Abstract

There is little understanding of how documented ethnic disparities in sentencing outcomes in England and Wales come to be and, consequently, how to address them. We argue that ethnic disparities in sentence outcomes could stem from how cases are constructed. We hypothesize that case characteristics determined through a high degree of judicial discretion and an assessment of the offender have a high risk of being racially determined, and therefore, operate as precursors of ethnic disparities in sentencing. We test this using data from the Crown Court Sentencing Survey combined with data on offenders’ ethnicity from the Ministry of Justice Court Proceedings database. We identify three sentencing factors (remorse, good character, and ability to rehabilitate) clearly favouring White offenders. We contextualize their operation and argue that all three should be classified as racially determined and sources of ethnic disparity. We conclude by setting out targeted policy solutions.

Introduction

A number of studies have now shown the presence of ethnic disparities in sentencing outcomes in England and Wales (Hopkins, 2015; Hopkins et al., 2016; Isaac, 2020). Ethnic disparities in sentencing can undermine the fundamental principle of equality under the law and damage trust and confidence in both the sentencing process and the criminal justice system more broadly (Tyler, 2001; Tyler, 2003; Hough et al 2013). Cognizant of its implications, the Lammy Review (2017, 7) proposed a principle for observed ethnic disparities in the criminal justice system, “explain or reform”; i.e. if an evidence-based explanation for apparent disparities between ethnic groups cannot be provided, then reforms should be introduced to address those disparities.

Despite this, in terms of sentencing, neither explain nor reform have been pursued sufficiently to date. We believe this is, in a large part, due to the lack of understanding as to where disparities are stemming from. Without knowing the sources of disparities, it is difficult to explain them and similarly difficult to develop targeted and effective policies to address them. The vast majority of the literature on sentencing disparities focuses solely on sentencing outcomes and as a consequence it neglects other key judicial decisions that impact the final sentence imposed, resulting in a gap in understanding as to sources of disparities. In this article, we seek to overcome this current impasse by changing the focus from outcomes to inputs and in doing so identify potential sources of disparities stemming from a key part of the sentencing process: the judicial construction of a case.

A large part of a criminal case will be ‘constructed’ prior to a sentencing hearing and will be presented to the judge by the prosecution at the outset of the hearing. Furthermore, for most offences, there will be a sentencing guideline which judges must follow in determining what sentence to impose, which might provide an impression of limited judicial autonomy. However, a judge’s role is not simply to apply a sentencing guideline to the case presented to them in order to determine a final sentence. Judges too have a role in the construction of a case. Judges decide whether or not a case characteristic/sentencing factor is considered relevant and taken into account in determining the final sentence (Ashworth and Kelly, 2021). For some factors, judges will have little discretion (e.g. whether the offence was committed while on bail is an objective reality) but for others, judges will have wide discretion (e.g. whether an offender is genuinely remorseful represents a subjective judgement). The exercise of this discretion in deciding whether or not to accept and take into account the sentencing factor in any given case will impact the final sentence imposed.

A further division can be made between sentencing factors in terms of whether they relate to the offence or the offender. Offence related factors involve an assessment of the criminal act or the circumstances in which it took place (e.g. committed against those working in the public sector or the presence of children). Offender related factors involve an assessment of the individual being sentenced (e.g. whether an offender is of good character). We believe these two criteria (high judicial discretion and an assessment of the individual) are key in exploring potential ethnic disparities stemming from the sentencing stage. By focusing on highly discretionary factors it will allow us to hone in on key judicial decisions at the sentencing stage. Within these highly discretionary sentencing factors, we hypothesis that it is particularly those that relate to offender (as oppose to offence) characteristics, that have the greatest risk of being racially determined. By this (‘racially determined’) we mean that the offender’s ethnicity can play a role in judges’ decisions whether or not to consider such factors as constitutive of the case to be sentenced (Omori and Petersen, 2020; Ugwudike, 2020; Pina-Sanchez et al, 2022). As a result, there is a high likelihood that factors satisfying both criteria can operate as sources of ethnic disparities in the construction of criminal cases, and in so doing explain some of the disparities in sentence outcomes reported in the literature.

To test our hypothesis, we first identify and select sentencing factors involving both a high degree of judicial discretion and an assessment of the individual being sentenced. We then use survey data from the Sentencing Council combined with administrative data from the Ministry of Justice capturing offenders’ ethnicity, to determine whether and to what extent the suspected factors are unevenly distributed across ethnic groups. We do this for two high volume offence groups: assault and public order offences, and drugs offences. For each of the factors identified as having a substantive disparity in their use in favour of the White group, we proceed to delve into how these factors are used in practice, so we can assess whether they should be classified as racially determined and a likely source of ethnic disparity in sentencing outcomes.

In the next section we will identify those factors that could be more susceptible to being racially defined. We will then present our analysis, based on a Freedom of Information request, showing whether or not the suspected factors were indeed less uniformly distributed across ethnic groups. This is followed by a review of the guidance and existing literature on how the factors are assessed by judges. We conclude by discussing of our findings and presenting a range of potential reform options to address ethnic disparities stemming from the sentencing stage.

Categorising Sentencing Factors

Table 1 lists the step one and step two sentencing factors contained in the Crown Court Sentencing Survey for assault and public order offences (hereafter referred to as assault offences), and the step two sentencing factors for drug offences. The data we received from the Sentencing Council did not include a detailed breakdown of individual step one sentencing factors for drugs offences. It only covered number and proportion of cases in each category for harm and culpability. Therefore, we restrict our analysis of drug offences to step two sentencing factors.

Step one sentencing factors comprise the principal factual elements of the offence. They assess the offender’s culpability and the harm caused, or intended, by the offence. When sentencing, a judge will first assess step one factors and place the case at a starting point and within a sentence/category range. Step two factors are aggravating and mitigating factors relating to the context of the offence, and to the offender. Each guideline provides a non-exhaustive list of these factors. Judges will identify whether any of the listed factors, or any other relevant factors, are present and then make an upward or downward adjustment from the starting point.

