Revisiting the punitiveness of deportation

Abstract

Immigration measures such as deportation are currently not regarded as punitive and there has been little exploration of this from a legal perspective. This paper will consider this issue in depth, covering little discussed case law from the European Court of Human Rights. It will also explore how this legal position on deportation does not reflect the findings of other disciplines such as criminology and sociology on how immigration measures are used and experienced as punitive. This paper will build on existing literature by demonstrating the significance of a recent development in UK law to this debate. Section 47 of the Nationality and Borders Act 2022 (NBA 2022) introduced a ‘stop the clock’ provision into the Early Removal Scheme for foreign national prisoners. This new provision may prompt the judiciary to revisit the question of whether deportation is punitive in some contexts.

Introduction

Criminologists and sociologists have been examining the increased trend towards harsh immigration measures for many years now¹ and often use the framework of punishment in attempting to make sense of the use and experience of such measures.² However, despite the importance of this question, there has been relatively little legal literature exploring the potential punitiveness of immigration measures in Europe.³ Although there is more developed legal literature on this point in the US, many of these are of limited value as they are heavily grounded in US laws, contexts and practices.⁴ This lack of European literature that engages closely with a doctrinal analysis of the issue may be explained by the perception that this legal avenue has remained firmly shut, with courts maintaining that immigration measures are administrative in nature. However, this gap means there is relatively little analysis of relevant case law and the reasoning underpinning this legal position. By contrast, this paper will provide an extensive discussion of the case law of the European Court of Human Rights (ECtHR) and its approach to deportation as punishment to situate a new UK legal development in context.

The UK government recently passed the controversial NBA 2022. One provision, on the early release and removal of foreign national prisoners (FNPs), contained in section 47 of the Act did not attract much debate or controversy but may prove to be legally significant. The provision may prompt the judiciary to re-examine the current legal position that deportation is never a punitive measure. Thus, this paper will examine the current legal position on immigration measures such as deportation and removal and will expand on this literature by approaching the issue from a brand new angle through explaining the significance of section 47 and why the resultant changes to the Early Removal Scheme (ERS) may persuade the courts to revisit the question of whether deportation may be punitive in certain contexts.

The paper will begin by providing a brief overview of the current power of deportation largely focusing on the UK. This will include setting out some of the ways in which deportation powers have shifted over the years. Next the paper will consider why we may want to revisit the question of whether deportation can be punitive. It will set out and critique the rather shaky legal reasoning that currently maintains that deportation is never punitive and bring together the work that criminologists and sociologists have done in examining the use and experience of immigration measures to demonstrate the severity of these measures. An overview of the ERS will be carried out, and the concept of ‘FNPs’ will be explored, to situate the new UK development in context. This will be followed by a discussion of section 47 of the NBA 2022 and the changes it makes. Finally, the paper will examine the importance of this change and whether the courts are likely to re-examine the issue in light of section 47, including a discussion of a little-known ECtHR case, Gurguchiani v Spain,⁵ which may prove helpful.

1. A brief overview of deportation and removal

First, it is important to distinguish between deportation and administrative removal as, although these practices seem very similar, they are distinct powers. In the UK, deportation is a particular type of removal which should only occur when specific public policy grounds, normally related to either proven or suspected criminal activity, are shown by the state.⁶ Administrative removal, on the other hand, is the removal of any person who does not have leave to enter or remain in the country on the basis of their immigration status.⁷ This distinction between two sets of powers is not unusual. Although the names may differ, other European countries also have separate sets of immigration measures that differentiate between removal based on status and removal based on public policy reasons.⁸

Deportation and administrative removal are thus separate powers but it is worth noting that their differences are less prominent than they used to be. The consequences of administrative removal used to be different to deportation orders because administrative removal used to create no obstacle to a person returning to the UK in the future. A deportation order, on the other hand, means that the deportee cannot return to the state until the order is revoked (which it may never be).⁹ There is now a presumption, however, that those who have been subject to administrative removal will be refused a UK visa.¹⁰ This means that most who have been removed will not be able to return to the UK, at least for a significant period of time. The line between deportation and administrative removal has thus become less clear now that the reality of their consequences – ie long term exclusion – is essentially the same. The line is further blurred by the fact that in many European states use of deportation and removal has become increasingly harsh and tied to restrictive processes.¹¹ For example, in the UK, deportations used to offer more process rights than administrative removal but since the 1980s the procedure, particularly possibilities for appeal, has been gradually eroded,¹² with the UK Government in 2014 introducing a policy of ‘deport first, appeal later’, which was ruled unlawful by the UK Supreme Court in June 2017.¹³ Administrative removal is a broad power: it does not require any kind of judicial process in the UK prior to removal and the actual power of administrative removal has not changed much since the 1970s but its use has grown significantly.¹⁴ These developments have also occurred in much of the rest of Europe, with many states expanding powers and budgets for both deportation and removal.¹⁵

