issues to the public: pledges to increase police-officer numbers and to toughen sentences have proved popular. It urgently now needs to turn its attention to re-establishing the link between offence and punishment – without which the justice system faces being permanently undermined.

Chapter 2

The Evidence on Deterrence and Why It Matters

Our previous policy reports on criminal justice have focused on reducing crime by addressing its underlying drivers, for example through earlier identification of vulnerability and a more targeted approach to supporting families with multiple needs. In this paper, we look more explicitly at the role of criminal-justice agencies in deterring offending and in bringing offenders to justice.

Deterrence Theory

In broad terms, punishment may be expected to affect deterrence in two ways. First, by increasing the certainty of punishment, the risk of apprehension may deter potential offenders from committing a crime. For example, if there is an increase in the number of traffic police patrolling motorways, some drivers may reduce their speed in order to avoid receiving a ticket.

Secondly, the severity of punishment may influence behaviour if potential offenders weigh the consequences of their actions against the potential benefits and conclude that the risks of punishment are too severe. This is part of the logic behind harsher sentencing laws, such as the “three-strikes” policy (where the third offence leads to a much longer prison sentence): to leverage the threat of severe sentences in order to deter some people from engaging in criminal behaviour.

For decades, the criminal-justice policy debate has been dominated by the question of severity, largely through the lens of sentencing reform. Since the 1990s, there has been a raft of bills introduced to amend sentencing policy including, most recently, the government’s Crime and Policing Bill, which introduces mandatory minimum sentences for serious sexual offenders.[_] This is reflected in prison population trends, which show how harsher sentencing has led to a steady rise in prison numbers. However, as the next section demonstrates, prioritising sentencing severity over certainty of punishment as a mechanism to deter offending is not necessarily supported by the evidence.

Certainty Versus Severity

Criminological research over several decades concludes that the certainty of punishment is a stronger deterrent than increasing the severity of punishment. In a landmark 2001 study, Nagin and Pogarsky surveyed a sample of college students to assess the likelihood of drinking and driving. The authors found that the certainty of punishment was a more robust predictor of deterrence than severity. Increasing the probability of apprehension by 10 per cent was predicted to reduce the likelihood of drink-driving by 3.5 per cent. No similar effect for severity was found.[_]

One of the most obvious ways of increasing the certainty of punishment is to increase the likelihood of police detection. A review by Sherman and Eck (2002) of studies examining changes in crime before, during and after industrial action by the police found that crime rates increased markedly during strikes, demonstrating the deterrent effect of a visible police presence. Similarly, a study conducted in the aftermath of the 2005 London terror attacks found that a 10 per cent increase in visible police activity reduced crime by 3 to 4 per cent.[_]

There is evidence of a more specific deterrent effect where prolific or repeat offenders are targeted by the police.[_] Increasing the likelihood of offenders being caught can reduce reoffending, particularly when that increased likelihood is effectively communicated to offenders, and pathways out of crime are provided by partners. There is a risk, however, that if offenders are not caught and think they have “got away with it”, they are more likely to reoffend.[_]

The implications in terms of police numbers are well understood by policymakers, even if they have not always been publicly acknowledged. (Between 2010 and 2018, a succession of home secretaries denied that there was any link between rises in crime and falling police-officer numbers. However, an unpublished Home Office research paper leaked to the press in 2018 concluded that “resources dedicated to serious violence have come under pressure and charge rates have dropped. This may have encouraged offenders”.[_]) Similarly, in a review of the evidence exploring the relationship between police numbers and crime rates, the criminologist Ben Bradford concludes that there is strong evidence to suggest a relationship between increased police numbers and reduced crime, particularly in terms of property-related crime, where a 10 per cent increase in police officers led to a 3 per cent reduction in property crime.[_]

Of crucial importance is not only the number of officers, but how effectively they are deployed. Random or reactive patrols, which involve officers patrolling places regardless of the crime rate, have been shown to have no crime-reduction effect.[_] However, visible police patrols can reduce crime provided they are targeted in the areas – or “hotspots” – where crime is concentrated. A systematic review by Braga et al (2019) concluded that, overall, hotspot policing is effective at reducing crime. In fact, 62 of the 78 studies included in the review reported that crime fell in the locations targeted with an increased police presence and activity, compared with other locations that were not targeted. One study found that property crime fell by 31 per cent in hotspots patrolled by marked police cars compared with hotspots where “business as usual” was maintained.[_] (It is notable that property crime did not fall in hotspots patrolled by unmarked police cars, highlighting the importance of the police providing a visible deterrent.) There is also evidence about the impact of community policing in reducing crime and disorder, a key feature of which is the involvement of community partners in identifying, prioritising and solving neighbourhood problems.

What might explain the primacy of certainty over severity in affecting offenders’ behaviour? Partly, it may simply be a function of the fact that humans are not always capable of effectively calibrating risk, which means that delayed consequences, even if severe, such as long prison sentences, are often given less weight than immediate consequences, possibly even more so for offenders. The fact that sentencing guidelines are not well understood by the general public, let alone the offending population, may also be a determining factor in the primacy of certainty over severity.

The Significance of Timeliness

There is a relationship between the principles of certainty and timeliness. Swift justice is likely to enhance perceptions of certainty. Conversely, the longer it takes to charge and prosecute offenders, the more that certainty is likely to be undermined. This is partly because victims will disengage from the process. Delays between the offence and the completion of a case undermine the effectiveness of the sentence in the eyes of victims and offenders. Victims become disillusioned, while offenders are encouraged to plead not guilty (on the grounds that the longer the case goes on, the more likely it is to collapse).

