Written evidence submitted by Dr Thomas Guiney and Professor Henry Yeomans (PEC0005)

Introduction

1. This submission is provided by Dr Thomas Guiney, Associate Professor of Criminology at the University of Nottingham, and Professor Henry Yeomans, Professor of Criminology at the University of Leeds. Between 2021 and 2023, Dr Guiney and Professor Yeomans collaborated on an inter-disciplinary study exploring the links between prison population forecasting, prison estate planning and new prison building in England and Wales. The results were published in the Howard Journal of Crime and Justice[1] and those findings are summarised here where relevant 

2. We welcome the publication of the recent NAO report ‘Increasing the capacity of the prison estate to meet demand’. There is a pressing need for greater democratic scrutiny of prison estate transformation and, in this response, we limit our observations to the findings set out in Part 3: Understanding and managing future demand (para 3.3.-3.7). Our central concern is that used uncritically, prison population forecasting can (a) normalise  a strategic assumption that prison population growth is inevitable and (b) amplify expansionary prison logics by connecting capital expenditure decisions about the overall scale, scope and reach of the prison estate with ‘business as usual’ planning cycles that assume considerable policy continuity with the past.
3. Our research demonstrates why it is so important that we challenge these self-fulfilling logics and create meaningful opportunities for democratic deliberation about the future of a prison system that is both morally justifiable and economically affordable. We cannot build our way out of current problems, and it will require considerable political commitment if we are to restore a more sustainable equilibrium between the supply and demand for prison places.[2]

Prison population forecasting.

4. In England and Wales, prison population forecasts are published annually by the Ministry of Justice as an ‘aid to policy development, capacity planning and resource allocation’.[3] The statistical bulletins report findings from a ‘dynamic flow model’ that tracks the movement of individuals through each decision-making stage of the criminal justice system and seeks to overlay these data with information pertaining to the anticipated impacts of certain policy positions.
5. We are not statisticians and we do not take issue with the statistical basis of these projections. Rather, we think it is important to recognise that, however sophisticated these statistical analyses may be, prison population forecasts are always underpinned by a series of assumptions about the world that inevitably privilege some ways of thinking about the future while discounting others. To elaborate:

(i)            Forecasting models typically adopt an uncritical posture towards current political assumptions and present a view of policy change that assumes the drivers of existing policy and practice will continue in the same direction and at a similar rate of change into the future. For this reason, prison population forecasts tend to emphasise policy continuity and discount the possibility of transformative policy change. Both the Criminal Justice Act 1991 and Criminal Justice Act 2003 introduced profound changes to the current sentencing framework and there is every reason to believe that the Independent Sentencing Review 2024 to 2025[4] led by the Former Lord Chancellor and Justice Secretary, the Rt Hon David Gauke will make similarly bold recommendations for change. 

(ii)          Prison population forecasts are frequently invoked by policymakers to promote a misleading impression of prison population growth as a natural phenomenon that is external, and therefore beyond the reach of government action. By treating trends in imprisonment as equivalent to weather forecasts, rather than the product of conscious government policy,[5] prison population projections translate moral and political problems about who, when and how to punish, into technocratic questions relating to capacity, demand, resilience and adaptation. In this way the job of government is reduced to a discussion of how to accommodate those sentenced to imprisonment by the courts, rather than recognising that these administrative challenges are the direct and foreseeable consequences of prior political and policy choices.

(iii)        While prison population forecasts are often presented as objective, scientific and neutral facts they are underpinned by a series of decisions about what is counted and what is not. The forecasts do not pose ‘what if’ questions relating to policy options that have not yet secured ministerial approval, nor do they explore the impact of policy changes that are considered unquantifiable (see discussion of the Female Offender Strategy below). Finally, we know very little about how prison population forecasts are themselves fed back into the annual cycle of prison planning and influence decision-making behaviour.

6. While prison population forecasts may appear neutral and value free, they are in fact mobilised by a wide range of policy actors including politicians, civil servants, penal reformers and the news media to shape public debates over prisons and imprisonment. It is curious then that these projections have such a low public profile. Neither the accuracy of these models, nor the assumptions that guide them regularly feature in public debates about prison building, public expenditure or the performance of our prison system. Despite billions of additional capital investment in new prison building, the Ministry of Justice does not routinely publish capacity projections making it difficult to scrutinise the delivery of these large capital infrastructure projects. For these reasons we agree with NAO recommendation 19a that, ‘To provide greater transparency to the public and parliament, MoJ should commit to routinely publishing capacity projections alongside its demand projections’.