Both powers have also become much more intimately related to the criminal justice process. This is especially true for deportation. Although protection of public security has long been a ground for deportation in many European countries, the use of immigration measures after criminal conviction has become increasingly harsh.¹⁶ Since 2000, the UK,¹⁷ Denmark,¹⁸ Germany¹⁹ and Italy²⁰ have all passed laws making deportation an automatic result of many criminal convictions where the convicted person is a foreign national. In Switzerland non-citizens who commit serious crimes can be subject to automatic deportation and there have been attempts to extend this to minor crimes.²¹ Even where deportation is not automatic, severe immigration consequences for committing a criminal offence have become the norm. For example, in 2002 the Netherlands passed a law allowing for the deportation of foreign residents who had been sentenced to one month in prison or community work.²² These developments are not isolated and are part of a wider pattern of the criminalisation of immigration or ‘crimmigration’ law as will be discussed below.²³

2. Why revisit the punitiveness of deportation?

In order to understand the potential legal importance of the new section 47 provision, it is necessary to examine how the courts have traditionally conceptualised and dealt with immigration measures like deportation. Despite the increasing overlap between removal, deportation and the criminal justice system, legal interpretations of deportation and removal continue to maintain that such measures are non-punitive. This is not only the case in the UK but is also the position of the ECtHR. The case law on this point has centred on the right to a fair trial, which immigration decisions – including removal and deportation decisions – do not benefit from.

(a) The current legal position of deportation and removal

To determine whether a measure is punitive the ECtHR normally looks at whether the proceedings concern a ‘criminal charge’. This is an autonomous concept²⁴ so its definition is independent of any meaning given to it in domestic legal systems. In Engel and Others v the Netherlands ²⁵ the ECtHR laid down three criteria that it would look at when considering whether proceedings amounted to a criminal charge: (1) the classification of the proceedings in national law; (2) the nature of the offence; and (3) the severity of the penalty that the individual may incur. The first element does not generally carry much weight, as that would endanger the autonomy of interpretation, and more weight is usually given to elements 2 and 3 of the Engel criteria.²⁶ The second and third criteria do not have to be considered cumulatively; one may be decisive for the Court.²⁷ If neither is decisive on its own, however, then the Court may also consider them cumulatively.²⁸ This seems relatively straightforward, but in the immigration context the Court does not appear to refer to the Engel criteria much when it is alleged that an immigration decision really concerns a criminal charge and therefore is punitive. The question of whether deportation or removal measures are punitive was brought to the old European Commission on Human Rights a couple of times²⁹ but the seminal ECtHR case on this point is Maaouia v France.³⁰

The case of Maaouia ³¹ concerned a Tunisian national who was living in France and convicted of armed robbery and armed assault with intent. He was released from prison in April 1990 and a deportation order was made. This was not served on him until October 1992 and he refused to comply with it, resulting in prosecution. For the refusal to comply he was given a one-year prison sentence and a ten-year ban from French territory. The initial deportation order was later quashed, however. As the deportation order was invalid, the applicant successfully sought the rescission of the ten-year ban but this took four years. The applicant argued that the length of time taken to make this decision violated his rights under Article 6 of the ECHR. The French Government argued that this claim was invalid, as deportation and exclusion orders do not concern criminal charges, so Article 6 did not apply – in other words, they argued that it was not a punitive measure.

In its judgment the Court emphasised that ‘criminal charges’ are autonomous concepts and cannot be interpreted only with reference to national classification. Yet, in its consideration of whether the proceedings could be said to involve the determination of a criminal charge – remember the ban from France for ten years was a result of a criminal conviction for refusing to comply with a deportation order – the Court appeared to rely exclusively on national classifications. It noted that such orders are not classed as criminal in France or in the Contracting Parties to the ECHR generally. The Court focused on the fact that such orders are often made by administrative authorities as well as by criminal courts and therefore that such measures constitute a special preventive measure for immigration control.³² The Court thus concluded ‘The fact that [the immigration measure] are imposed in the context of criminal proceedings cannot alter their essentially preventive nature. It follows that proceedings for rescission of such measures cannot be regarded as being in the criminal sphere either.’³³

Thus, the Court based its opinion that immigration decisions do not concern criminal charges on the fact that states usually do not classify exclusion orders as criminal. This seems to contradict the idea that the way in which states classify proceedings carries very little weight in determining whether a criminal charge is at stake. The fact that administrative authorities decide such proceedings is also dependent on state classification. This reasoning has been criticised as a ‘reversal of the ECtHR's long-established independence regarding the autonomous concepts within Article 6’.³⁴ The Court neglected to touch on other evident issues in this case. For example, throughout its submission the French Government explicitly said that the purpose of the measure was to act as a deterrent but the Court did not even address this point.³⁵ This is odd because the Court has found on several occasions that a measure which has the purpose of deterrence indicates a criminal charge.³⁶ The Court also did not examine the severity of the measure at hand, which again seems unusual given the importance of the third Engel ³⁷ criterion.