This is one of the reasons why the current backlog in the courts, which has been exacerbated (though not caused) by the pandemic, represents such a risk to the legitimacy of the justice system. More detail on this is set out in the next chapter.

Case Study

The Application of Swiftness and Certainty in Policy

The Application of Swiftness and Certainty in Policy

In a criminal-justice context, the principles of swiftness and certainty are most commonly associated with a probation project pioneered in Hawaii (“Project HOPE”) in 2004. The premise behind the HOPE model was that, following a warning hearing from a judge at the beginning of the probation period, the offender would undergo regular mandatory drugs testing (with rehabilitative support available if requested). Any breach would result in an immediate custodial sentence, even if only for one or two nights.

The lessons from Project HOPE also, in part, drove the drug-court movement in the US. Drug courts employ a programme designed to reduce drug use and the risk of recidivism through judicial supervision and graduated sanctions. The core components of drug courts are:

  • Judicial supervision and monitoring of the offender

  • Sanctions generally being graduated depending on the number of breaches and their seriousness

  • Use of incentives and available rehabilitative support

In 2012, following the 2011 riots, the UK government published a White Paper on Swiftness and Certainty, setting out a range of measures designed to:

  • Provide an effective deterrent to crime

  • Respond swiftly and effectively when crime does take place, so that offenders are quickly made to face the consequences of their actions

  • Ensure that offenders are punished and supported to reform

Measures included schemes that encouraged faster processing of cases through magistrates’ courts, using the early-guilty-plea scheme and the “Stop Delaying Justice” initiative. Police-led prosecutions were also extended and simplified to reduce the amount of time spent engaging with the Crown Prosecution Service at the point of charge. Neighbourhood-justice panels were also rolled out in some areas to intervene early in low-level cases through restorative-justice approaches to prevent escalation. However, most of these initiatives were dropped or discontinued after 2015, and the government has largely failed to follow through on this agenda.

Deterrence Versus Labelling

In recent years, some academics have questioned the usefulness of the deterrence concept on the grounds that it enhances the risks of “labelling”. This is the idea that contact with the criminal-justice system is in itself a negative turning point, exacerbating chances of future offending by labelling individuals as “offenders”. Indeed, one recently published study, drawing on a nationally representative sample of British adolescent twins, concluded that, rather than deterring offenders, formal engagement with the criminal-justice system made offenders more likely to commit further offences.[_]

However, while the risks of labelling ought to be taken into account, it is difficult to see how they should be prioritised over the principle of deterrence within criminal-justice policy. The logic of such an approach would effectively mean choosing to decriminalise certain offences entirely, on the grounds that such offenders would be more likely to stop offending if unpunished. Leaving aside the impact on deterrence, such a position is unlikely to garner public consent.

Having considered the evidence around deterrence, the next chapter considers the context within which policing and criminal-justice agencies are operating.

Chapter 3

Context: Rising Demand, Rising Expectations

In recent years, the operating environment for policing and criminal-justice services has become more challenging owing to budgetary pressure and rising demand. Between 2010 and 2019, spending on English and Welsh police forces declined, with the number of police officers falling by more than 20,000, while net spending on the Crown Prosecution Service fell in real terms by more than 30 per cent.

In recent years, some of that budgetary pressure has been eased. One of Boris Johnson’s first pledges on becoming Prime Minister in 2019 was to increase police-officer numbers by 20,000 (restoring them to levels last seen in 2010). The 2021 Spending Review reinforced the direction of travel, confirming real-terms increases to policing and criminal-justice budgets over the current parliament, including funding to ease Covid-19-related backlogs. However, the increase in spending will also bring an increase in expectations. Reform will be critical to make sure the money is well spent.

Rising Demand

It is impossible to assess the performance of the police and criminal-justice system in deterring, catching and convicting offenders without considering the various ways in which demand has changed.

Changing Patterns of Crime

The continuing fall in the overall volume of crime, as measured by the Crime Survey for England and Wales, has masked important changes in the pattern of modern crime. Since 2014, the number of crimes recorded by the police has risen and there has been a growth in high-harm offences such as sexual offences, knife crime and domestic abuse (Figure 2). In addition, more crime has shifted from the public into the private sphere, including online. These offences tend to be more complex to investigate and costly to deal with, yet the democratic pressure, resulting from the priorities of the public, tends to focus on crimes that take place in the public realm. As a result, while the total volume of crime has fallen, pressure on the police has actually risen in recent years.

Figure 2

Indexed trends in crime nationally from 2011 to 2020

Source: Office for National Statistics (28 October 2020), Crime in England and Wales: Appendix tables, Year ending March 2020 (Note: The numbers show the percentage change in comparison to year ending March 2014, which is represented by 100)

A More Challenging Cohort of Offenders

The above trend is reinforced by analysis of offenders entering the criminal-justice system. While the volume of people (children and adults) entering the system for the first time declined by 58 per cent between 2010 and 2019, the seriousness and regularity of their offending has increased. The proportion of “first-time entrants” (FTEs) committing possession-of-weapons offences has increased by 16 percentage points over the past ten years and is now the second most common offence committed by FTEs.[_] At the same time, the proportion of FTEs committing theft offences fell from 35 per cent to 14 per cent.