To gain the necessary analytical focus we only select sentencing factors which involve a high degree of judicial discretion (subjective), and can only be applied based solely on an assessment of the individual. Five factors are deemed to satisfy both criteria.

• Remorse: This is probably the most subjective sentencing factor. Judges have wide discretion in how remorse is assessed and ultimately in deciding whether or not it is accepted as a mitigating factor (Ashworth and Kelly, 2021; Field and Tata, 2023). Identifying genuine remorse is purely a personal assessment of whether or not the individual being sentenced is remorseful for the crime committed.

• Good character: The Court of Appeal (see R v Howells 1999) and the sentencing guidelines expanded explanation (Sentencing Council, 2021) offer some guidance on how to define a ‘good character’ offender (see section ‘Racially Determined Factors?’ below). The guidance, however, is somewhat vague and judges are essentially left with significant discretion in deciding what acts demonstrate good character. The assessment is of the individual and their past actions.

• Determination or demonstration of steps taken to address addiction/offending behaviour (hereafter referred to as ability to rehabilitate): When assessing this factor, judges have to decide what acts are sufficient to demonstrate steps taken to address addiction/offending behaviour. Judges also retain significant discretion in deciding whether to accept a person’s commitment to address their addiction/offending behaviour as sufficient for the mitigating factor to be accepted and applied (Sentencing Council, 2021). The assessment a judge has to make is whether the individual being sentenced can and will likely address the root causes of their offending and as a consequence, avoid committing further offences.

• Mental disorders/learning difficulties: When sentencing a person with a mental disorder or learning difficulty, judges are required to follow the ‘mental disorders, developmental disorders, or neurological impairments’ sentencing guideline (Sentencing Council, 2021). While the guideline is detailed and sets out criteria judges should consider, mental disorder is a complex factor that can present in many forms and degrees. Therefore, the guideline leaves a considerable degree of discretion to judges when it comes to deciding whether a person’s sentence should be mitigated due to the presence of a mental disorder or learning difficulty. We deemed the factor to have met the criteria of ‘an assessment of the individual’ because the factor can apply even where the mental disorder/learning difficulty is not linked to the offence. In those circumstances, the Court of Appeal (see R v PS 2020) stated a judge should assess the individual’s mental health at the time of sentence and determine whether the individual will experience a sentence more severely due to the mental disorder/learning difficulty.

• Age/lack of maturity: The assessment of age/lack of maturity becomes subjective when sentencing a person in their late teen or early twenties. Judges have to consider and assess as person’s maturity. The factor can apply if a judge considers the lack of maturity to have affected the persons responsibility for the offence. However, it can also be applied independent of the offence committed, and solely on the basis that the person’s age/lack of maturity will give rise to additional challenges and cause a sentence to be experienced more severely (Sentencing Council, 2021).

In assessing the above factors, a judge will often have a pre-sentence report from a probation officer and/or an expert report from a medical practitioner to assist them. However, these reports are only advisory and the ultimate decision of whether the factors are accepted and taken into account rest with the sentencing judge.

Table 1: sentencing factors for assault and drug offences categorised as both subjective and involving a judicial assessment of the individual (in bold)

Distribution of Sentencing Factors across Ethnic Groups

In order to analyse the distribution of sentencing factors across ethnic groups, we obtained data from the Sentencing Council through a freedom of information request. It combines Crown Court Sentencing Survey data1 with data on ethnicity from the Ministry of Justice Court Proceedings Database2. The ethnicity variable used is the offender’s ethnicity as perceived by the police officer dealing with the case. This is preferred to self-identified ethnicity data, as a police officer’s perception of ethnicity is a more accurate measure of judicial perceptions of ethnicity. The ethnicity categories used in the data are: White, Black, Asian and Other. The data covers the period 1st January 2013 to 31st March 2015 and includes adult offenders only.

The data we received set out the frequency each factor was ticked by a sentencing judge in the Crown Court Sentencing Survey, broken down by ethnic group (See Appendix 1). T-test are carried out in order to identify whether there is a statistically significant difference in the prevalence of any of the selected factors. We also consider the direction and extent of the group difference. Where this group difference is statistically significant, bigger than two percentage points (i.e. substantially significant in absolute terms), and bigger than ten percent (i.e. substantially significant in relative terms), we take it as evidence of a substantive disparity.

Table 2 shows there is evidence of substantive disparities between the White group and at least one ethnic minority group for four out of five of the selected factors. ‘Mental disorder/learning difficulty’ is the only factor not flagged as showing a substantive disparity in its use between ethnic groups.

Table 2: Significant disparities in prevalence of sentencing factor across ethnic groups

Assault Offences

Drug Offences

Most, but not all, of the disparities identified as substantive, favour the White group and disadvantage an ethnic minority group. ‘Age/lack of maturity’, was the only exception. This personal mitigating factor was for example present in 13% of drug cases in the Black group, but only 7% in the White group. For ‘good character’ we find substantive disparities that both favour and disadvantage the White group. This mitigating factor is more present in the White group (16%) than the Black group (11%) for drug offences. However, it was only present in 14% of assault cases in the White group, with the Asian group and Other group reaching 22% and 20% respectively.

For ‘remorse’ and ‘ability to rehabilitate’ all of the disparities identified favour the White group and disadvantage an ethnic minority group. For ‘remorse’, there is evidence of substantive disparity in favour of the White group against the Black group for assault offences (34% of cases in White group and 29% in Black group) and against Black, Asian and Other for drug offences (28% of cases in White group, 19% in Black group, 21% in Asian group and 19% in Other group). For ‘ability to rehabilitate’, there is evidence of substantive disparity in favour of the White group against all other ethnic minority groups for both assault offences (9% of cases in White group, 6% in Black group, 4% in Asian group and 5% in Other group) and drug offences (15% in White group, 9% in Black, 9% in Asian group and 7% in Other group).