The Court's decision in Maaouia ³⁸ still stands and deportation and removal remain classed as non-punitive. However, Maaouia is not the only case which considered this problem; another case touched on the issue of immigration measures as punitive in more depth. The case of Uner v the Netherlands ³⁹ focused on a Turkish national who had been living in the Netherlands since he was a child. He and his Dutch partner had two children. He had committed several violent offences and in 1994 was convicted of manslaughter and assault resulting in seven years in prison. In 1997, in light of his criminal record, the authorities rescinded his Dutch residence and gave him a ten-year exclusion order prohibiting him from returning. This case primarily focused on Article 8 of the ECtHR (the right to private and family life), with the Grand Chamber ultimately deciding that there was no violation. However, the case also touched on the issue of immigration measures as punitive. The applicant argued that rescinding his residence and imposing an exclusion order were a ‘second punishment’ and complained that this violated Article 6 and Article 4 of Protocol No 7 to the Convention (the right not to be tried or punished twice). These arguments were declared inadmissible by the Chamber⁴⁰ with the court simply stating that Article 6 does not apply to immigration decisions, citing Maaouia and dismissing the Protocol 7 complaint on the grounds that the Netherlands had not ratified it. Despite this, several judges and parties to the Grand Chamber judgment did consider this argument. For example, the German government, intervening as a third party, emphasised that expulsion orders are not punishments because they are ‘aimed at guaranteeing public safety in the future without the intention of inflicting a punishment’.⁴¹ The Grand Chamber judgment likewise stated that immigration measures such as removing a residence permit or imposing an exclusion order after a migrant has been convicted of a criminal offence does not constitute a double punishment generally.⁴² It reiterated the idea that these measures are preventive ‘rather than punitive in nature’.⁴³ However there was a significant dissenting opinion on this point, with Judges Costa, Zupancic and Turmen strongly disagreeing with the view of the majority:

… we believe a question of principle to be at stake, one which we would like to conclude. The principle is that of ‘double punishment’ or rather the discriminatory punishment imposed on a foreign national in addition to what would have been imposed on a national for the same offence. We do not agree with the assertion in paragraph 56 that the applicant's expulsion was to be seen as preventative rather than punitive in nature. Whether the decision is taken by means of an administrative measure, as is this case or by a criminal court, it is our view that measures of this kind which can shatter a life or lives – even where, as in this case, it is valid at least in theory, for only ten years (quite a long time incidentally) – constitutes as severe a penalty as a term of imprisonment, if not more severe.⁴⁴

Again, in this case there was no in-depth consideration of the Engel criteria: both the Grand Chamber and the Chamber judgment were generally content to cite Maaouia. The idea that immigration measures were preventive in nature was reiterated but with little discussion or explanation of what that meant. The Dissenting Opinion of several judges on this point demonstrates that the decision was not uncontroversial.

In the UK policy developments and judicial and political rhetoric have likewise demonstrated the controversy and confusion around the use of removal and deportation and their increased intertwining with criminal justice. UK judges have found that recommending deportation orders does not constitute part of the punishment for the offender⁴⁵ but noted that the need for the judge to consider whether the offence was serious enough to merit deportation did point towards deportation being part of the punishment.⁴⁶ This is reflected in the actual use of deportation orders by judges in criminal trials in the UK: Home Office guidance describes how judges are reluctant to recommend a deportation order unless a particularly serious offence has been committed.⁴⁷ This relationship between criminal conviction and deportation has been further emphasised by the fact that deportation is now an automatic result of certain criminal convictions,⁴⁸ regardless of whether the judge recommended it or not. When deportation is explicitly considered by a judge, the rationale is often expressed in terms that seem very difficult to separate from punitive aims. For example, Lord Justice Rix stated that ‘The public interest in deportation for those who commit serious crimes goes well beyond depriving the offender in question from the chance to re-offend in this country: it extends to deterring and preventing serious crime generally and to upholding public abhorrence of such offending.’⁴⁹