Over the same period, offenders have also become more prolific. In the latest year for which data are available, children and young people who reoffended committed an average of 3.95 reoffences each (frequency rate), substantially higher than ten years ago (3.06 reoffences).[_] The picture is similar for adults: the number of repeat offenders with more than 46 previous cautions or convictions has more than tripled since 2010.[_]

The fact that offenders are entering the justice system with a more serious and prolific level of offending than in the past suggests a failure to deal with lower-level offending earlier in their criminal careers. In recent years, policing leaders have questioned whether the fall in sanctions for low-level youth offending may have reduced deterrence, inadvertently fuelling more serious crime. For example, Metropolitan Police Commissioner Cressida Dick stated that some children who offend are “simply not fearful of how the state will respond to their actions”, and that “harsher, more effective” sentences are required to deter this group of repeat child offenders.[_]

Changing Demands on the Police

The demands facing the police are more wide-ranging than a decade ago. A growing proportion of time is spent identifying, protecting and supporting those who are vulnerable, whether responding to domestic-abuse victims, dealing with exploitation or investigating missing children. Several officers spoken to as part of the research for this report told us that safeguarding demands – managing the risks around vulnerable victims – took up a significant amount of time and resources.

“The safeguarding jobs can be the biggest jobs we get in the office in terms of, for example, trying to get people back into the country that have been flown out for forced marriages. They can take teams of people to try and resolve them. It can take weeks. If you look at it from a cold-hearted detections point of view – are we going to solve the crime? Probably not, but it will take a lot of our time.”

Investigation and Complex Crime Team, unnamed urban police force

There is also some evidence that “non-criminal” demand is sucking up more resources, including the need to respond to mental-health crises and, more recently, the need to enforce social-distancing rules during the pandemic.

Lack of available data makes it difficult to precisely quantify whether mental-health-related demand has grown, but there is a good deal of circumstantial evidence to suggest that it has. A 2018 study found that the Metropolitan Police Service receives a call about a mental-health concern once every four minutes and sends an officer to respond every 12 minutes.[_] Officers responding will invariably spend time with the person to understand what they need: a report by the Inspectorate of Constabulary suggested that, on average, this can take around three hours. They may need more officers to help, and the person might end up being detained under Section 136 of the Mental Health Act 1983 and taken to a place of safety. It is the police, not the ambulance service, that transports people to this place of safety in half of these cases. The length of the process depends on the availability of health agencies and specialist hospital beds. Often it can result in a long wait of several hours in accident and emergency for the person in crisis and the police officers accompanying them.

More recently, there has been a significant increase in anti-social behaviour with police recording 2 million incidents in the year ending March 2021, an increase of 48 per cent compared with the previous year.[_] The largest increase was seen between April and June 2020 — the period of the first lockdown — which is likely to reflect the reporting of breaches of public-health restrictions.

Having examined the context facing policing and criminal-justice agencies, the next chapter considers how effectively the system deters offenders from criminality.

Chapter 4

Assessing Deterrence in the Justice System

A core purpose of the criminal-justice system is to bring offenders to justice and, in so doing, to deter other would-be offenders and reoffenders from pursuing criminality. How effective is the system is in achieving this objective?

Responsibility for bringing offenders to justice lies across several agencies. Once a victim or third party reports a crime, the police force records it. The police then investigate the incident to decide whether to charge a suspect. If a suspect is identified and arrested, and the crime is minor – for example, it does not involve bodily harm – the police charge it themselves. Where the crime is serious, forces must seek advice from the CPS, which decides which cases to prosecute and what the charges should be, and prepares the case for court.

Figure 3

The processes involved in reporting, investigating and charging crime as well as sentencing

The processes involved in reporting, investigating and charging crime as well as sentencing

Source: TBI

Once charged, a defendant will be summonsed to appear in a magistrates’ court (where most cases are heard), with the most serious cases or those attracting a longer sentence passed on to the Crown Court. Suspects found guilty of a crime are then sentenced. Most people who are found guilty receive a fine, while others serve a community sentence, overseen by the probation service (who are also responsible for enforcing breaches), and some are sent to prison.

Below, we assess the performance of each agency against some key metrics, broken down by the three main stages of the offender journey:

  • Charging offenders

  • Convicting offenders

  • Offender management

Figure 4

Measuring effectiveness of each stage of the offender journey from charging through to management

The charging process


  • Incident-response times

  • File quality


  • Proportion of cases referred to CPS that are charged


  • Victim withdrawal rate

  • Time from offence to charge

The sentencing process


  • Prosecution rate


  • Outstanding cases (backlogs)

  • Effective trial rate

The offender-management process


  • Breach enforcement

Source: TBI

The Charging Process

In headline terms, fewer offenders are being charged than five years ago, with reductions across every major offence type (Figure 5).

Figure 5

Proportion of offences leading to charge or summons in 2014 to 2015 versus 2020 to 2021

Source: and 2020–2021

But the headline figures only tell part of the story. To truly understand the effectiveness of the charging process, it is necessary to break it down into its component parts, starting with the initial police response to an incident (how quickly the police deploy to the scene of a crime), followed by the decision on whether to make an arrest, the process of gathering evidence and the charge itself.