In addition to analysing the distribution of selected factors (those with a high degree of judicial discretion and involving an assessment of the individual being sentenced), we also conducted the same analysis on all other sentencing factors listed but not selected in Table 1, which we expected to be less prone to become racially determined. Findings show that the ethnic distribution of these other factors that were not selected is far more balanced than the selected factors. Specifically, out of 70 factors that were not selected in our analysis, 27 (39% of the total) show a substantive disparity between the White group and at least one ethnic minority group (these are listed in Table 3), but only 16 of those (23% of the total) favour the White group.

Table 3: Other sentencing factors showing substantive disparity between the White group and an ethnic minority group (listed) and the ethnic group the disparity favours.

The difference in ethnic disparities in the use of selected and non-selected factors is therefore quite clear. Whereas, we find that 60% of selected factors show evidence of disparities favouring White offenders, that is only the case for 23% of non-selected factors. That difference in ethnic disparities across groups of factors seems too large to be explained by mere chance, and supports our general hypothesis that selected factors are at a high risk of being racially determined.

Racially Determined Factors?

Having found empirical evidence showing three of the high discretion and individual assessment factors clearly favouring the White group, we now move on to consider how these factors operate in practice. We rely on Court of Appeal and Sentencing Guideline guidance, as well as on past research exploring how these factors are applied, to help us determine whether they should be classified as racially determined factors, and as such a likely source of unjustified ethnic disparities in sentencing outcomes.

Remorse

Remorse is one of the most commonly used sentencing factors. The data we obtained from the Crown Court Sentencing Survey shows remorse to be the most used mitigating factor for assault and drug offences, cited in 33% and 25% of cases respectively. There are substantial variances in the prevalence of the factor across ethnic groups. For drugs offences, it was cited in 28% of cases in the White group but only in 19% of cases in the Black group. Remorse has also been shown to have a significant impact on sentencing outcomes (Maslen, 2015), with the odds of receiving a custodial sentence more than halved when remorse is taken as a defining characteristic of the case (Pina-Sánchez et al, 2018).

Remorse is a highly subjective factor. Judges have wide discretion in how it is assessed and ultimately in deciding whether or not to accept it as a mitigating factor in any given case. International research on how judges assess remorse, have highlighted an offender’s demeanour and performance in court are central to the assessment. Weisman (2009) reviewed 178 Canadian cases in which remorse was a matter of dispute. The author suggest that public enactments of offender remorse are ‘moral performances’ and argued that ‘the offender’s ‘body’ is unequivocally central to the judiciary’s assessment of remorse’ (2009, 49). Johansen’s (2019) study in Danish courtrooms consisting of field notes from 80 criminal cases as well as interviews with 20 of the presiding judges emphasised the important role behaviour and appearance plays in judges’ assessments of emotions such as remorse. Rossmanith’s (2015) research, which involved interviews with 18 Judges in New South Wales, reported that judges will often use ‘visual clues’ from a person’s performance when assessing remorse. Various judges referred to the offender’s demeanour and the judge’s capacity to read the external, physical signs when assessing the genuineness of remorse. The research also highlighted that in some cases remorse assessment can operate in experiential terms, as a feeling. ‘Something gets felt by judges at the level of embodied affect that then enables them to declare: ‘This person is remorseful.’ (2015, 171).

If, as the literature suggests, an offender’s appearance, demeanour and performance in court is central to the assessment of remorse, then this raises concerns when the assessment is cross-racial. Bennett (2016, Sec. 5) argues: ‘different cultures may have different rules about who can display remorse, and how, and indeed what appropriate remorse looks like when it is displayed. We need to be open to the possibility that what may look like a lack of remorse to one observer may look different to someone attuned to the norms of the culture from which the person being observed comes’.

This is supported by Johansen’s study (2019, 250), who concluded that ‘Judges’ assessments of emotions are mediated through their own cultural understandings, and what counts as ‘appropriate’ emotion is dependent on how the defendant is culturally and systemically situated’. For example, judges appeared to place significant weight on the defendant looking the judge/legal actor in the eye when being questioned, however, some defendants, particularly those from ethnic minority backgrounds, did not do so but instead looked down. Johansen (2019) suggests that this was done as a sign of respect and is linked to specific cultural expectations and upbringing. Challenges with cross-racial assessment are also evident in Rossmanith’s (2015) study where many of the judges acknowledged that cultural difference makes it especially hard to read how other people are feeling and therefore risks a person’s actions or demeanour being misinterpreted. A study by Veiga et al (2023) involving interviews with criminal barristers in England and Wales noted similar concerns.

As at April 2022, just 5% of judges in England and Wales were from Asian backgrounds, 1% were from Black backgrounds, 2% were from mixed ethnic backgrounds and 1% were from other ethnic minority backgrounds (Ministry of Justice, 2022). The highly subjective nature of remorse, the concerns raised above about cross-racial assessment of remorse and the high likelihood of ethnic minority offenders being sentenced by White judges in England and Wales, provides a strong basis for identifying remorse as a racially determined factor.

A recent study by Proeve and Tanvir (2022) which involved the coding of 262 sentencing remarks from the higher courts in South Australia using content analysis, to examine the frequency in which judges mentioned various pieces of evidence for remorse, found a strong link between guilty pleas and remorse being accepted and applied as a mitigating factor. It is reasonable to assume a similarly strong link in England and Wales. This would indicate that remorse may also be indirectly racially determined through guilty pleas, as data from England and Wales consistently shows lower rates of guilty pleas amongst ethnic minorities compared to Whites (Lymperopoulou, 2022; Ministry of Justice, 2021, Sec.5.6).

