The Immigration Bill, which received its third reading in the House of Lords on Tuesday of this week (12/04/16), is a wide-ranging measure, running to 98 sections and 15 schedules ostensibly designed to make immigration control more effective. It included measures on, inter alia, the labour market, residential tenancies, bank accounts, driving licences and taxi licensing, enforcement powers of immigration officers, immigration detention and bail, and transfer of support for migrant children. The Immigration Bill and its precursor, the Immigration Act 2014, are of interest to constitutional lawyers for several reasons. This blog concentrates on one particular issue: the intrusions made by the Bill, and the 2014 Act which it amends, into the devolved competence of the Scottish Parliament (SP) and the implications for the delineation of the SP’s powers, the operation of the Sewel convention and the relationship between the UK government and devolved governments. It considers these issues in relation to the Bill’s provisions on residential tenancies and focuses in particular on Scottish devolution, although the implications extend also to Northern Ireland and Wales.
In our view, these provisions have such a large effect on devolved matters that the consent of the SP should have been sought.
Provisions on Residential Tenancies
One of the striking features of both the current Bill and the 2014 Act is the extent to which they seek to enlist private sector actors as agents of immigration control, although this not a completely new theme. The 2014 Act’s provisions on residential tenancies aimed to make it more difficult for those with an irregular immigration status to rent accommodation. It created a scheme whereby persons subject to immigration control either have a limited “right to rent” or are disqualified by virtue of their immigration status from renting accommodation, and landlords were forbidden to authorise persons to occupy premises under a residential tenancy agreement where they were disqualified by virtue of their immigration status. The Secretary of State was given power to impose financial penalties on landlords or their agents who contravened the prohibition. These provisions and indeed the entire Act apply without distinction to the whole of the UK, although thus far the residential tenancy provisions have only been brought into force in England.
The current Immigration Bill takes the aim of enlisting the aid of private sector landlords still further but takes a different approach to the 2014 Act in that there is separate provision for different parts of the UK. The principal measures relating to rented housing in the Bill are:
the creation of new offences of leasing premises;
new powers for landlords to terminate tenancy agreements; and
new powers for landlords to obtain possession.
Clause 41 of the Bill amends the 2014 Act by making it a criminal offence for the landlord to permit an adult to occupy premises in England under a residential tenancy agreement where (i) that person is disqualified from renting as a result of his/her immigration status, and (ii) the landlord knows or has reasonable cause to believe that the adult occupier is thus disqualified. In practice, this means that a person will be disqualified from renting unless s/he has leave to enter the UK or remain in the United Kingdom and the terms of that leave allow him/her to rent accommodation (2014 Act, s. 33A). The landlord also commits an offence where the tenant in question was not disqualified from renting when the tenancy was granted but becomes disqualified later, and there are further offences which may be committed by landlords’ agents (s. 33B).
Landlord’s power to terminate tenancy agreement
Clause 42 amends the 2014 Act to allow the landlord to terminate the tenancy agreement on grounds of immigration status. The landlord may terminate the agreement by giving 28 days’ notice where the Secretary of State has notified the landlord that the occupier is disqualified from renting as a result of their immigration status (s. 33D). Clause 42 also makes it an implied term of a residential tenancy agreement that the landlord may terminate the tenancy if the occupier is disqualified (s. 33E).
Landlord’s power to obtain possession
Clause 43 amends the Housing Act 1988 by adding a new mandatory ground for possession of rented accommodation in assured tenancies (Ground 7B). Where the landlord raises an action seeking possession, the court must order possession where (i) the occupier in question is disqualified as a result of their immigration status, and (ii) the Secretary of State has given the landlord notice to that effect.