The Initial Police Response

The starting point of a robust criminal-justice response is how quickly and effectively the police are able to respond to a specific incident. While there are no nationally published data, there is evidence that police-response times are getting longer. For example, data submitted to the BBC by the Greater Manchester Police show that the average response time for grade-one rated calls increased from 6.5 minutes in 2011 to nearly 12.5 in 2018. Similar information provided by the Metropolitan Police shows an increase from just under eight minutes to well over nine between 2013 and 2018. Response times for lower-graded calls had also increased.[_]

The extent to which response times determine detection rates is contested. Historically, criminologists have tended to assume that the police-response time had no meaningful effect on the likelihood of catching offenders. However, a recent study by Kirchmaier et al[_] challenged that assertion by finding a significant impact: a 10 per cent increase in response times led to a 4.7 percentage point decrease in the likelihood of clearing a crime. In other words, longer response times led to fewer crimes being detected. The authors of the study concluded that the likelihood of an immediate arrest, and that a suspect will be named by a victim or witness, both increase as response time becomes faster.[_] Police officers have told us that this is especially true in domestic-abuse cases, with the speed of the initial police response having a significant impact on the likelihood of a victim supporting a charge.

As well as longer response times, there is evidence that the police have reduced the rate at which they deploy officers to the scene of the crime, with some forces having seen a 16 per cent decrease in their deployment rate since 2014 (see Figure 6). This may be the result of police forces having reformed their call-handling systems, with greater triaging of calls leading to fewer crimes being “screened-in” for further investigation.

Figure 6

Average proportion of incidents that continue to deployment in a mid-sized, urban police force

Source: Control-room data provided by an anonymous force

Arrest and Investigation

Once the police have been deployed to an incident, they need to decide whether to make an arrest and to begin the process of gathering evidence, which precedes a charge. In recent years, the trend for both arrests and charges has been falling, even while recorded crime has risen. In the year ending March 2020, the volume of arrests made was only 12 per cent of the volume of crime recorded, while only 7 per cent of offences led to a charge. This has fallen from 30.4 per cent and 16.2 per cent respectively in 2011.

Figure 7

Recorded-crime versus arrest rates for the year ending March 2020

Source: Home Office’s police powers and procedures, England and Wales, year ending 31 March 2020, and police recorded crime and outcomes open data tables 2020

It is at this point in the process – the handover between the police and CPS – that performance is arguably the most concerning. While national data is hard to come by, there is evidence to suggest that the quality of police referrals to the CPS has declined. For example, a Criminal Justice Joint Inspection Report into the police and the CPS response to rape, published in 2021, suggested there had been a decline in the quality of files being referred to the CPS by the police. The inspectors found that in over half of all the cases submitted to the CPS for a charging decision, the police file didn’t comply with national file standards. The inspectorate concluded that “the police don’t always understand what is needed for a charging decision”.[_] On the other hand, there is also evidence, including from the 2020/2021 Victims’ Commissioner report, that the CPS is charging a smaller proportion of cases that are referred to it.[_] The next chapter looks at the drivers of these declines in more detail.

Time From Offence to Charge

As well as a drop in the volume of offenders being caught and charged, the process itself has become slower. Between 2010 and 2019, the average number of days from offence to completion (the time from an offence being committed through to an offender receiving an outcome) increased from 133 to 164 days. However, as can be seen from Figure 8, the part of the process that takes the longest, and is driving the overall delays in bringing offenders to justice, is the time taken from offence to charge. Offenders wait on average nearly four months to be charged: almost a doubling in the median wait time.

Figure 8

Average numbers of days from offence to completion, including specifically to charge, in 2019 versus 2011


In cases of serious sexual offences, the time taken from offence to charge is even longer. The inspectorate’s rape review pointed to lengthy delays in CPS advice being given on charging decisions, finding that prosecutors took an average of 209 days after the first file submission from the police, with the case often being sent back and forth between each agency, before a decision was finally reached.[_]

The delay in charging suspects has serious consequences for both offenders and victims. For offenders, delays are likely to reinforce the lack of association between the offending behaviour and the consequence, weakening deterrence. At the same time, the longer it takes to bring an offender to justice, the more likely victims are to disengage from the process, which in turn reduces the likelihood of a successful outcome.

Victim Withdrawal

It is notable that the proportion of crimes with “evidential difficulties” has grown significantly over the past five years. In particular, the proportion of crimes with an identified suspect, but which did not proceed to charge owing to victims withdrawing their support, has increased by more than 12 percentage points since 2015.

Figure 9

Proportion of crimes in England and Wales resulting in charges, summons or evidential-difficulties outcomes, 2015 to 2021


A 2020 review suggested that there was a greater likelihood of victim withdrawal in domestic abuse-related cases, which in some police force areas now represent around 40 per cent of all crimes. When asked in a 2019 survey at what point in the process victims were more likely to withdraw, police officers perceived withdrawal as most common pre-charge, either during the investigation or awaiting a charging decision.[_] By contrast, most participants perceived less victim withdrawal as cases progressed through the criminal-justice system, confirming that the more quickly the process began, the less likely a victim was to withdraw.

The Sentencing Process

Prosecutions and Effective Trials

Since 2009, prosecutions have fallen by 19 per cent. In 2019, 1.37 million defendants were prosecuted at magistrates’ courts — the lowest since records began in 1970.[_]

Successful prosecutions rely on trials administered effectively by Her Majesty’s Courts & Tribunals Service (HMCTS):

  • An effective trial is one that commences and reaches a conclusion on the scheduled dates.

  • An ineffective trial does not commence on the due date and will be rescheduled.