Another more recent judgment in the UK demonstrates the continued confusion over the purposes of deportation. The case concerned whether a Jamaican man (SC) could be deported.⁵⁰ He had lived in the UK since the age of 10 and been recognised as a refugee because his mother is a lesbian and both he and his mother had been violently persecuted by a gang in Jamaica because of her sexuality. In June 2012 SC had been sentenced to two years in a young offender institution, triggering the automatic deportation process. As a refugee, it was accepted that he could not be returned to his home area in Jamaica, but the question was whether he could be required to internally relocate in Jamaica. The usual test for this is whether this would be ‘unduly harsh’; the First-Tier Tribunal found that it would, as he would face destitution, homelessness and be in need of psychological treatment. The Court of Appeal, however, took a different approach and suggested that this ‘unduly harsh’ test should apply differently to someone who has been convicted of a crime. This was rationalised by stating that ‘[t]he phrase “unduly harsh” imports a value judgment of what is “due” to the person’.⁵¹ In other words, criminal conviction would justify imposing greater hardship on the person because they deserve it, which indicates some retributive aims associated with punitiveness. This point was overturned in the UK Supreme Court as incorrect but demonstrates the difficulties of understanding these measures outside of a punitive lens. These are not the only examples and others such as Sheona York and Jonathan Collinson have also noted similar punitive language and justifications in UK deportation cases.⁵² This confusion is not limited to UK judges either; others, such as the US Supreme Court, have struggled – stating that deportation is ‘uniquely difficult to classify’⁵³ as it is ‘intimately related to the criminal process’.⁵⁴

Thus, the legal reasoning that underpins the idea that deportation and removal are never punitive is a shaky foundation. The rationales often seem confused or weak and circular. This is particularly problematic given the extensive study of ‘crimmigration’ by other disciplines such as criminology and sociology which suggest that the reality behind the use of these measures is often severe and difficult to conceptualise without a punitive framework.

(b) ‘Crimmigration’ and the punitiveness of immigration measures

Although how to define and identify punishment is hardly a settled matter in academic debate,⁵⁵ many criminologists have turned to this prism as a tool for understanding immigration measures. These comparisons, especially in terms of their individual impact, often seek to demonstrate that such measures are punishment in all but name.⁵⁶ Immigration detention, where most migrants are held prior to removal or deportation, has been widely compared to prison by criminologists and is considered to have punitive elements. If we look to the use of immigration detention in the UK, it is clear why. In the UK most immigration detention takes place in Immigration Removal Centres (IRCs). Many of the IRCs (current and former) had a previous life as a prison or were designed to have a prison-like appearance and feel.⁵⁷ Morton Hall IRC, a former prison, was run by HM Prison and Probation Service on behalf of the Home Office, despite the fact that it held immigration detainees rather than prisoners, and was surrounded by large quantities of razor wire. At the end of 2021 it was quickly converted back into use as a prison, further demonstrating the similarity.⁵⁸ Throughout many IRCs, HM Inspectorate of Prisons (who also inspect IRCs) has continued to find disproportionate restrictions on movement for detainees as well as widespread handcuffing and routine room searches.⁵⁹ In the most recent HMIP inspection of Dungavel IRC the need for the use of patrol dogs⁶⁰ was questioned, as was the carrying of batons by staff in Morton Hall.⁶¹ In recent years IRCs such as Yarls Wood and Brook House have been the subject of exposé documentaries showing a culture of violence and degradation by IRC staff against detainees.⁶²

Empirical criminological and sociological work examining how detainees experience immigration detention have found it to be punitive.⁶³ In the UK, detainees who have experienced both prison and IRCs report their overall experience in IRCs to be worse than prison.⁶⁴ Those who were FNPs and held beyond their sentence release date under immigration powers were angry and confused as to why they continued to be detained, if it was not to punish them again.⁶⁵ The fact that immigration detention may be indefinite in the UK also compares unfavourably with prison as the detainees had no release date and so lived with constant uncertainty.⁶⁶ This lack of certainty creates widespread fear, not just for FNPs but all immigration detainees, as noted by HM Inspectorate of Prisons in many of its reports.⁶⁷

Deportation has received somewhat less attention than immigration detention. Perhaps this state power is less controversial than immigration detention, or lacks a clear comparator, unlike immigration detention and prison. Or maybe it is due to the very practical difficulty of keeping track of and in touch with deported participants. Nevertheless, there have been studies demonstrating the potential physical severity of deportation and administrative removal practices.⁶⁸ Moreover, the potentially severe and profound effects it has on those subject to it have also been well documented. Deportation ‘disrupts the whole tenor of life’⁶⁹ and can ‘shatter a life or lives’.⁷⁰ In popular discourse, deportation is seen as returning a person to their ‘home’ and therefore issues of reintegration on return are rarely discussed. This is misleading. Deportees, even those who have maintained ties to the receiving country, suffer a range of problems on return to their ‘home’ state, including economic, psychological and social trouble and general issues of reintegration.⁷¹ The problem is considered to be particularly acute for those who left their country of origin as children and some scholars have shown increased interest in recent years for legal reform to protect this particular group from deportation.⁷²