When we add the evidence that remorse is unevenly distributed across ethnic groups in favour of the White group and that it has a significant impact on sentencing outcomes, one can reasonably flag remorse as a likely source of ethnic disparities in sentencing outcomes.

Good Character

Good character is another of the most commonly raised mitigating factors. In the data we obtained from the Crown Court Sentencing Survey, good character was cited as a mitigating factor in 15% of both Assault and Drugs cases, though its prevalence varied across ethnic group. For drugs offences, it was cited as a mitigating factor in 16% of cases in the White group but only 11% of cases in the Black group. Good character has also been shown to have a significant mitigating impact on sentencing outcomes. The sentencing Council (2015, 30) found that in the Crown Court in 2014 domestic burglary cases involving good character were associated with a custody rate of 27%, compared to the average of 77%, and an average custodial sentence length that was one year and one month shorter than the overall average for the offence.

Good character is an elusive and vaguely defined concept (Belton, 2018). It is often used to connote merely an absence of previous conviction (Hungerford-Welsh, 2019). However, in ‘R v Howells’ (1999) the Court of Appeal stated that: ‘a measure of leniency will ordinarily be extended to offenders of previous good character, the more so if there is evidence of good character (such as a solid employment record or faithful discharge of family duties) as opposed to a mere absence of previous convictions’. Here, the Court of Appeal is defining good character in a way that covers more than just a lack of previous convictions, highlighting examples of employment and performance of family duties as evidence of good character. Ashworth and Kelly (2021, 162), discuss good character primarily in the context of one-off good deeds referencing cases such as Reid (1982) and Wenman (2015) where sentences were reduced as a result of acts of bravery unrelated to the offence being sentenced. The latest extended explanation attached to the mitigating factor in sentencing guidelines (Sentencing Council, 2021) state: ‘This factor may apply whether or not the offender has previous convictions. Evidence that an offender has demonstrated positive good character through, for example, charitable works may reduce the sentence’. In sum, when assessing good character, the focus is primarily on whether the person being sentenced has demonstrated good character through positive acts. What counts as a positive act sufficient to demonstrate good character can be wide and varied. Judges are essentially left with wide discretion in deciding what acts demonstrate good character, whether it should be included as a mitigating factor in any given case and the extent of the mitigation that should be given.

Belton’s review of judges’ sentencing reasons relating to good character (Belton, 2017 cited in Belton 2018, 50) and his study involving interviews with six judges in England Wales (Belton 2018, 50) found: ‘that the everyday “bread and butter” of good character is character testimonials from family, friends, employers, community leaders, and so on. These testimonials deal with a broad range of character-related issues including public service but also matters such as trustworthiness and reliability as a worker or partner, caring responsibilities, and status within the community.’

Saccomano (2019) argues white cultural values are deeply embedded in many of the character-related issues considered and assessed as part of the process of weighing ‘character’.3 She points in particular to issues such as offender’s educational attainment, employment record/status, and family stability. Data in the UK shows lower levels of employment, education attainment, family stability amongst ethnic minorities (Social Metrics Commission, 2020; Centre for Social Justice, 2020; Equality and Human Rights Commission, 2016). One could also point to issues such as public service or charitable work which have clear links with a person’s wealth and financial security. A person who is financially secure is in a far better position to engage in public service or charity work than a person living in poverty, whose primary focus will be on ensuring themselves and their family have the basics needed to survive. In the UK, people from ethnic minorities are twice as likely to live in poverty compared to white British people (Edmiston, 2022; Social Metrics Commission, 2020; Equality and Human Rights Commission, 2016).

The subjective nature of ‘good character’ assessment combined with the argument that many character-related issues often considered as part of the process of assessment stem from the dominant cultural values and have a higher likelihood of being present in white British offenders, provides an evidentiary basis for identifying good character as a racially determined factor. Add to this the stark uneven distribution in its use across ethnic groups in favour of the White group, together with the significant mitigating impact it has on sentencing outcomes, and one can reasonably flag good character as a likely source of ethnic disparities in sentencing outcomes.

Ability to rehabilitate

Ability to rehabilitate is yet another commonly raised and impactful mitigating factor. In the data we obtained from the Crown Court Sentencing Survey it shows it was cited in 8% of assault cases and 13% of drugs cases. There are substantial variances in its use across ethnic groups, with drugs offences showing the largest variance. It was cited as a mitigating factor in 15% of cases in the White group but only in 9% of cases in the Black and Asian group. Both sentencers and defence lawyers have pinpointed it as a highly influential factor (Chen 2023) and this is corroborated by a quantitative studies based on the Crown Court Sentencing Survey (Lightowlers and Pina-Sánchez, 2017; Sentencing Council, 2015). For example, Lightowlers and Pina-Sánchez (2017) show a four-fold reduction in the odds of receiving a custodial sentence amongst assault offenders who were deemed to have this mitigating factor.

The current guidance for judges on this factor contained in sentencing guidelines (Sentencing Council 2021, no pagination) states: “Where offending is driven by or closely associated with drug or alcohol abuse (for example stealing to feed a habit, or committing acts of disorder or violence whilst drunk) a commitment to address the underlying issue may justify a reduction in sentence. This will be particularly relevant where the court is considering whether to impose a sentence that focuses on rehabilitation. Similarly, a commitment to address other underlying issues that may influence the offender’s behaviour may justify the imposition of a sentence that focusses on rehabilitation. The court will be assisted by a PSR [pre-sentence report] in making this assessment”.

The guidance suggests that a commitment to address addiction/offending behaviour is sufficient in order for this mitigating factor to be applied. This leaves judges with wide discretion when considering whether to accept this factor. However, the Court of Appeal in ‘R v Howells’ (1999) previously indicated that where steps have been taken to demonstrate this commitment, sentencing courts will be more inclined to accept it and give greater weight to it. So, while this factor can be applied based on a person’s commitment alone, if evidence can be provided to support it, then the likelihood of it being accepted and the weight given to it will increase. The judges interviewed as part of Belton’s study (2018, 151) appeared very much to approach the assessment of the factor in line with the guidance set out in ‘R v Howells’ (1999).

