Skip to content

Produced by the Sentencing Academy

Community Orders

Peer-reviewed research on the effectiveness and impact of sentences, orders, and requirements.

Eoin Guilfoyle, Brunel University of London

Published: 18 Sep 2025
Last updated: Sep 2025

Summary

This bulletin examines the effectiveness of Community Orders (COs) in England and Wales, focusing on their core aims of punishment, reparation, rehabilitation, as well as their cost-effectiveness compared to custody. The Community Order is a flexible sanction that can combine a wide range of requirements. Despite ongoing policy interest in increasing the use of community sentences as an alternative to short-term imprisonment, their use has more than halved since 2010.

COs can impose restrictions on a person’s liberty and a range of burdens that are experienced as punitive in practice. Through unpaid work, COs provide symbolic reparation by delivering tangible benefits to local communities, though challenges remain around resourcing, project availability, and ensuring placements meet community needs. In terms of rehabilitation, COs are associated with lower reoffending rates than short custodial sentences, and they avoid many of the harms linked to imprisonment.

While there are gaps and limitations in the research evidence that hamper definitive conclusions, evidence suggests that COs are achieving their core aims. Financially, COs are significantly cheaper than custody and likely more cost-effective overall, when reduced reoffending and wider social impacts are taken into account. Further research is needed to capture rehabilitative outcomes beyond reoffending rates and to assess their suitability for offenders with multiple or complex needs. Strengthening these areas will be key to building confidence in COs as a credible alternative to short-term prison sentences.

A Community Order can have a variety of requirements attached

The Community Order (CO) is one of the non-custodial sentencing options available to sentencers in England and Wales. It allows sentencers to pick and choose from a wide range of requirements (16 in total), combining them if needs be, when sentencing a person (section 201 of the Sentencing Act 2020).  The requirements that can be attached are:

  • unpaid work requirement (40 – 300 hours to be completed within 12 months)
  • rehabilitation activity requirement (the court does not prescribe the activities to be included, but will specify the maximum number of activity days the offender must complete).
  • programme requirement
  • prohibited activity requirement (must refrain from participating in specified activities for up to 3 years)
  • curfew requirement (2 – 20 hours in any 24 hours; maximum term 24 months, maximum of 112 hours in any period of 7 days.)
  • exclusion requirement (from a specified place/places; maximum period 2 years: may be continuous or only during specified periods)
  • residence requirement (to reside at a place specified or as directed by the responsible officer)
  • foreign travel prohibition requirement (not to exceed 12 months)
  • mental health treatment requirement (may be residential/non-residential; must be by/under the direction of a registered medical practitioner or chartered psychologist and the offender must have expressed willingness to comply).
  • drug rehabilitation requirement (treatment can be residential or non-residential, reviews must be attended by the offender at intervals of not less than a month (discretionary on requirements of up to 12 months, mandatory on requirements of over 12 months) and offender’s consent is required)
  • drug testing requirement
  • alcohol treatment requirement (residential or non-residential; offender’s consent is required)
  • alcohol abstinence and monitoring requirement (up to 120 days)
  • attendance centre requirement (12 – 36 hours. Only available for offenders under 25).
  • electronic compliance monitoring requirement
  • electronic whereabouts monitoring requirement

A CO must include at least one of the above requirements for the purpose of punishment or a fine must be imposed in addition to the CO, or the court can do both. It is a matter for the court to decide which requirements amount to a punishment in each case.

The CO was designed to be flexible and capable of fulfilling any of the purposes of sentencing. The breadth of requirements that can be attached to a CO, and the fact that the sanction can pursue multiple aims at once, can complicate efforts to assess the effectiveness of the sanction as a whole. For the purposes of this bulletin, effectiveness will be considered against what are generally seen as the CO’s primary aims: punishment, rehabilitation, and reparation. A short discussion of cost-effectiveness, compared with custodial sentences, is also included.

The Declining Use of the Community Order

The number of COs imposed each year has more than halved (down 56%) since 2010 (Ministry of Justice, 2024). During this same period, Suspended Sentence Orders declined by 9% and the use of immediate custody decreased by 27% (Ministry of Justice, 2024).  These changes have occurred during a period (2010-2023) when the overall number of sentences imposed on adults fell by 19% (Ministry of Justice, 2024).