Beyond the ‘hard treatment’ that such measures entail,⁷³ the use of such detention and deportation is often bound up with aims of criminal justice. As mentioned above, this can be seen in the legal rationales for deportation⁷⁴ and some studies have found immigration detention is used in a similar manner – punitively for retribution or to act as deterrent.⁷⁵ Likewise, there have been comparisons made between historical forms of punishment, such as banishment and transportation, and the modern practice of deportation.⁷⁶ There is also greater awareness of the importance that immigration status has for determining an individual's experience of the criminal justice system and that it is increasingly difficult to separate out immigration measures and the criminal justice system.⁷⁷

It is worth noting that reconceptualising certain immigration measures as punitive and determining which do and do not fall within that categorisation is not straightforward. There is already much debate and internal critique in the field as criminologists struggle to conceptualise these measures that do not fit neatly into either criminal or administrative frameworks.⁷⁸ The levels of discretion and ‘asymmetric incorporations of criminal justice norms’⁷⁹ make it difficult to see many immigration measures as purely criminal and the stigma, use and intermeshing of the criminal law and practices drawn from the criminal justice system make it hard to see them as administrative. Moreover, this phenomenon is not isolated to immigration measures: the increased relevance and use of ‘preventive measures’ in other areas also suffers from similar complications in conceptualisation.⁸⁰ Yet, despite the conceptual uncertainty it is worth bearing in mind the similarities to punishment. It is important that human rights act as proper check on state power, especially when it comes to vulnerable groups such as migrants. The ECtHR must make good on its ‘motto … that the Convention must be interpreted in a manner which renders its rights “practical and effective, not theoretical and illusory”’.⁸¹

3. Foreign national prisoners and the early removal scheme

Recent changes to UK law should re-open the debate on the punitive power of deportations. Section 47 of the NBA 2022 changes the way FNPs are dealt with. Before examining the provision itself, the ERS and the concept of the FNP must be explained. FNPs are persons without an absolute right to remain in the UK (ie those without British citizenship⁸²) who are remanded or sentenced to a UK prison. For a measure which is non-punitive, the introduction of automatic deportations mentioned above certainly had a significant effect on the prison architecture of the UK and made a prisoner's nationality or citizenship status increasingly salient. As criminal convictions and immigration consequences have become more closely related to one another, this has been followed by changes which separate prisoners into different institutions dependent on their (actual or suspected) nationality. Automatic deportation prompted changes to the structure of the UK's prison estate where non-citizen offenders are placed in certain prisons: dedicated ‘foreign national offender only’ prisons.⁸³ These prisons are set up to deal with the immigration system with full-time immigration staff onsite. There is evidence, however, that they are lacking in many other respects such as facilities and rehabilitation support.⁸⁴ This concept of FNP-only prisons and the perception that they provide a lesser service than typical prisons is not unique to the UK – there is a similar system in Norway.⁸⁵

What section 47 of the NBA 2022 does is shift the way in which the ERS for FNPs works. To facilitate their deportation, FNPs may be released early from prison and deported to their country of origin through the ERS. This scheme allows for the removal of FNPs up to 12 months prior to their earliest release point, so long as they have served a certain portion of their sentence; for many this will be a quarter of their sentence.⁸⁶ For example, someone with a two-year sentence would typically have to spend 12 months in prison before being considered for release. ERS means that if that person is an FNP, they need to spend six months in prison (a quarter of their sentence) before being considered for early removal. In other words, they can be removed six months earlier than they would have been considered for release from prison if they were a British national.⁸⁷

The ERS has very few limitations on who is included with little restriction on offence type.⁸⁸ All FNPS with a determinate sentence must be considered under this scheme except those convicted of certain offences connected with terrorism.⁸⁹ The ERS includes those in young offender institutions,⁹⁰ but those who are serving indeterminate sentences for public protection and life sentences are not eligible under this scheme and instead will be considered for removal after their tariff has expired under the Tariff Expired Removal Scheme (see below). Prison governors can refuse permission for the ERS in any of the following situations: (1) there is clear evidence that the prisoner in question is planning another crime (including immigration crimes); (2) there is evidence that, while in prison, the prisoner engaged in violence or threats of violence numerous times; (3) the prisoner has been dealing class A drugs while in custody; and (4) the prisoner is serving a sentence for a terrorism or terrorism-connected offence (this is broader than the exclusion above as it includes any terrorist offence and the Joint Extremist Unit will be involved in any final decision). They may also refuse due to other matters of similar gravity relating to public safety or where early removal under the ERS would undermine public confidence in the criminal justice system. Guidance provides that the latter is anticipated to be used very sparingly and only for ‘notorious’ offenders in cases of national profile.⁹¹ Prisoners are informed, in writing, of their early removal once the prison governor has approved it.⁹²