These are far-reaching changes to landlord and tenant law, but, as noted above, they apply only to England and no clauses in the Bill make equivalent provision for the rest of the UK. Instead Clause 44 enables the Secretary of State by regulations “to make such provision” as s/he “considers appropriate for enabling any of the residential tenancies provisions to apply in relation to Wales, Scotland or Northern Ireland.” Clause 44 also states that the Secretary of State may by regulations make provision which “has a similar effect to any of the residential tenancies provisions” and applies in relation to Wales, Scotland or Northern Ireland. Such regulations may amend, repeal or revoke any enactment. This is, therefore, a cross-border Henry VIII clause which enables the Secretary of State to make important changes to criminal and private law in an area of policy which has generally been regarded as devolved (i.e. landlord and tenant law) with minimal parliamentary scrutiny and without the agreement of the devolved governments or the devolved assemblies.
We know that Scottish Ministers had concerns about the Bill that became the 2014 Act but do not know how far the constitutional points were pressed. We do know from an exchange of correspondence released by the UK Government on 18th March 2016 that the Scottish Housing Minister, Margaret Burgess MSP, was firmly of the view that the residential tenancy provisions of the Immigration Bill affected devolved matters and so required the consent of the SP. She urged James Brokenshire MP, Minister for Immigration, to amend the Bill so as to impose a duty to consult Scottish Ministers and to seek the Scottish Parliament’s consent before making any regulations to translate the residential tenancy provisions into Scots law. The same argument has been made by others including a coalition of Scottish housing bodies. James Brokenshire’s response was that, “as immigration is a reserved matter” it was not necessary to seek the Scottish Parliament’s consent. We agree with Margaret Burgess; these provisions do indeed concern devolved matters and the consent of the SP should have been sought.
The Constitutional Issues: Legislative Competence and the Sewel Convention
The Sewel convention is that the UK Parliament should not normally legislate with regard to devolved matters in Scotland without the consent of the Scottish Parliament. The convention is an essential feature of the devolution settlement as, despite the Scotland Act 1998 (SA), the UK Parliament retains authority to legislate for Scotland on any issue. If devolution is to achieve its primary purpose of giving the people of Scotland a large measure of self-government, it is necessary for the UK Parliament to refrain from exercising its undoubted legal power to legislate for Scotland on devolved matters in the normal course of events. It is, therefore, important that the UK Government and Parliament are scrupulous in their observation of the convention, and that whenever, they proceed without seeking the consent of the SP, that they have a sound constitutional basis for doing so which they are willing to present for public scrutiny.
However, deciding on the on the scope of application of the convention is not straightforward. It is in part a question of law because the distinction between devolved and reserved matters was created by, and is defined in the Scotland Act 1998. But it cannot be treated as solely a question of law, because as with any convention, its content depends in part upon political practice. It is necessary to take account of what the practice is and the content of the convention may change over time.
So, we will consider first the legal question of how the residential tenancy provisions of the Bill relate to the competence provisions of the SA, and then the broader question of whether the Sewel convention applies.
The SA contains a number of rules for deciding whether a matter falls within the competence of the SP or is exclusively within the competence of the UK Parliament. The primary rule is that a provision is beyond competence if it “relates to reserved matters” (s. 29(2) SA). That provision is amplified by s. 29(3) which states that:
"... the question whether a provision of an Act of the Scottish Parliament relates to a reserved matter is to be determined, subject to subsection (4), by reference to the purpose of the provision, having regard (among other things) to its effect in all the circumstances."
There was, for a time, some uncertainty about the general approach that would be taken to interpretation of the competence provisions in the devolution legislation. However, the Supreme Court appears to have settled on the following general principles: (i) the question whether a provision is outside competence must be determined according to the particular rules laid down in the legislation (in the case of the SP, these are s. 29 and Sch. 4). (ii) The description of the SA as an Act of great constitutional significance cannot be taken, in itself, to be a guide to its interpretation. It must be interpreted in the same way as any other statute. (iii) When enacting the SA Parliament had to define, necessarily in fairly general and abstract terms, permitted or prohibited areas of legislative activity. The aim was to achieve a constitutional settlement. It is proper to have regard to that purpose if help is needed as to what the words mean (see Imperial Tobacco Ltd v Lord Advocate 2013 SC (UKSC) 153; Attorney General v National Assembly for Wales Commission  1 AC 792). The Court has also made clear that for a provision to relate to a reserved matter, there must be more than a loose or consequential connection (see Martin v Most  UKSC 10, 2010 SC (UKSC) 40; Imperial Tobacco v Lord Advocate  UKSC 61, 2012 SC (UKSC) 153; In re Agricultural Sector (Wales) Bill  UKSC 43,  1 W.L.R. 2622).