Figure 1: Use of Community Orders, Immediate Custody and Suspended Sentence Orders, 2010 – 20231

When considering the proportionate use of the four main sanctions for more serious offences (fines, COs, Suspended Sentence Orders and Immediate Custody), it is COs that have experienced the greatest decline. Whilst the proportionate use of SSOs and immediate custody remained roughly constant between 2010 and 2023, the proportionate use of COs has approximately halved, with the use of fines appearing to increase at the expense of COs.

Figure 2: Proportionate use of Community Orders, Immediate Custody, Suspended Sentence Orders and Fines, 2010 and 20232

The average (mean) number of requirements attached to a CO was 1.6 in 2023 (Ministry of Justice, 2024b). The most common requirement attached to a CO was the Rehabilitation Activity Requirement, which was attached to 71% of COs that commenced during the year 2023. Unpaid work was the second most common requirement at 51%. Accredited Programme, Drug Treatment and Alcohol Treatment were the next most commonly used requirements at 9%, 6.6% and 4% respectively. 

Figure 3: Relative Frequency of Requirements Imposed, 20233

*Does not include all standalone curfews as most of these are not supervised by the Probation Service.

Community Orders and Their Core Aims: Punishment, Reparation, Rehabilitation

The punitive element

As a starting point, all criminal justice sanctions are punishments in the sense that they compel an individual to do something they would not otherwise choose to do (Bottoms, 2017). In the context of community sentences, the punishment element has most often been framed in policy discussions as restrictions on liberty, and more specifically on free time. When unpaid work was first introduced in the early 1970s, it was understood that the loss of leisure time, rather than the work itself, was intended to provide the sanction with its punitive bite (Kilcommins, 2014). This logic can be applied to most CO requirements, whether attendance at rehabilitative programmes, participation in drug or alcohol treatment, or compliance with curfews and electronic monitoring: all involve obligations to be in a particular place at a particular time, thereby restricting liberty and leisure time. Failure to comply can result in the imposition of further sanctions.

To equate punishment solely with the loss of free time, however, is to overlook many of the other burdens that COs impose. A growing body of research documents other punitive elements of community sanctions. Hayes (2015) identifies six categories of such elements: enforced lifestyle changes as part of rehabilitation; restrictions on liberty; welfare issues (such as disrupted relationships or reduced employability); submission to interventions by external agencies; institutional marginalisation (linked to perceived unfairness or being labelled the ‘usual suspect’); and stigmatisation. Specific requirements can also create their own distinctive hardships. Curfews and electronic monitoring, for example, bring additional punitive elements including privacy intrusions, the sense of being constantly monitored, and the stigma of wearing a visible device (Berry, 2019; McNeill, 2019; Gainey and Payne, 2000). These studies that have considered the experiences of individuals carrying out community sentences consistently highlight that the punitive impact of the sentences goes well beyond simple time restrictions.

There is not a single universally accepted metric that can be used to assess the extent to which a CO achieves the aim of punishment. Nonetheless, any realistic assessment of what a CO involves, and how its requirements are experienced by those subject to them, leaves little doubt that it is punitive in nature. There is a base level punitiveness with the degree of hardship experienced, which varies depending on the requirements imposed and the circumstances of the individual sentenced.

Despite this, political and public perceptions of COs as a “soft option” persist. Over the past three decades, there has been a sustained policy drive to make community sentences more punitive (Bottoms, 2017). There is, however, little evidence that such measures have altered use of community sanctions. For example, the Crime and Courts Act 2013 required sentencers to include at least one punitive element in a CO, yet the use of requirements typically regarded as punitive have not increased significantly since its introduction (Bottoms, 2017). Nor is there evidence that making COs more punitive changes how they are viewed by the public. With prison serving as the default point of comparison, community sanctions will almost inevitably be seen as less punitive. Hayes (2018) argues that community-based penalties are not necessarily less punitive, but punitive in different ways (original emphasis), but this nuance tends not to be captured in public and political debate.