The numbers potentially involved in the scheme are significant. FNPs currently make up 12% of the prison population.⁹³ According to the immigration statistics for the year ending June 2023, foreign national offenders (FNOs) make up the biggest proportion of enforced returns (72%), though this has fallen from pre-Covid levels, when they were far and away the vast majority of enforced returns.⁹⁴ Enforced return numbers are steadily increasing compared to the pandemic years but are yet to get close to pre-pandemic levels. The numbers removed may increase in the future as the number of total (whether released early or not) FNP deportations appeared to be increasing prior to the pandemic, with 4,700 deported between 2019 and 2020.⁹⁵ It is worth pointing out that references to FNO removals in statistics do not automatically mean that they were all removed through the ERS – an FNO is defined as ‘someone who: (a) is not a British citizen; and (b) has been remanded in custody, convicted and given a custodial sentence in the UK for any offence. An FNO can be convicted and have served their sentence while on remand, so would not necessarily have been sent to prison.’⁹⁶ The majority of returned FNOs are EU nationals (58% in 2023) though Albanian is by far the single most prominent nationality of returned FNOs.⁹⁷

Brexit might also increase the number of FNP deportations as it changed the process of deportation of EU nationals. Its impact is difficult to assess and might turn out to be a major or minor shift. It is potentially major because prior to 31 December 2020 EU nationals who committed an offence benefited from stricter proportionality rules surrounding their deportation that focused on their personal conduct and length of residency.⁹⁸ Now any EU national who commits an offence after 31 December 2020 will be subject to the same rules as any other FNO,⁹⁹ where it is assumed that their deportation is in the public interest, and this generally results in a less generous proportionality test with less regard to their personal conduct and length of residence.¹⁰⁰ Thus EU nationals will now be worse off if they offend and in theory easier to deport. On the other hand, the impact is arguably somewhat minor as EU nationals made up a significant proportion of all deported FNPs prior to Brexit anyway¹⁰¹ and it is debateable how much the old test really constrained the UK's ability to deport EU FNOs.¹⁰² Nevertheless, we may see an increase in the number of total FNOs deported because of the less strict test for EU nationals.

Turning to the ERS specifically, according to a parliamentary question asked and answered in 2023: from January 2010 to June 2022, the Home Office apparently removed 22,707 FNPs through the ERS, with 1,322 of those being in the year ending June 2022.¹⁰³ A further 571 FNPs serving indeterminate sentences have been removed since 2012 under the similar Tariff Expired Removal Scheme, but due to the fact that their sentence was indeterminate (ie without a set end date) their situation is somewhat different to the typical FNP.¹⁰⁴ If we take all the developments mentioned above and couple it with the general expectation that the whole prison population is projected to grow significantly and increasing FNP deportations is being touted as a solution for overcrowding, then it is clear that the changes brought by section 47 could affect a large number of people.¹⁰⁵

4. Section 47 of the Nationality and Borders Act 2022

So, what did section 47 of the NBA 2022 change? Previously, if a FNP was deported through the ERS and then returned to the UK before their original criminal sentence end date expired, they could then be detained in prison again for the remainder of time until their original sentence end date.¹⁰⁶ Section 47 changes this so that the original sentence end date is no longer relevant. Now if an FNP is deported and returns to the UK after their original sentence end date they can still be detained for the remainder of time left on their sentence at the point they were deported. This means that when a foreign national is removed as part of the ERS, the threat of recall to prison does not expire when their original sentence time expires: instead it goes on indefinitely. In other words, the sentence is ‘paused’ while the foreign national is in their country of origin. This is confirmed by the government's own explanatory notes:

this clause introduces a ‘stop the clock’ provision, which will apply to FNPs removed under ERS. The new provision will, in effect, pause the sentence at the point a person is removed from prison under ERS. If the removed FNP returns to the UK at any point in the future, they would be liable to be detained and returned to custody to serve the balance of their sentence, ‘ignoring’ any days after they were removed from prison…¹⁰⁷

Thus, a FNP who is released early for deportation will never be able to return to the UK without incurring imprisonment for the remainder of their sentence. This is an important shift. It makes the consequences of offending for foreign nationals more severe with longer term consequences than a British citizen would face. Most importantly, it opens up the possibility that the deportation is intended to replace the prison sentence and is therefore a punitive decision.