Policy legacies and self-fulfilling logics

7. In sum, the interpretation and application of prison population forecasts can have significant and lasting policy implications. Expansionary criminal justice policies relating to policing, prosecution, sentencing or prison release are fed into the annual cycle of prison population forecasting and translated into value-free or politically neutral ‘facts’ about future ‘demand’ for prison places. These projections are then relied upon by officials, government ministers and committees of inquiry, such as the landmark May (1979)[6] and Carter (2007)[7] reviews of the prison system in England and Wales, to inform public expenditure decisions that drive long-term expansion or contractions in the prison estate. These moments of strategic self-reflection establish new baselines for the overall scale, scope and reach of the penal system, upon which successive phases of subsequent planning are then based.
8. Unlike other forms of public spending on criminal justice, such as the recruitment and training of staff, prison buildings endure across generational divides. They are less amenable to policy adaptation and the way we interact with these physical spaces can help to shape the administration of prisons for decades, if not centuries, as in the case of existing Victorian prisons.[8] As Spelman[9] found in his careful analysis of the US prison boom, criminal justice spending stood out as one of the strongest predictors of state–level prison population growth. Those states that invested most heavily in prison estate capacity and new prison building tended to experience higher and more sustained population growth over the long term than their neighbours. While capital spending did not in and of itself cause prison populations to increase, Spelman found that it did remove a very significant institutional counterweight against the sort of inflationary sentencing that drives up prison population.[10]
9. Interestingly, the National Audit Office reports that the Ministry of Justice’s prison population projections are premised on the assumption that current plans to increase prison places will be fully implemented.[11] As such, any potentially deflationary effect on prison population of limited prison estate capacity is discounted. This is particularly curious in the current context as the recent growth in prison population has been halted due to the introduction of new early release schemes.[12] These schemes were created because of a severe shortage of prison places. In other words, limited prison estate capacity has effectively constrained penal expansion.

Case Study: Prison population forecasting and women’s imprisonment.

10. We can demonstrate some of these points with respect to recent debates over women’s imprisonment. The female prison population decreased in size in the 2010s and the Ministry of Justice’s 2019 prison population forecast projected this downward trend would continue.[13] The 2020 prison population forecast, which predicted steep overall rises due largely to government plans to recruit 20,000 new police officers, then projected an increase in the number of women in prison (from 3217 in September 2020 to 4500 by September 2025).[14] Subsequent forecasts have projected similar increases. Importantly, these forecasts were used to secure additional capital investment, and justify new prison building for women.[15]
11. In this context it is noteworthy that while recent prison population forecasts assess the likely impact of police recruitment and statutory change, the influence of measures intended to reduce the women’s prison population are discounted. For instance, the Female Offender Strategy was launched in 2018 but, despite its aim for custody to be treated as a last resort for female offenders, its potential influence has yet to be factored into the prison population forecasts. The latest forecast report states that work to assess the impact of the Female Offender Strategy is still ongoing.[16]
12. Here we have further evidence that what is counted counts. Not only are projections for the future of the prison tied to the overall growth rate of the male prison population, but the decision to exclude the Female Offender Strategy from the baseline projection in effect unravels the distinct policy frameworks that should now exist to manage these very different prisoner cohorts. In our view this creates a dangerous self-fulfilling logic that militates against any realistic prospect of shrinking the women’s prison estate at a time when the prison system is gearing up for a period of sustained growth in the adult male population.

Recommendations

13. On the basis of our research, we make several recommendations which we hope will encourage greater public understanding of prison population forecasts and a more informed debate about the future of the prison system in England and Wales:

(i)            There is a strong and pressing need for serious and sustained scrutiny of prison population forecasts. We call upon key criminal justice opinion formers – government advisory boards, parliamentary select committees, penal reform organisations and academic research networks – to work together to improve the visibility of the annual prison population forecasts and discuss the political assumptions that sit behind these projections. This will help to mitigate the potentially inflationary effects of the forecasts upon prison population.

(ii)          We agree with the NAO that the MOJ should commit to routinely publishing its prison capacity projections. There is currently an information asymmetry between supply and demand side projections that stifles evidence-led policy making. Publishing the capacity projections will improve public knowledge of the challenges facing the criminal justice system and support the development of more effective penal policy.

(iii)       With the UK Government still committed to construction of thousands of new prison places, it is imperative that space is created for greater democratic deliberation about the continuing use of prison as a punishment and the extent to which this is morally and economically justifiable. We believe there is a need for a more pronounced ‘modesty’ about the very limited capacity of prisons to help solve longstanding social problems.[17] As we have already noted with respect to women’s imprisonment, new prison building is frequently justified on dubious humanitarian grounds as a way to ‘modernise’ the prison estate, while prison closures are often made conditional upon future reductions in the prison population that are treated as external and beyond the scope of government action. Our analysis demonstrates why the institutional link between prison population forecasting, prison building and ‘business as usual’ prison planning cycles must be subject to effective democratic scrutiny.

(iv)        The Independent Sentencing Review 2024 to 2025, led by the former Lord Chancellor and Justice Secretary, the Rt Hon David Gauke, represents a once in a generation opportunity to reflect upon the aims of sentencing and recalibrate how we balance the supply and demand for prison places in England and Wales. In this context, we see value in agile and imaginative form of strategic planning that embrace the possibility of change, as well as continuity. This means drawing upon a range of decision-making tools - including, but not limited to, prison population forecasting - that are designed to support those with responsibility for prison policy and planning to prepare for a wider-range of possible futures. In particular, we recommend greater use of more qualitative or narrative forms of scenario planning that draw upon international and historical experiences[18]. These innovative techniques have been used effectively by many successful businesses to prepare for an uncertain future and should be part and parcel of the governments decision-making repertoire.[19] The sustained use of scenario planning would be additional and complementary to prison population forecasts and provide alternative visions of the future, which can feed constructively into processes of policy formulation, evaluation and scrutiny.

January 2025