In this case, the reserved matter relied on by the UK Government is immigration control. These provisions are not immigration measures in the conventional sense; they do not concern the substantive rules under which persons are admitted or refused entry to the UK or the administration of immigration control. But they are intended to make immigration control more effective. Clearly, the purpose of the provisions relates to the reserved matter of immigration control, but the question of competence cannot be decided by reference to purpose alone. Regard must also be had (among other things) to the effect of the provisions in all the circumstances. The effects of these provisions will be felt in the field of landlord and tenant law. They will change existing laws including a number which are currently contained in ASPs. It will be necessary for regulations made under clause 44 to amend a number of provisions of ASPs. These effects might also be described in terms of purpose, i.e. that the purpose of the measure is to change the rights and obligations of landlords and tenants in Scots law. On that view, the provisions would have two purposes.
The SA itself does not provide any general guidance on the relative importance to be attached to the purpose(s) and effect(s) of a measure. Lord Kerr suggested, in Martin v Most, that purpose was more important than effects: “It seems to me obvious that the way in which the subsection is structured signifies that the effect of the provision is subordinate to its purpose in the inquiry as to whether it relates to a reserved matter.” If that is correct, it supports the UK government’s position on the legal question. The dominant purpose appears to be to enhance the effectiveness of immigration control which is a wholly reserved matter. If instead, the UK Government had requested that the SP enact equivalent provisions, the SP would then be acting to enhance immigration control. The dominant purpose of the legislation would, therefore, relate to a reserved matter and the provisions would be beyond its competence.
But, accepting that only the UK Parliament has legal competence to enact measures equivalent to those in the Immigration Bill does not settle the question whether the Sewel Convention applies to the Bill.
When does the Sewel convention apply?
The UK Government’s views on when convention applies are set out in Devolution Guidance Note 10 (DGN 10). According to that note, “The convention applies when legislation makes provision specifically for a devolved purpose … ; it does not bite when legislation deals with devolved matters only incidentally to, or consequentially upon, provision made in relation to a reserved matter” (para. 1). That statement is amplified by para. 4 which states that the consent of the SP is required only where a Bill “contains provisions applying to Scotland and which are for devolved purposes, or which alter the legislative competence of the Parliament or the executive competence of the Scottish Ministers”, although where other Bills have effects on devolved matters UK departments are expected to consult the Scottish Government.
It is not clear how this applies to these provisions of the Immigration Bill. If para. 4 is taken literally, then the convention does not apply to them because (assuming we concentrate on the dominant purpose of making immigration control more effective) they do not have a devolved purpose. On the other hand, it is difficult to argue that the changes to the devolved matters of landlord and tenant law are merely incidental to, or consequential upon provision made for a reserved matter, especially if these terms are given the same meaning as in the SA. Lord Rodger commented in Martin v Most that these terms “suggest the kinds of relatively minor modifications which are often to be found in schedules, rather than cluttering up the body of a traditional UK statute” and which raise no separate issue of principle.
The problem is that the DGN 10 does not address the full range of situations that arise in considering legislation. In particular, it does not address the situation where a Bill relates to reserved matters and so is outside the competence of the SP, but also has major effects on devolved areas of law and policy, or the where a Bill has multiple purposes, some of which are devolved and others reserved. The problem is magnified where the Bill affects a number of devolved areas simultaneously. We have concentrated on the residential tenancy provisions but the Bill also has effects on local authority child protection and welfare functions, and licensing of private hire vehicles (all devolved areas).