One of the most significant and impactful steps a person can take would be to begin to address their risk factors/criminogenic needs, particularly those identified in the pre-sentence report. Where a person’s offending is associated with drug or alcohol abuse this would be to begin, and ideally to be doing well in, drug or alcohol treatment. In Belton’s study (2018,151) this appeared to be the key issue judges focused in on when assessing and weighing up this factor.

One of the ways in which this factor might be racially determined is due to the additional barriers that exist for ethnic minorities compared to white British in accessing drug or alcohol treatment. These barriers have been highlighted in a number of studies and reports as well as by those working in addiction treatment in England and Wales. Ismail and Williams (2012, 19) when studying drug use amongst BME groups in Bristol noted: ‘there were many barriers to access drug services by the BME communities. These were divided into real barriers (such as waiting time, lack of information and advice about illicit drugs and lack of trust in the service providers) and perceived barriers (stigma surrounding access to drug service and fear that disclosing drug use would negatively affect immigration status).’

The existence of these additional barriers for ethnic minority groups were also outlined in a series of reports, each of which focused on a specific ethnic minority group, published by the National Treatment Agency for Substance Misuse (see Fountain, 2009). Many of these same barriers have also been shown to exist for alcohol treatment (Gleeson et al, 2029). Sohan Sahota, the managing director of the drug and alcohol recovery support service BAC-IN, notes in particular, language barriers, stigma and cultural differences, mistrust of mainstream services, and experiences of racial prejudice, can all prevent ethnic minorities from accessing and engaging with treatment services (Alcohol Health Alliance, 2020). Statistics of those being treated for addiction in England reflect these concerns, with 90% of all people in treatment for addiction recorded as white, highlighting an underrepresentation of ethnic minorities in treatment services nationally (Public Health England 2015; 2016)

The strong emphasis that judges place on whether a person has access to or begun treatment in order to demonstrate their commitment to addressing their addiction/offending behaviour associated with addiction, combined with the additional barriers ethnic minorities face in terms of access to treatment, means that this factor can become indirectly racially determined. The uneven distribution in the use of this factor between the White group and all ethnic other minority groups in favour of the White group, along with the significant impact it has on sentencing outcomes provides sufficient grounds to flag it as likely source of ethnic disparities in sentencing outcomes.

Discussion

Research on ethnic disparities in sentencing has predominantly concentrated on sentencing outcomes and has overlooked other key judicial decisions made as part of the sentencing process, resulting in a gap in understanding as to sources of disparities. Sentencing does not consist of just one judicial decision. Judges do not simply decide the final sentence to impose based on the facts of the case presented to them. Judges also play a vital role in the construction of a case. It is the judge who will decide whether sentencing factors (aggravating or mitigating) should be considered constitutive of the case to be sentenced and these decisions will have a direct impact on the final sentence imposed. Our study addresses the existing knowledge gap by focusing on these other judicial decisions made as part of the sentencing process. Our findings point to three sentencing factors being racially determined, and as such likely sources of unjustified ethnic disparities in sentencing outcomes.

We hypothesised that sentencing factors involving a high degree of judicial discretion and an assessment of the individual are at a high risk of being racially determined and therefore, a potential source of ethnic disparities in sentencing outcomes. Our findings support this hypothesis and show that out of the five sentencing factors at a high risk of being racially-determined, three of them - remorse, good character and ability to rehabilitate – were used more favourably to the White group. When we contextualised the use of these factors relying on sentencing guidelines and Court of Appeal guidance, as well as on past literature on how the factors are assessed, we found further evidence that supported our hypothesis.

A corollary of our findings is that remorse, good character, and ability to rehabilitate should explain some of the observed ethnic disparities in sentencing outcomes identified in previous studies where these factors were not controlled for (Hopkins, 2015; Hopkins et al., 2016). Perhaps more interestingly, the true extent of ethnic disparities reported in studies where these factors have been controlled for (e.g. Isaac, 2020), is underestimated. This would be the result of the common assumption inadvertently invoked by most researchers, who see sentencing factors as race neutral (Pina-Sánchez et al., 2022; Ugwudike, 2022; Omori and Petersen, 2020). If some sentencing factors are indeed racially determined and they are controlled for, then it risks some disparities being controlled away.

The identification of racially determined factors is therefore important for explaining past research findings and for enhancing the accuracy of future studies. Besides that, identifying factors as sources of ethnic disparities and explaining how they can operate as such, is vital for policy development. It allows for targeted and effective policies to be developed in order to reduce ethnic disparities in sentencing outcomes.

The Lammy review (2017) put forward the principle of ‘explain or reform’ for apparent disparities between ethnic groups within the criminal justice system. We have identified a substantive disparity in the use of remorse, good character and ability to rehabilitate and provided an evidence-based explanation as to how each can operate as a source of ethnic disparity in sentencing outcomes. Unless the disparities in the use of these factors can be explained in some other way and justified, then we argue that the discussion must move to considering what can and should be done to address the disparities stemming from these factors. We will now consider the various policy options that could be introduced to achieve this objective.

Policy options

A major challenge with addressing ethnic disparities in sentencing outcomes is that even when sources of disparities are identified, the root causes of those disparities can be wide and varied. This is evident from our explanation as to how remorse, good character and ability to rehabilitate can operate as sources of ethnic disparities in sentencing. As such, there is no single policy option capable of addressing disparities stemming from all three sentencing factors. For each factor, there will be different considerations and the suitability of policy options will vary between factors. It is necessary, therefore, to not only consider different policy options but also to consider their suitability for each individual factor.