  • A cracked trial does not commence on the scheduled date and the trial is not rescheduled because it is no longer required. This is usually due to a late guilty plea being entered or the prosecution being dropped due to, for example, lack of evidence.

We can see from Figure 10 that the rate of effective trials has remained broadly flat over the past decade, falling only slightly since 2015. However, in 2021, around a fifth of trials (18 per cent) was categorised as “ineffective or cracked due to prosecution” and a similar proportion (22 per cent) resulted in a late guilty plea.

Figure 10

Proportion of all trials that were effective, ineffective or cracked, including the reason why, 2010 to 2021


Some legal experts have suggested that the recent increase in late guilty pleas may itself be an indirect result of longer delays. There is the notion among lawyers that defence teams may be advising suspects to withhold their plea for as long as possible on the basis that the longer the trial is delayed, the more likely it is to end up collapsing. In 2019, the top two reasons given for ineffective trials both in magistrates’ courts and in the Crown Court were “prosecution availability” (29 per cent) and “prosecution not ready” (25 per cent).[_]

Outstanding Cases (Court Backlogs)

The magistrates’ courts and Crown Court have been dealing with a substantial backlog of cases for many years. Unsurprisingly, these backlogs have become worse as a result of the early months of the pandemic when courts were forced to close during the first lockdown, although they had started to increase before this.

The backlog in the Crown Court is particularly striking: at the end of the first quarter of 2021, there were 59,532 outstanding cases, representing more than 50 per cent of the total caseload in the Crown Court – the highest level of outstanding cases since the statistics were first published. The average number of outstanding cases in relation to the length of delay has also increased. For example, the proportion of Crown Court cases that have been outstanding for a year or more has increased from 7 per cent in 2020 to 21 per cent in 2021 (see Figure 11).

Figure 11

Volume of outstanding Crown Court cases, grouped by length of delay, 2014 to 2021


The government clearly recognises the urgency of this situation, having made available almost £500 million over the next three years to address the backlog in the courts and to reduce waiting times.[_]

Offender Management

Breaches of suspended sentences and community orders are not consistently enforced. While statistics on breach outcomes are not routinely published by the Ministry of Justice, a response to a parliamentary question in 2018 revealed that for post-custodial, sentence-supervision breaches:

  • Just over a third (35 per cent) of those offenders were sent back to custody.

  • Nearly a quarter (24 per cent) had either an “invalid outcome'” recorded or no further action taken.

Figure 12

Outcome of post-custodial, sentence-supervision breaches in 2017


 There is a similar level of inconsistency applied to breaches of community orders. The most likely disposal for a breach in 2018 was continuing with the order.

Figure 13

Outcomes of community-order breaches, 2016 to 2018


This appears to contradict Sentencing Council for England and Wales guidelines, which state that “it is important that penalties for [these] breaches are consistent and sufficiently robust to avoid the objective of the sentence being undermined and to promote public confidence in these sentences.”

In summary, the criminal-justice system is, on nearly every available measure, failing to provide a meaningful deterrent to would-be criminals. The police are taking longer to respond to incidents; fewer suspects are being arrested and charged; victims are having to wait longer for investigations and for court cases to conclude; and even when offenders are caught, they are often able to breach their sentence conditions with impunity. With more money now set to come into the system, it is more important than ever to understand and address these failures to ensure value for money. That is what the next chapter sets out.

Chapter 5

Understanding the Deterioration in Performance

In this chapter, we examine some of the reasons behind the sharp deterioration in performance of the justice system, as set out in the previous chapter. They include:

  • Lack of resources

  • Changing crime mix

  • Growing complexity (failure to adapt to technological change)

  • Structural problems and misaligned incentives

  • Absence of a victim-centred approach (leading to victim withdrawal)

  • Policy changes with unintended consequences

  • Capability and skills shortfall

Lack of Resources

In examining the decline in the number of crimes solved over the past decade, an important consideration is the very significant reduction to policing and criminal-justice budgets, which has forced agencies to ration finite resources.

Between 2010 and 2019, spending on English and Welsh police forces declined and the number of police officers fell by more than 20,000, while the workload of the police (demand) demonstrably increased as a result of an increase in more serious crimes, such as sexual offences and serious violence. Over the period, government spending per person on policing fell by 27 per cent, while the number of recorded crimes per police officer increased by 50 per cent (Figure 14). The CPS’s net funding also fell by 33 per cent in real terms over the decade.

Figure 14

Number of recorded crimes per police officer in England and Wales, 2010 to 2018

Source: Home Office’s police recorded crime and police workforce statistics

In addition to workforce reductions, the closure of police-custody suites to release cash savings, particularly in London, may have had the unintended consequence of disincentivising arrests. This has meant that processing arrests often involves taking the detainee to a station based more than an hour away in another borough, where the officer is unlikely to have his or her case prioritised.

Following the Spending Review 2021, which has confirmed real-terms increases in policing and criminal-justice budgets over the rest of this parliament, one would expect these pressures to ease in future years. However, the onus will be on government to ensure that new money is accompanied by reform to deal with the systemic problems outlined in the previous chapter.

Changing Crime Mix

The squeeze on budgets has coincided with a period in which complex crimes, such as sexual offences and serious violence (which take longer to investigate and solve), have grown as a proportion of total police demand.