This argument can seem counterintuitive. It would be understandable to see deportation under the old system, where the deportation formed part of a sentence, as more punitive than the new system, where it separated from the sentence. Yet this is not how punitiveness is typically conceptualised. Under the old system, the early release of FNPs to their country of origin was analogous to the early release of other offenders who may also be required to stay within or away from certain areas or be limited in their movement at certain times of day for the remainder of their sentence. These early release and recall to prison decisions for all prisoners are not generally classed as punitive but administrative.¹⁰⁸ The reasoning behind this focuses on the fact that this is not a ‘fresh’ decision on punishment but rather an administrative decision on the implementation of the punishment from the original sentence and is aimed at preventing further reoffending.¹⁰⁹ Lady Hale has made clear her discomfort on this point¹¹⁰ and some academics have pointed out that these decisions seem like ‘back-door’ sentencing which determine the extent and character of punishment in important ways.¹¹¹ Nevertheless, the legal position remains that in that context, these decisions are not punitive but administrative. Section 47, however, shifts this context quite significantly. Deportation is no longer part of the sentence in the same way, if the sentence does not carry on counting down while it happens. Deportation is now what happens instead of the sentence – this is made clear by the fact that once deportation occurs, the sentence is paused and only resumes when the deportation is no longer effective – ie when the person returns to the UK. Pausing the sentence makes it seems more likely that the deportation measure is now divorced from the implementation of the original sentencing decision and thus may be a ‘fresh’ punitive decision.

5. Section 47 and re-evaluating deportation

If the courts, including the ECtHR were to examine the deportation provisions in section 47 of the NBA 2022 it is possible that they might shift their interpretation of this measure as non-punitive.

First, as discussed at length above, the deportation measure seems much harder to see as part of the original sentence decision when the original sentencing decision is paused and does not resume unless the deportation is rendered ineffective. This seems likely to undermine much of the judicial reasoning as to why such decisions in that context are not punitive. Secondly, it is difficult to see how this does not fundamentally alter the criminal punishment of non-nationals versus the punishment of nationals. Non-nationals subject to this measure will indefinitely have a sentence hanging over their heads. Despite being sentenced by a criminal court, it is unlikely in many cases that their sentence will ever be spent, as even if they return to the UK after being deported, the process will likely start again and the sentence will again be paused and so on and so forth. An analogy may be drawn with those under a life sentence. Only life sentence prisoners currently exist with the threat of recall to prison hanging over their lives indefinitely and, given that FNPs only need a sentence of a year to qualify for automatic deportation, the extension of that to other categories of offender seems potentially very disproportionate to the offence. This is especially true now that section 40 of the NBA 2022 criminalises all irregular entry and stay with a punishment of up to 12 months in prison, meaning anyone with that status may be subject to these provisions.

The proportionality of this response by the state to a crime is particularly open to question given the approach of the ECtHR on the issue of life sentences without prospect of rehabilitation and release.¹¹² While not directly analogous (because in our case the FNP is released) the Court's reasoning that such life sentence are incompatible with the ECHR was partly underpinned by the fact that the offender lacked the possibility to be recognised as having been rehabilitated and suitably atoned for his offence.¹¹³ A criminal sentence that can never be spent does not seem to fit well with that reasoning. The power of this reasoning has been somewhat diluted, however, by the more deferential approach taken by the ECtHR in subsequent case law such as Hutchinson.¹¹⁴ Nevertheless, the potential severity of this measure should not be quickly dismissed; as mentioned briefly above, the pains of constant uncertainty in the immigration system are well-documented.¹¹⁵ Many migrants who have experienced both prison and immigration detention find that the latter compares unfavourably, and the lack of a clear date to work towards is a frequent source of psychological pain. As stated by Melanie Griffiths it removes a sense of purpose ‘… And without a maximum threshold or date of release, the waiting of immigration detention has no cumulative purpose. There is no goal that one is working towards or end point to which the clock is ticking down.’¹¹⁶ Although of course, the FNP will be released back into their country of origin, further work will be needed on how having this unspent sentence hanging over them will impact both their time in UK prison and their life afterwards.

