Making ‘reserved powers’ work for Wales

Alan Trench says key issues around the implementation of a reserved powers model for Wales could be ducked.

The debate about a ‘reserved powers’ model for the National Assembly has to be one of the most obscure legal issues to enter public debate.  The basic idea is straightforward: that the powers of the National Assembly should be defined by setting out what it cannot do, rather than by defining ‘subject areas’ where it does have power to pass laws.  The idea is scarcely novel – it was mooted by theRichard Commission in 2004, and repeated by the Silk Commission in 2014 – but it has acquired political legs following the St David’s Day process with all the parties agreeing to adopt it.

A ‘reserved powers’ approach would offer a number of significant advantages.  It would mean that Welsh devolution works in a similar way to that in Scotland and Northern Ireland – important both symbolically and as a way of making it clearer to the public how a devolved UK works.  It also offers a way to resolve the puzzle created by the UK Supreme Court’s jurisprudence about devolution, and particularly its judgment in the reference about the NHS Recovery Of Medical Costs for Asbestos Diseases (Wales) Bill, [2015] UKSC 3, by enabling the Assembly to legislate for all matters save expressly those reserved to Westminster.  In broad terms, the Scotland Act 1998 provides a valuable model – not necessarily so much in the list of reserved matters in Schedule 5 as in the provisions of the Act for identifying the scope of those reservations and interpreting them in the courts.   At present, the list of proposed reserved matters suggests a list of matters Whitehall departments do not wish to see devolved, unsupported by any wider rationale or principle.  That is not the right way to proceed when drafting a constitution.  It needs some clearer and stronger basis, rooted in a conception of what the UK needs to do at the centre (and why), and what is best done by devolved governments.

The difficulty is that the reserved powers model is not straightforward to apply.  It raises a complex set of legal issues as well, beyond simply deciding what matters should be reserved and which should not.  (That issue is already causing a number of problems – as discussed earlier HERE.)  As part of an august group assembled by the Wales Governance Centre at Cardiff University and theConstitution Unit at UCL (and including Sir Paul Silk and Sir Stephen Laws, former First Parliamentary Counsel), I’ve been looking at these issues in some detail over the summer, and our report considering them has just been published.  It can be found here.

There are three key difficulties.  One is the UK Government’s proposal that criminal law and procedure be reserved.  The problem is that criminal offences are often created to ensure that ‘ordinary’ legislation is effective – creating the possibility of a fine or something more serious for breach of regulations, for example.  This could be dealt with by reserving the law relating to key criminal offences such as murder, theft or treason, but otherwise allowing the National Assembly to legislate for criminal matters as well.  That will lead to a significant shift in how the criminal law works in Wales, though, even if policing, the courts and the criminal justice system remain under Westminster’s control.

The second difficulty is the proposal to reserve civil law and civil procedure.  Under the present arrangements, the National Assembly has no direct powers in this area, but equally it has changed the rules considerably in this area through its existing powers.  Legislation like the new rules on organ donation involve a change in the law of agency to enable consent to donation to be deemed; the Renting Homes (Wales) bill makes significant changes to the law of landlord and tenant.  Again, the Assembly needs these sorts of powers in order to be able to make effective laws.  The civil law, like the criminal law, is not a subject – it is a mechanism used to achieve certain objectives, many of which are now devolved.  Remove powers over the civil or criminal law, and the result is a hobbled legislature that cannot in fact do its job.

The third difficulty relates to the legal nature of Wales, and the single legal jurisdiction of England and Wales.  Scotland and Northern Ireland have their own, separate, legal jurisdictions.   This means that laws passed by the Scottish Parliament or Northern Ireland Assembly have effect only within those jurisdictions.  Wales doesn’t, and that causes confusion.  Laws passed by the National Assembly are as much part of the law affecting Teignmouth as that for Tenby, and in some circumstances can have legal effect in England as well.  That needs to be regulated, in the interests of people in both England and Wales.  One can devise specific mechanism to regulate issues of applicability of laws and their enforcement, within the framework of a legal and court system shared between England and Wales.  Alternatively, one could establish a Welsh legal jurisdiction, distinct from that in England if not separate from it.  That raises the question of what a Welsh legal jurisdiction needs to involve or might involve.  (For an argument for a separate jurisdiction, the legal group Justice for Wales has just published a paper – see here for a news report.)

What worries us – the group of lawyers, former civil servants and advisers who drafted the report – is that these issues may be ducked in the desire to produce another devolution bill quickly.  (Indeed, the Wales Office’s initial reaction suggests that is exactly their intention.)  If these matters aren’t got right, the likelihood is another round of constitution-making for Wales in a few years’ time, during which the working of Welsh devolution will be hampered and messy, to the detriment of the people of Wales and indeed the whole of the UK.

Alan Trench is an academic, associated with the University of Ulster, the University of Edinburgh and the Constitution Unit at University College London. This post first appeared on his blog (

Also within Politics and Policy