Figure 15

Proportion of complex crimes, including serious violence and sexual offences, as part of the overall crime mix, 2014 to 2023 (projected)

Source: Crest Poliscope crime forecasting analysis

While demand pressures are undoubtedly real, the deterioration in criminal-justice outcomes cannot be solely attributed to the growth in complex crimes: the charge/summons rate has fallen across the board, including for less serious offences, such as theft, which has dropped from 10.8 per cent to 5 per cent, and public-order offences, from 31.4 per cent to 8.4 per cent (Figure 1).

Growing Complexity (Failure to Adapt to Technological Change)

Both the availability and volume of data on digital devices have a practical impact on the majority of criminal investigations, increasing the resources necessary and length of time needed to review the material. In the HM Crown Prosecution Service Inspectorate’s Rape Inspection 2019, the findings of the report included “requests for forensic examination of phones taking up to 11 months to complete”.

As the volume of data held on devices continues to increase, police can potentially identify and obtain greater volumes of data in the course of an investigation. Once obtained, police need to review the data held as part of the investigation. Situations can quickly arise when there is more data held by the police than can realistically be viewed, affecting timescales for an investigation and for criminal-justice outcomes. Indeed, policing leaders have cited this as being one of the main changes and challenges experienced in adult rape cases in the past five years.

A review of digital evidence in appeal-case summaries in 2010 and 2018 showed a marked increase in the proportion of cases in which digital evidence was identified as being relevant to the court case, especially for sexual offences, including rape (Figure 16).

Figure 16

Percentage of appeal cases in which digital evidence was identified as relevant to the court case[_]




Total number of cases reviewed



% of all sexual offences



% of rape cases



Source: George and Ferguson, “Review into the CJS response to adult rape and serious sexual offences across England and Wales’, June 2021

In a government review of the police and CPS response to rape, the inspectorate reported that delays in case progression were often attributed both to a lack of appropriate technology, and training in accessing and analysing digital material. “The police digital units had seemingly not been resourced or equipped with the necessary IT to manage the change in use of digital media and the increased focus on digital material in disclosure which creates a backlog of material.[_] The increasing complexity and storage size of mobile phones added to this problem, with some police officers raising the concern that there were phone models the police were unable even to access.

“Everything else has moved on so fast and so quick we are so far behind it is embarrassing to sit here in 2019 and say that nobody has given us the tools. Give us the right capability, give us the right technology, we know what we need to investigate, we are literally still in 1989 or 1990.”[_]

Rape and Serious Sexual Offences (RASSO) Panel participant

Structural Problems and Misaligned Incentives

The relationship between the CPS and the police is a frequent source of friction. One accusation frequently levelled against the CPS by the police is that prosecutors regularly undercharge defendants in order to boost their conviction rates with easier cases to prove; for example, charging GBH or wounding as common assault. However, hard evidence of such practices is hard to come by.

A more substantial complaint about the CPS is that overly risk-averse prosecutors will often send case files back to the police for relatively trivial reasons, for example minor administrative issues, even if the file is substantively complete. There is some evidence to back this up: in their review of the police and CPS response to rape, the inspectorate found that just under half of the CPS’s action plans were deemed unsatisfactory. Problems with the standard-of-action plans included requests to examine digital devices when they were not needed and failing to set meaningful parameters when some examination was required.[_]

Often this tension spills over into the public domain. In response to criticism in the media about low charge rates for thieves, the chairman of the Police Federation implied the blame lay elsewhere: “Charging decisions for the majority of crimes are generally made by the Crown Prosecution Service (CPS), not the police. What will drive these decisions is evidence: sometimes we simply don’t have enough to satisfy the CPS which is not only frustrating for victims but also for those police officers investigating the crimes.”

The CPS has its complaints about the police, too. Many CPS lawyers complain about the poor quality of files they receive from the police. There are complaints that the police will often hide or exploit evidence in order to “twirl” the CPS into laying a charge they would not bring if they had all the evidence. There are also complaints that the police are concerned only with securing a charge, rather than whether the defendant is convicted or not, meaning that the CPS often finds the relevant police officers are unavailable during the pre-trial stage.

Absence of a Victim-Centred Approach

As discussed in the previous chapter, there has been a rise in victim withdrawal, which is hampering the progress of investigations. Without the support of the victim, it is very difficult to meet the evidential threshold needed to proceed with a case. There are several factors to explain this.

First, there is evidence that the increasing length of the charging process is leading to victims withdrawing, since the delay prevents victims from being able to move forward with their lives. For example, victims’ advocates have discussed how lengthy delays may prolong the trauma for victims, meaning some have to withdraw in order to seek support.

“An investigation that goes on for years has a detrimental effect on wellbeing and I think that comes into play at some point… it stops, and they’ve got to look after themselves before they can even think about continuing through the criminal-justice system.”[_]

Independent Sexual Violence Advisor, 2020

Independent Sexual Violence Advisors have also raised questions about the combined impact of the removal of bail protections (meaning the rise in people being released while under investigation) and the process taking longer, which can leave victims at greater risk of harassment or intimidation by perpetrators and thereby encourage victim withdrawal.

Second, there is a perception that cases remain overly focused on the victim’s credibility rather than that of the suspect. This is particularly the case for rape cases, where the intrusive nature of investigations and anxiety around being disbelieved can lead to victims deciding to withdraw.

“We do a lot more around the complainant, it’s all around her credibility, how credible the victim is, but that’s never replicated for the suspect.”