The reparative element

While COs are undoubtedly punitive, punishment is only one aspect of their effectiveness. Another central purpose is to provide reparation. One of the main ways in which COs are reparative is through the performance of unpaid work. In 2023, an unpaid work requirement was included in just over half of all COs imposed (Ministry of Justice, 2024b). When imposing the requirement, a judge or magistrate can compel a person to carry out between 40 and 300 hours of unpaid work in the community, to be completed within 12 months.

Unpaid work can take the form of group projects or individual placements. Group projects are supervised by probation staff and cover a wide variety of tasks, such as removing graffiti, decorating public buildings, maintaining local parks, or growing food on allotments (Jackson et al., 2024; Ministry of Justice, no date). Individual placements allow those who are unable or unsuitable for group work (for example, people with disabilities) to complete their hours in a more flexible way. These often involve working in charity shops, with supervision provided by the beneficiary organisation (Jackson et al., 2024).

Between October and December 2023, 1,116,660 hours of unpaid work were completed – around 372,000 hours per month during that quarter. Over a 12-month period across 2023/24, a total of 4.7 million hours were delivered (Jackson et al., 2024; Ministry of Justice, 2024c). This work ensures that local spaces such as parks, churches, and schools across England and Wales receive essential maintenance and improvements that might not otherwise have been carried out. In addition, individual placements provide much-needed support to local charities, helping them deliver vital services to their communities. While unpaid work does not directly repair the harm caused to victims of crime, it can be considered symbolically reparative: by undertaking work of benefit to the wider community, people on probation are “paying back” society for the harm caused by their offending.

That is not to suggest, however, that the operation of unpaid work is without problems. Delivery was privatised under the Transforming Rehabilitation reforms (2014–2020), a period marked by significant difficulties, and although responsibility has since returned to the unified Probation Service, challenges remain. Post-unification, both a House of Lords Justice and Home Affairs Committee inquiry into community sentencing (2023) and a detailed Ministry of Justice evaluation of the unpaid work requirement (Jackson et al., 2024) have identified obstacles that need to be addressed to further strengthen the COs reparative potential. These include understaffing, system backlogs that delay the start and completion of hours, limited availability of engaging projects in some areas, and the need for stronger collaboration with local communities to ensure placements respond to local priorities and deliver the greatest possible benefits.

The rehabilitative element

A common metric used in policy discussions relating to the rehabilitative impact of sentencing options is reoffending rates. On this measure, community orders consistently perform better than short-term prison sentences. Figures from 2022/23 show that the reoffending rate for those sentenced to a CO was 38%, compared with 62% for those given a prison sentence of less than 12 months (Ministry of Justice, 2025). Frequency of reoffending is another measure used along with reoffending rates, with COs showing a lower average number of reoffences per person (4.1) than prison sentences of less than 12 months or less (7.5) (Ministry of Justice, 2025). 

These measures, however, are potentially misleading if used to directly compare the effectiveness of different sanctions. They do not account for key differences between the populations receiving each sentence. In practice, individuals assessed as higher risk of reoffending are more likely to receive a custodial sentence, while those assessed as lower risk are more likely to receive a CO. If more of the high-risk group reoffends after custody than the low-risk group after a CO, the difference cannot be attributed simply to the sanction imposed.

Some studies have sought to control for population differences. A Ministry of Justice study, for example, used over 150 variables to create matched groups for comparison. After matching, those sentenced to prison terms of 12 months or less were found to have had a 4% higher reoffending rate than those sentenced to a CO (Eaton and Mews, 2019). These findings broadly align with other international research showing community sentences to be associated with either similar or lower rates of reoffending than short custodial sentences (Villettaz, 2015; Klement, 2015).

There is, however, another limitation with using broad reoffending rates as a measure of rehabilitative effectiveness – one that likely understates or misses some rehabilitative achievements of COs. Reoffending rates do not capture important markers of desistance such securing accommodation, beginning education or employment, addressing addiction or mental health issues, building positive relationships etc. For example, a person who commits a minor offence during the 12 months after completing a CO will be recorded has having reoffended, but they may also have taken important steps towards long-term desistance that go unrecorded.