The most straightforward option would be to amend the existing explanation attached to factors in sentencing guidelines. This would involve little financial cost and a low risk of conflict with core principles of sentencing. Such an approach has already been used by the Sentencing Council for other factors thought to have the potential to contribute to ethnic disparities in sentencing. For example, a new expanded explanation was recently introduced for the mitigating factor of mental disorder/learning difficulty, which in part states that there is ‘a range of evidence that people from ethnic minority backgrounds may be more likely to experience stigma attached to being labelled as having a mental health concern, may be more likely to have experienced difficulty in accessing mental health services and in acknowledging a disorder and seeking help’ (Sentencing Council, 2021, no pagination).4 Following a consultation with guideline users, the Sentencing Council has indicated their intention to make changes to the expanded explanations for number of factors including remorse, good character and ability to rehabilitate, in order to try address issues of equality and diversity in sentencing (Sentencing Council, 2023).

In our view, amending the expanded explanation is an approach that would be particularly suited to addressing ethnic disparities stemming from good character, less so for remorse and most likely not suitable for ability to rehabilitate. For good character, a minor amendment could be made to include additional examples of ‘good character’ that do not disadvantage individuals from different cultural and socio-economic backgrounds. These could be broad and include such things as integrity, trustworthiness and kindness to others. This minor change could have a significant impact on how the factor is assessed and used in cases involving individuals from ethnic minority groups. (Chen et al, 2023)

This same approach could also be used for remorse. The explanation could be amended to specifically highlight the risks of cross-racial assessments of remorse. Currently, it highlights the danger of an offender’s demeanour being misleading but does not discuss this in the context of ethnicity, different cultural backgrounds or research on the risks of cross-racial/cultural assessments of remorse. However, for this to be an effective approach we have to be confident that highlighting the risks of cross-racial assessments of remorse would be sufficient in order for those risks to be eliminated. Absent any research that we are aware of to support this position, there remains a considerable degree of doubt as to whether this approach alone would be sufficient to address potential ethnic disparities stemming from remorse.

Amending the expanded explanation would likely not be a suitable approach for addressing potential ethnic disparities stemming from ability to rehabilitate. While an expanded explanation could highlight the additional challenges for some ethnic minority groups in accessing treatment, due to the important role of rehabilitation as a purpose/aim of sentencing (Sentencing Act 2020, Section 57), it is much more difficult to see how simply making judges aware of these additional challenges would impact how the factor is assessed and used. If a judge does not believe a person can/will access treatment, and as a result has a high risk of reoffending, even if they are aware of the additional challenges, it is unlikely they will consider this mitigating factor as a constitutive part of the case, and take it into account in determining the final sentence.

Due to the low cost, the low risk of any knock-on unintended negative consequences, and the relative ease at which these amendments could be introduced, there is an argument to be made that the expanded explanations should be amended for each factor in the manner outlined above. This could be done as a first step while also considering additional approaches for remorse and ability to rehabilitate. If amending expanded explanations is the approach used, then it is important that this is done alongside initiatives to raise awareness amongst the judiciary. Chen et al (2023) in a recent report into equality and diversity in the work of the Sentencing Council, found that many judges were not aware of the content of the expanded explanations and recommended changing the format of the display on the webpage to make them more easily accessible and to also increase their effectiveness through ‘lived experience’ education for judges. The Sentencing Council accepted these recommendations and have already made some changes to their website and online guidelines in order to make both more user friendly and to make the expanded explanations more accessible.

An alternative or additional approach, which in our view should be explored for addressing ethnic disparities stemming from remorse would be to alter how the factor is assessed. Instead of a judge assessing whether a person is remorseful, remorse would be accepted/presumed once asserted and only rejected if there is clear evidence to the contrary, for example, inappropriate or intimidating behaviour towards the victim, or a record of repeatedly committing similar offences. If remorse is rejected, then a judge would be required to state the reasons for rejecting the person’s assertion of remorse.

The most challenging factor to address, but arguably also the most important, is ability to rehabilitate. Our analysis of the literature above suggests that disparities flowing through this factor stem primarily from the additional barriers, both real and perceived, for some ethnic minority groups in accessing drug or alcohol treatment. Addressing this factor as a purely sentencing issue may not therefore be the most effective approach. If the search for solutions is expanded out to the broader criminal justice system, then the National Probation Service stand out as the most suitable agency to tackle disparities stemming from this factor.

The National Probation Service could commission treatment programmes specifically for ethnic minorities to ensure there are appropriate programmes available for ethnic minority service users, ideally with staff from similar ethnic backgrounds, to reduce mistrust and the reluctance to engage. A recent review of race equality in probation highlighted that currently, the Nation Probation Service commission few services specifically for ethnic minority service users (HM Inspectorate of Probations, 2021, 24). At the pre-sentence report writing stage, probation officers should also be aware of and discuss with service users the barriers to treatment and then, if treatment programmes specifically for ethnic minority service users were available, to make them aware of this. The HM Inspectorate of Probations review (2021, 29) found that the quality of pre-sentence reports on Black, Asian and minority ethnic individuals was insufficient in 21 of the 51 reports inspected, with not enough consideration of the service user’s diversity. HM Inspectorate of Probation (2021, 12) has recommended the development of a national race equality strategy for probation service delivery, supplemented by strategic needs assessments in each probation region, to ensure that ethnic minority service users are not disadvantaged and receive culturally appropriate support. While ultimately, eliminating disparities stemming from this factor would require broader societal reforms aimed at breaking down real and perceived barriers to treatment for ethnic minority groups, developing and implementing a race equality strategy for the Probation Services would be a significant step in the right direction and would go a long way towards reducing disparities stemming from this factor.