Police Investigator

Third, the historically poor relationship between police and victims has been cited as a central factor in victim withdrawal.[_] There have long been concerns raised by victims’ groups that the police can be insensitive in their approach and may use poor judgement in their interactions with victims. In recent years, victim-liaison services – and wider victim support – have been cut back due to lack of resources, which may have left victims feeling abandoned and unwilling to participate in the process. Research with rape victims suggests that at least some of those who initially supported a prosecution, but subsequently withdrew from the process, would probably have been willing or able to continue with the case if they had been provided with better specialist support.

More research is required to understand the phenomenon of victim withdrawal. The current system of recording cases, where the police have made the decision to take no further action, fails to identify at what point the victim withdraws support. This is a missed opportunity to gather and use data in a focused way and provide tailored support to victims.

Policy Changes With Unintended Consequences

In recent years, there have been several attempts to streamline and improve the front end of the criminal-justice process, yet the siloed application of these reforms has often had unintended consequences.

Changes to Pre-Charge Bail

Police bail places conditions on individuals to restrict where they go and with whom they interact (for example, requiring them to live at a particular address or give up their passport to the police). The aim is to protect victims and witnesses, and reduce the risk of further offending. Prior to 2017, concerns with the bail system were that suspects were subject to stringent conditions for long periods without being charged, and that the system was leading to lengthy delays and lethargic investigations.

Provisions in the Police and Crime Act 2017 restricted bail to a maximum of 28 days and also set out a presumption against the use of police bail “in almost all cases”. As a result of those changes, most suspects are now released under investigation (RUI), rather than bailed. However, in contrast to the bail system, RUI places virtually no conditions on suspects. The police have no powers to require RUI suspects to report to them or to place conditions on them, and no time limits within which to complete their investigation. This is potentially compromising the safety of the public, including victims, who may be targeted again by the same perpetrator.

Many police forces do not collect RUI data, so it is difficult to gain a proper understanding of its scale. However, data released under the Freedom of Information Act in 2019 suggest that the use of RUI has increased significantly and is now being used by police forces as an alternative to bail.[_] Indeed, research undertaken by the Law Society suggests that since the use of RUI started, the average length of investigations has increased, leaving suspects and victims in limbo for longer and contributing to longer delays.

Figure 17

Average length of time on pre-charge bail versus release under investigation across 10 police forces, 2016/2017 and 2018/2019

Source: The Law Society (2018)’s Release Under Investigation Data (freedom of information request)

Changes to the Threshold on Arrests

In 2005, the Serious Organised Crime and Police Act introduced “necessity criteria” for arrests, requiring officers to prove why an arrest is needed in each circumstance. The criteria were further updated in 2012 (“Code G”), with officers required to consider in each case whether voluntary attendance at a police station for an interview under caution may be more suitable than an arrest. Anecdotally, we have heard from police officers that this has raised the threshold for the necessity of arrest, meaning, for example, that custody sergeants are applying it more rigorously than before, although there is no hard data to support this. We recommend the government carries out further research on this subject as a matter of urgency.

Capability and Skills Shortfall

Another possible explanation for the fall in charge rates is a lack of experienced and trained personnel, meaning the quality and speed of investigations has declined.

Analysis of police-demand data reveals that a growing proportion of demand is being taken up by investigations of more serious and complex crimes. Forces have to make decisions about whether and how to investigate a crime, and put the right investigators (police detectives, police-staff investigators, intelligence analysts and forensic-support staff) on the right cases. However, in recent years, police forces across England and Wales have faced significant shortages in trained investigators, including approximately 5,000 vacant detective roles.  HMICFRS has described the detective shortage as a “national crisis”.[_]

A recurring issue affecting the standards of investigation is the quality of supervision. Poor or inconsistent supervision means that too many cases do not make the progress they should. Good supervision develops investigators’ skills and makes sure that workloads are manageable. A 2017 inspectorate report[_] found that lack of supervision was particularly severe in so-called low-level or volume crimes (for example, theft from the person or common assault), with those investigations more likely to be allocated to less experienced, non-specialist investigators. These will often be response officers who have an investigative caseload in addition to many other responsibilities.

There are several reasons why the police service has struggled to keep pace with the increase in demand for investigative capacity, which largely relate to the structure of the workforce:

  • Officers are increasingly reluctant to specialise as detectives because of high workloads, a lack of experienced mentors to support them during training, unsociable working hours and lack of reward with pay increments, and the intense scrutiny they come under if there is some form of investigative failure.

  • It takes time to gain investigative skills and experience. To become qualified as a detective takes several years, and involves examinations and work-based assessment. By contrast, other career opportunities in the police service may seem more attractive.

  • It is difficult to retain detectives. Their skills are becoming increasingly attractive to the commercial sector. For example, highly trained financial investigators have been lost to banks and other financial institutions, which offer greater personal rewards.

  • To balance budgets, some forces have cut back on programmes to introduce more police-staff investigators. This has increased the pressure on detectives.

Chapter 6


The Spending Review 2021 has signalled the government’s commitment to tackling the justice gap, providing for real-terms increase in policing and criminal-justice budgets. It is critical that this funding is directed to the problems identified in this report, specifically the falling quality and speed of investigations. With new funding should come an expectation of reform. We recommend that government implements three reforms immediately.

New Minimum Standards

The criminal-justice policy debate remains overly focused on sentence severity rather than the certainty of being caught, and there is a lack of clarity about what success looks like with regards to the charging and prosecution of offenders.