Although the case law outlined near the beginning of this paper points towards deportation remaining characterised as a non-punitive measure, the case of Gurguchiani indicates a different route. The little-known ECtHR case, Gurguchiani v Spain,¹¹⁷ concerned an applicant who was sentenced to 18 months’ imprisonment for attempted burglary. The police requested that the judge responsible issue directions for the applicant's removal from the country and this request was accompanied by a government decision ordering the administrative removal of the applicant. The judge decided not to issue removal directions, however, as he thought that the enforcement of the initial sentence imposed by the judgment would be more appropriate. The state appealed and was successful; the applicant was deported and excluded from Spain for ten years. This deportation and exclusion order were based on a new law which provided that when an illegal immigrant in Spain was given a prison sentence of up to six years, there was an obligation to replace that sentence with deportation, save in exceptional cases. This legislation was not in force until a year after the applicant's initial trial. The applicant argued that this was retroactively applying a harsher penalty, in breach of Article 7 of the ECHR (which prohibits the retrospective application of criminal penalties). He was successful and the Court found that there had been a violation of Article 7. The Court considered the fact that the almost automatic application of the replacement of a prison sentence with a deportation order which took away judicial discretion and the imposition of a ten-year exclusion ban when the original penalty was an 18-month prison sentence, indicated that a harsher criminal penalty was being retrospectively applied. Interestingly, the applicant also argued that his rights under Article 6 had been breached. The Court found that Article 6 did apply here but, as it had already found a violation of Article 7, it decided not to examine the complaint under Article 6. Yet by saying that Article 6 applied, and by finding a violation of Article 7, the Court found that an immigration measure – deportation and exclusion – was tantamount to a criminal penalty. The rationale appeared to be that where the law provides for a deportation order to replace a prison sentence following a criminal conviction, the deportation will be considered a criminal punishment. Judge Zupancic dissented but only on the matter of compensation rather than any finding in principle,¹¹⁸ whereas Judges Myjer and Fura dissented on the ground that the prior cases of Maaouia and Uner should have been followed and no violation of Article 7 found.¹¹⁹

Is it likely that such an interpretation will be applied to the new section 47 provisions of the NBA 2022? There are some indications that, despite the clear overlap between the logic behind the Gurguchiani decision and what section 47 does, the courts will be reluctant to go down this route. It has been well-documented that the ECtHR is reluctant to interfere in this area – two significant studies of the migration case law of the ECtHR have found that the Court tends to give primacy to the legal principle that states have the right to control immigration over human rights considerations.¹²⁰ Marie Dembour refers to this as the ‘Strasbourg reversal’ or ‘state control principle’¹²¹ while Costello calls it the ‘statist assumption’.¹²² Some recent case law on migrants’ rights has emphasised this stance.¹²³

Secondly, the Court may be unlikely to take particular note of this issue because, as mentioned above, the Gurguchiani decision is not well-known. It is notable that the Gurguchiani case is in only available in French and Russian, ‘signifying that the Court does not attribute much importance to [it]’;¹²⁴ a lack of academic interest and subsequent citation has meant this case has been largely unnoticed despite its potentially transformative importance for the criminalisation of immigration. Thus, there is no guarantee that section 47 will create the impetus for re-evaluating the punitiveness of deportation and removal.

There are, however, indications that the Court would take a more robust stance. Although not a well-known or much discussed decision, the Gurguchiani case is included in the Guide on Article 6 (criminal limb) of the ECHR prepared by the Registry for the Court, which indicates that it may be regarded as important for this issue.¹²⁵ As mentioned above, not recognising the potential punitiveness of deportation has not been without controversy within the Court itself. Some judges strongly dissented with the majority judgment that the right to a fair trial did not apply to certain deportation cases¹²⁶ and there is some evidence that a few judges in the Court think that it is high time for this approach to be revisited.¹²⁷ Given that in recent years we have seen the ECtHR revisit controversial immigration judgments such as the N v United Kingdom ¹²⁸ decision, which was revised to provide much greater protection for seriously ill migrants facing deportation in Paposhvili v Belgium,¹²⁹ there is certainly a possibility that the ECtHR might shift its stance on this issue.

Conclusion

In conclusion, the current legal conception of deportation orders as non-punitive may be revisited in light of section 47 of the NBA 2022. This paper has demonstrated that the current legal interpretation rests on reasoning which seems circular, as it focuses on state classification of proceedings which seems incompatible with the Engel criteria. It has also shown that this legal position is already out of step with the findings of other disciplines such as criminology and sociology on how deportation is used and experienced. Section 47 represents a new chance to revisit this interpretation for at least some deportation orders. By pausing a prison sentence, it is difficult to see how a deportation remains part of the implementation of the original sentence and not a distinct punitive order which replaces a prison sentence. Yet, given the judiciary's historical deference on immigration issues, it remains to be seen whether this opportunity will be grasped, and the issue revisited.