Conclusion

The absence of understanding of the underlying factors contributing to ethnic disparities in sentencing outcomes has been a major obstacle to the crafting of effective policies to rectify them. Without a clear grasp of the root causes, policy makers are essentially operating in the dark and their ability to develop of targeted, evidence-based responses significantly hindered.

We focused on one part of the sentencing process: the judicial construction of a case. We have highlighted how ethnic disparities in sentencing are not only apparent in terms of outcomes, but they can also be observed in the application of certain case characteristics. Sentencing factors suspected to be racially-determined, due to their discretionary judicial assessment of the offender, indeed tend to be so. These factors are remorse, good character and ability to rehabilitate. We argue that much of the observed disparities in sentence outcomes could be derived from the differential application of these sentencing factors. Furthermore, having identified these factors as racially determined and likely sources of ethnic disparities in sentencing outcomes, we can also put forward targeted policy solutions.

Based on our analysis, we suggest that the most effective approach to address disparities stemming from good character would be to amendment to the expanded explanation in sentencing guidelines to include more inclusive examples of good character. For remorse, the way in which the factor is assessed should be altered, to introduce a presumption of remorse once asserted and only rejected with clear evidence to the contrary. To effectively reduce disparities stemming from ability to rehabilitate, reforms should be targeted at the delivery of probation services, in particular at the pre-sentence assessment and the commissioning of services and treatments specifically for ethnic minority service users.

It is imperative that proactive steps are now taken to alleviate ethnic disparities in sentencing outcomes. Addressing disparities stemming from remorse, good character and ability to rehabilitate would be a significant step towards achieving this objective.

Acknowledgements: This work was supported by the Secondary Data Analysis Initiative of the Economic and Social Research Council (Grant Ref: ES/W00738X/1).

References

Alcohol Health Alliance (2020), Race and alcohol addiction treatment services. available online at https://ahauk.org/race-and-alcohol-addiction-treatment-services/ (accessed 24 October 2023).

Ashworth, A. and Kelly, R. (2021), Sentencing and Criminal Justice. London: Bloomsbury Publishing.

Belton, I.K. (2017), Judges’ perceptions of good character: An analysis of England and Wales sentencing reports. Unpublished manuscript.

Belton, Ian (2018), The role of personal mitigating factors in criminal sentencing judgments: an empirical investigation. Middlesex University. PhD.

Bennett, C.D. (2016), The Role of Remorse in Criminal Justice. In: Tonry, M., (ed.) Oxford Handbook Online in Criminology and Criminal Justice. Oxford Handbooks Online: Oxford University Press.

Centre for Social Justice (2020), Facing the Facts: Ethnicity And Disadvantage In Britain: disparities in education, work, and family. available online at https://www.centreforsocialjustice.org.uk/wp-content/uploads/2020/11/CSJJ8513-Ethnicity-Poverty-Report-FINAL.pdf (accessed 24 October 2023)

Chen, Q., Vuk, M., Kuppuswamy, C., and Kirsch, D. (2023), Equality and diversity in the work of the Sentencing Council. Sentencing Council.

Edmiston, D (2022), Falling Faster Amidst A Cost-Of-Living Crisis: Poverty, Inequality and Ethnicity in the UK. Runnymede Trust.

Equality and Human Rights Commission (2016) Healing a Divided Britain: The Need for a Comprehensive Race Equality Strategy. Equality and Human Rights Commission

Field, S. and Tata, C. (2023), Criminal Justice and The Ideal Defendant in the Making of Remorse and Responsibility, London: Bloomsbury Publishing.

Fountain, J. (2009), ‘A series of reports on issues surrounding drug use and drug services among various Black and minority ethnic communities in England’, Drugs and Alcohol Today, 9(4), 41-42.

Gleeson, H., Thom, B., Bayley, M. and McQuarrie, T. (2019), Rapid evidence review: Drinking problems and interventions in black and minority ethnic communities. London: Alcohol Change UK.

Hopkins, K. (2015), Associations between police-recorded ethnic background and being sentenced to prison in England and Wales. Ministry of Justice.

Hopkins, K., Uhrig, N., Colahan, M. (2016), Associations between ethnic background and being sentenced to prison in the crown court in England and Wales in 2015. Ministry of Justice.

Hough, M., Jackson, J. and Bradford, B. (2013), ‘Trust in Justice and the legitimacy of legal authorities: topline findings from a European comparative study’, in Body-Gendrot, S., Hough, M., Levy, R., Kerezsi, K. and Snacken, S., eds., European Handbook of Criminology. London: Routledge, 243-65.

HM Inspectorate of Probation (2021), Race equality in probation: the experiences of black, Asian and minority ethnic probation service users and staff: A thematic inspection by HM Inspectorate of Probation. HM Inspectorate of Probation.

Hungerford-Welch, P. (2019), Criminal Procedure and Sentencing. London: Taylor & Francis Group

Isaac, A. (2020) Investigating the association between an offender’s sex and ethnicity and the sentence imposed at the Crown Court for drug offences. Sentencing Council.

Ismail, N., & Williams, K. (2012), Prevalence of Drug Use among Black and Minority Ethnic (BME) Communities in Bristol. Bristol City Council.

Jacobson, J., & Hough, M. (2007), Mitigation: The role of personal factors in sentencing. Prison Reform Trust.

Johansen, L.V. (2019), ‘Impressed’ by Feelings-How Judges Perceive Defendants’ Emotional Expressions in Danish Courtrooms, Social & Legal Studies, 28(2), 250-269.

Lammy, D. (2017), The Lammy Review: an independent review into the treatment of, and outcomes for, Black, Asian and minority ethnic individuals in the criminal justice system.

Lightowlers, C. and Pina-Sánchez, J. (2017), ‘Intoxication and assault: an analysis of Crown Court sentencing practices in England and Wales’, British Journal of Criminology, 58(1), 132-154.