As part of a commitment to drive up speed and certainty within the system, the Home Office and Ministry of Justice should consult on the creation of new national standards that set out what the public should expect from the police and CPS when they respond to crimes. These could include a commitment to:

  • Improving response times and the police’s ability to exploit the “golden hour” (first hour) of evidence windows.

  • Reduce the time taken from offence to charge, which will increase the likelihood of keeping victims engaged in the process.

  • Increase the proportion of serious crimes, such as violence and rape, that are charged and prosecuted.

National Taskforce on Use of Technology in Investigations

One of the biggest barriers to bringing offenders to justice in a timely fashion is the growing complexity of the charging process. In particular, the volume of material that the police are required to extract from mobile phones is proving to be an intolerable burden for many forces. The potential for new technology, such as the use of AI, to streamline the process is significant yet the police remain ill-equipped to maximise such opportunities.

At the same time, many forces are facing a shortage of experienced and trained investigators, resulting in decreased standards of initial investigation and worsening file quality.

The government should appoint a national-investigations taskforce with the sole purpose of driving up the quality and speed of investigations. The taskforce would seek to achieve this in two ways:

  • Accelerate partnerships between government, technology companies and universities, focused on designing technological solutions to reduce the complexity of charging and improve timeliness.

  • Review the College of Policing’s approach to recruitment, supervision and training of specialist investigators in order to identify any gaps and recommend reforms.

Reforming Accountability Structures

The front end of the justice system suffers as a result of outcomes that are not in step. On the one hand, the CPS blames the police for not being focused enough on the likelihood of conviction, resulting in case files being returned on the basis of insufficient file quality. On the other hand, the police suspect that the CPS is setting overly high thresholds for charges, by weeding cases out as early as possible rather than taking them through to court, since they are incentivised to maintain conviction ratios rather than charge rates.

Government should commit to reforming accountability structures, for example by making the police and CPS jointly accountable for both charge and conviction rates. To begin with, this could be piloted in a single force area, with seed funding to facilitate greater co-location of staff to improve police-prosecutor relationships.

Government should also ask Her Majesty’s Inspectorate to conduct a review of police forces’ decision-making around arrests (as opposed to voluntary interviews), and the potential impact on evidence-gathering and early guilty pleas.

All three of these reforms would make a difference, but they need to be seen as part of a bigger effort to fundamentally rebalance the system. For over half a century, the justice system has been driven more by a focus on the rights of those accused (and the need to avoid egregious miscarriages of justice) than by the needs of victims. As a result, we are now in a world where the complexity of the criminal-justice process makes successful prosecution very tough. Central government needs to use every lever at its disposal to remedy this problem, and deploy a laser-like focus on improving the quality and speed of investigations. Nothing less than the legitimacy of our justice system is at stake.

Lead Image: Getty Images

Charts created with Highcharts unless otherwise credited.


  1. 1.

  2. 2.

    Nagin and Pogarsky, “Integrating Celerity, Impulsivity, and Extralegal Sanction Threats into a Model of General Deterrence: Theory and Evidence,”


    , 39(4), 2001.

  3. 3.

  4. 4.

    Weisburd et al, 2010.

  5. 5.

    Pogarsky et al, 2005, Earnhart and Friesen, 2014.

  6. 6.


  7. 7.

    Bradford, 2011, ‘Police Numbers and Crime Rates – A Rapid Evidence Review’, HMIC.

  8. 8.

    Sherman and Eck, 2002.

  9. 9.

    Ratcliffe et al, 2020.

  10. 10.

    Motz, “Does contact with the justice system deter or promote future delinquency?” Dec 2019,


  11. 11.

  12. 12.


  13. 13.

  14. 14.

  15. 15.

    Cost and Demand of Mental Health to the Metropolitan Police Service and the Public, Metropolitan Police Service, 2018

  16. 16.

  17. 17.


  18. 18.

    Jordi Blanes i Vidal and Tom Kirchmaier, “The effect of police response time on crime clearance rates,” 2017,

  19. 19.

    This may partly be a function of the fact that the “golden hour” for evidence collection is the first hour after a crime incident has occurred.

  20. 20.

  21. 21.

    For example, see the section on the rape review within the Victims’ Commissioner’s annual report, 2020-21,

  22. 22.

    Justice Inspectorate, Joint Thematic Response to CPS and Police Response to Rape, 2021.

  23. 23.


  24. 24.

  25. 25.

    Court statistics for England and Wales, House of Commons Briefing Paper, Dec 2020,

  26. 26.

    Autumn Budget and Spending Review 2021

  27. 27.

    Richardson et al, forthcoming, Using appeal case summaries to assess changes in the presence of digital evidence in police investigations Home Office publication.

  28. 28.

    George and Ferguson, “Review into the CJS response to adult rape and serious sexual offences across England and Wales,” June 2021,

  29. 29.


  30. 30.

  31. 31.

  32. 32.

    Hohl and Stanko, 2015

  33. 33.

    The Law Society (2019),

    Release under investigation

  34. 34.

    Peel Police Effectiveness 2017, A National Overview,

  35. 35.


Article Tags


Practical Solutions
Radical Ideas
Practical Solutions
Radical Ideas
Practical Solutions
Radical Ideas
Practical Solutions
Radical Ideas
Radical Ideas
Practical Solutions
Radical Ideas
Practical Solutions
Radical Ideas
Practical Solutions
Radical Ideas
Practical Solutions