So, the UK Government has plausible grounds for arguing that the Sewel convention does not apply to the residential tenancy provisions of the Immigration Bill but this leaves us in a very unsatisfactory situation. We have several concerns.
First, it seems incompatible with the spirit of the Sewel convention to enact legislation which has major effects on devolved areas of law and policy without any input from the SP. If it is accepted that the convention does not apply whenever the dominant purpose is a reserved one, that would potentially justify many other dubious incursions into the devolved sphere whether based on the immigration rationale or other reserved subject matters. Thus, a future UK Government might, propose legislation requiring Scottish education authorities to ensure that places at schools in their areas were not allocated to the children whose immigration status was precarious, or conferring power on the Home Secretary to direct the Scottish Police Authority to allocate greater resources to the enforcement of the Misuse of Drugs Act 1971.
Secondly, the method of law reform chosen is wholly inappropriate. The use of a Henry VIII clause – something that always carries a heavy burden of justification and which was criticised by the House of Lords Delegated Powers and Regulatory Reform Committee (17th Report of Session 2015‒16) – means that there has been full parliamentary scrutiny of the residential tenancy provisions as they apply to England but there will be only minimal parliamentary scrutiny of the equivalent measures applying to Scotland. The SP has not examined the Bill and the Sewel convention will not apply to any regulations made under Clause 44. To bypass the SP by a measure that also effectively bypasses the UK Parliament is to add constitutional insult to constitutional injury.
Thirdly, in addition, to the objection in principle to having differential levels of parliamentary scrutiny for different parts of the UK, the use of the Henry VIII clause creates problems of relative expertise. This is a complex and technical area of law and it is the Scottish Government and not the UK Government which has expertise in Scots landlord and tenant law, so the possibilities of ill-considered legislation which subsequently causes legal confusion are magnified by leaving changes to the law entirely in the hands of UK ministers. Whilst they have been and will continue to be in communication with the Scottish government, it is significant that the lead department will be a UK department with no in-house expertise in Scots landlord and tenant law.
The Future of the Sewel Convention
The UK government’s current approach to the Sewel convention is constitutionally unsound. Until recently, the operation of the convention had been relatively uncontroversial. As a result the details of the convention were not examined too closely. The Immigration Bill demonstrates that important questions had been left unresolved, notably whether the convention applies when a Bill clearly has a reserved purpose but also has substantial effects upon devolved matters. It also demonstrates that the UK government has too much freedom to determine the scope of application of the convention when there ought to be a shared understanding between the UK and the devolved parliaments and executives.
The Scotland Act 2016 provides further reason to reconsider the convention. It extends the competence of the SP substantially and makes one major tax – income tax – and one major public function – social security benefits – shared functions, potentially increasing the possibilities for conflict over the operation of the convention. The 2016 Act also put the Sewel convention on a statutory footing, amending s. 29 SA so that it states:
“(8) But it is recognised that the Parliament of the United Kingdom will not normally legislate with regard to devolved matters without the consent of the Scottish Parliament.”
The legal effects of this provision, and how far it will prove to be justiciable when it comes into force (23rd May 2016), are unclear but it can be seen as confirming the importance of the Sewel convention in the constitution.
Conventions are capable of evolution; the Sewel convention has already evolved so that it now covers Bills which alter the legislative competence of the Parliament. It would make sense for the Sewel convention to evolve further in order to keep pace with changing nature of the constitution as we move towards a quasi-federal legal order. We think that the convention ought to apply, not only where UK Bills have devolved purposes, but also where they have reserved purposes which have substantial effects on devolved areas of law and policy.
Tom Mullen is a Professor and Sarah Craig a Senior Lecturer at Glasgow University School of Law