Lymperopoulou, K. (2022) Ethnic Inequalities in Sentencing in the Crown Court - Evidence from the MoJ Data First Criminal Justice datasets. Swindon: ADRUK

Maslen, H. (2015), Remorse, Penal Theory and Sentencing. Oxford: Hart.

Ministry of Justice (2022), Diversity of the judiciary: Legal professions, new appointments and current post-holders - 2022 Statistics, available online at https://www.gov.uk/government/statistics/diversity-of-the-judiciary-2022-statistics/diversity-of-the-judiciary-legal-professions-new-appointments-and-current-post-holders-2022-statistics#:~:text=As%20at%201%20April%202022%2C%205%25%20of%20judges%20were%20from,and%20above)%20compared%20to%20others (accessed 24 October 2023).

Ministry of Justice (2021), Ethnicity and the Criminal Justice System 2020, available online at https://www.gov.uk/government/statistics/ethnicity-and-the-criminal-justice-system-statistics-2020/ethnicity-and-the-criminal-justice-system-2020#:~:text=Across%20the%20past%205%20years%2C%20White%20defendants%20consistently%20had%20the,and%20Black%20defendants%20at%2066%25 (accessed 24 October 2023).

Omori, M. and Petersen, N. (2020), ‘Institutionalizing inequality in the courts: Decomposing racial and ethnic disparities in detention, conviction, and sentencing’, Criminology 58(4), 678–713.

Phillips, C. and Bowling, B. (2003), ‘Racism, Ethnicity and Criminology. Developing Minority Perspectives’, British Journal of Criminology, 43(2), 269– 290

Pina-Sánchez, J., Geneletti, S., Veiga, Morales, A., Guilfoyle, E. (2022), ‘Ethnic Disparities in Sentencing: Warranted or Unwarranted?’, Center for Open Science.

Pina-Sánchez, J., Brunton-Smith, I., and Li, G. (2018), ‘Mind the step: A more insightful and robust analysis of the sentencing process in England and Wales under the new sentencing guidelines’, Criminology & Criminal Justice, 20(3), 268–301.

Pina-Sánchez, J. and Grech, D. (2017), ‘Location and sentencing: To what extent do contextual factors explain between court disparities?’, British Journal of Criminology, 58(3), 529–549.

Proeve, M. and Tanvir, T. (2022), ‘When ‘I did it’ is not enough: content analysis of remorse in sentencing remarks’, Current Issues in Criminal Justice, 34(3), 291-306.

Public Health England (2015), Adult Substance Misuse Statistics from the National Drug Treatment Monitoring System (NDTMS) 1 April 2014 to 31 March 2015. London: Department of Health.

Public Health England (2016), Adult Substance Misuse Statistics from the National Drug Treatment Monitoring System (NDTMS) 1st April 2015 to 31st March 2016. London: Department of Health

‘R v Howells’ (1999) 1 Cr. App R (S) 335

‘R v PS’ (2020) 2 Cr. App. R. (S) 9

‘R v Reid’ (1982) 4 Cr. App. R. (S) 280

‘R v Wenman’ (2015) 2 Cr. App. R. (S) 13

Rossmanith, K. (2015), ‘Affect and the Judicial Assessment of Offenders: Feeling and Judging Remorse’, Body & Society, 21(2), 167-193.

Rossmanith, K. (2014), ‘Getting into the box: Risky enactments of remorse in the courtroom’, About Performance 12, 7–26.

Saccomano, L. (2019), ‘Defining The Proper Role Of “Offender Characteristics” In Sentencing Decisions: A Critical Race Theory Perspective’, American Criminal Law Review, 56(4), 1693-1734.

Sentencing Council (2015), Crown Court Sentencing Survey, Annual Publication 2014, London: Sentencing Council.

Sentencing Council (2020) Sentencing offenders with mental disorders, developmental disorders, or neurological impairments Sentencing Guideline, available online at https://www.sentencingcouncil.org.uk/overarching-guides/magistrates-court/item/sentencing-offenders-with-mental-disorders-developmental-disorders-or-neurological-impairments/ (accessed 24 October 2023).

Sentencing Council (2021), Common Assault Sentencing Guideline, available online at https://www.sentencingcouncil.org.uk/offences/crown-court/item/common-assault-racially-or-religiously-aggravated-common-assault-common-assault-on-emergency-worker/ (accessed 24 October 2023).

Sentencing Council (2023) Miscellaneous amendments to sentencing guidelines: consultation 2023, available at https://www.sentencingcouncil.org.uk/html-publication/item/miscellaneous-amendments-to-sentencing-guidelines-consultation-2023/ (accessed 24 October 2023)

Social Metrics Commission (2020), Measuring Poverty 2020, A report of the Social Metrics Commission. Social Metrics Commission.

The Sentencing Act 2020, London: The Stationery Office.

Tyler, T.R. (2001), ‘Public trust and confidence in legal authorities: What do majority and minority group members want from the law and legal institutions?’, Behavioural Sciences and the Law, 19, 215-35.

Tyler, T.R. (2003), ‘Procedural justice, legitimacy, and the effective rule of law’, Crime and Justice, 30, 283-357.

Ugwudike, P. (2020), ‘Digital prediction technologies in the justice system: The implications of a ‘race neutral’ agenda’, Theoretical Criminology 24(3), 482–501.

Ugwudike, P. (2022), ‘Predictive Algorithms in Justice Systems and the Limits of Tech-Reformism, International Journal for Crime’, Justice and Social Democracy, 11(1), 85-99.

Veiga, A., Pina-Sánchez, J., and Lewis, S. (2023), ‘Racial and ethnic disparities in sentencing: What do we know, and where should we go?’, The Howard Journal of Criminal Justice, 62(2), 167-182.

Weisman, R. (2009), ‘Being and doing: The judicial use of remorse to construct character and community’, Social & Legal Studies, 18(1), 47–69.