The available evidence is broadly positive in terms of the ability of COs to achieve the sentencing aim of rehabilitation. This is further strengthened when we consider the realities of participating on a CO versus serving a short prison sentence: the opportunities COs offer for addressing criminogenic needs (e.g. through education, treatment or training) alongside the well-documented harms of custody that include loss of housing, disruption of prosocial relationships, and damage to employment prospects – all of which can contribute to the risk of future offending (Sapouna et al., 2015).

Yet, limitations in available data and research evidence restrict more detailed and definitive conclusions regarding COs rehabilitative ability. Due to the importance of rehabilitation to public and judicial confidence in the CO and to achieving the policy goal of increasing the use of the sanction as an alternative to prison, additional data and further research would be useful. In particular, this area would benefit from additional data to allow for more compressive measures when assessing the effectiveness of COs in terms of rehabilitation, measures that extend beyond reoffending rates and capture key markers of desistance. With criminal justice datasets continuing to be linked with datasets from other government department as part of the Ministry of Justice Data First Project, tracking of some of these markers at a national level may soon become more feasible.  Another approach would be a longitudinal study of a cohort of people sentenced to COs in which these additional markers of desistance are tracked and studied. Doing so would provide valuable insights into some rehabilitative achievements of the CO that may currently be going unnoticed and unrecorded. 

A Cheaper Alternative to Prison

In 2021–22, the average cost of a prison place was £47,434 per year, or around £4,000 per month (Ministry of Justice, 2023). The Ministry of Justice does not publish the unit cost of COs annually. The most recent directly comparable figures are from the 2014/15 National Offender Management Service Business Plan, which reported that the average annual cost of holding a prisoner in 2012/13 was £34,766, compared with £4,305 for a community order or suspended sentence order.

On the basis of short-term pecuniary costs alone, COs are significantly cheaper than prison sentences. There are, however, broader factors to consider when assessing cost-effectiveness, including the impact of sentences on outcomes such as reoffending and the associated financial and social costs (Gormley et al., 2023). A full assessment of these wider factors is beyond the scope of this bulletin, but given the available evidence on the rehabilitative capacity of COs compared with prison, and even allowing for the limitations noted above, there can be reasonable confidence that COs represent a more cost-effective option than short prison sentences.

Conclusion

COs are an important non-custodial sentencing option in England and Wales, offering flexibility to punish, rehabilitate, and provide symbolic reparation through unpaid work. The evidence indicates that COs are punitive in practice, can deliver meaningful benefits to communities, and achieve rehabilitative outcomes, while also being significantly more cost-effective than prison sentences.

Given the current policy goal to reduce the use of short-term prison sentences (Ministry of Justice, 2025b), future research in England and Wales might examine the suitability of COs for individuals who are currently receiving prison sentences of 12 months or less. Existing evidence suggests that Judges and magistrates often lack confidence in COs, particularly in relation to their rehabilitative capacity (Crest 2017, Jackson et al., 2025). This concern is especially relevant for higher-risk offenders with multiple and complex needs, a cohort that tends to receive short prison sentences.

Much of the existing evidence on rehabilitation relies on reoffending rates, which are a limited measure that do not capture wider achievements, such as securing stable housing, education, employment, or progress in addressing addiction and mental health needs. Further research is therefore needed both to develop a fuller picture of rehabilitative outcomes and to assess the suitability of COs for individuals currently receiving short prison sentences. Addressing these gaps would help strengthen their operation and build greater confidence in their use as a credible alternative to custody.

This Bulletin is part of a series on the effectiveness of sentencing in England & Wales. Other Effectiveness Bulletins can be found here.

For further information, contact Annalena Wolcke at a.wolcke@sentencingacademy.org.uk

References

  1. Ministry of Justice (2024) Criminal justice system statistics quarterly: December 2023, Outcomes by Offence data tool: December 2023 ↩︎
  2. Ministry of Justice (2024) Criminal justice system statistics quarterly: December 2023, Outcomes by Offence data tool: December 2023. Excludes discharges, compensation and ‘otherwise dealt with’. ↩︎
  3. Ministry of Justice (2024b) Offender management statistics quarterly: October to December 2023, Probation: 2023, Table A6.9. ↩︎