Nick Barber: The Virtues of Advisory Referendums

The question of the capacity of the Scottish Parliament to hold an advisory referendum is one that has exercised contributors to this blog.  The dispute turns on a piece of statutory interpretation.  The law in question is found in section 29 of the Scotland Act 1998.  This states, in brief, that an Act of the Scottish Parliament is not law if it ‘relates to reserved matters’ (s.29(2)(b)), and this question must be answered by considering ‘the purpose of the provision, having regard (amongst other things) to its effect in all the circumstances’ (s.29(3)).  Turning to Schedule 5, we find that ‘the Union of the Kingdoms of Scotland and England’ falls within these reserved matters.

Those arguing that the Scottish Parliament does possess a power to hold a referendum contend that the referendum is advisory, and consequently would not ‘relate’ to the Union as the Union would continue, unaltered, after the vote. Those arguing the contrary view contend that the any referendum would, nevertheless, ‘relate’ to the Union as the future of the Union is the subject-matter of the vote.

This post does not revisit these textual questions, but instead reflects on the consequences for the Scottish Parliament if the Supreme Court found that it lacked the power to hold referendums of this type.  If the Scottish Parliament were denied the power to hold advisory referendums on matters that are reserved to Westminster, quite a large number of questions would be excluded from its jurisdiction.  To take a few examples from Schedule 5, a narrow interpretation of section 29 would bar advisory referendums on:

the state funding of political parties in Scotland;

the stationing of American troops on Scottish soil;

the legalisation of cannabis in Scotland;

the sale of hard-core pornography in Scotland;

the legality of animal experimentation in Scotland;

the construction of ‘mega-casinos’ in Scottish cities;

the building of nuclear power stations in Scotland;

the stationing of nuclear weapons on Scottish soil;

whether Scotland should be in the same time-zone as England.

One factor that the courts should consider when choosing between a broad and narrow reading of s.29 is whether there is any advantage in allowing the Scottish Parliament to hold referendums of this type.  If such referendums are useful, if they are a potentially valuable part of the devolution framework, this is an argument in favour of the broader interpretation.  If not, this is a consideration that weighs for a narrower reading.

Whilst the Scottish Parliament is a body of limited jurisdiction, and unable to unilaterally legislate on reserved matters, it does not follow that it is constitutionally barred from having an opinion on these matters, or even campaigning within the structures of the Union on these topics.  If, for example, MSPs were particularly concerned with the problems that arise from the criminalisation of cannabis in Scottish cities, it is hard to argue that it would be inappropriate for them to debate this question, order an inquiry into the issues raised, or even pass a resolution expressing their concerns.  Just because a matter is reserved does not mean that the Scottish Parliament cannot have its say.  The advisory referendum should be understood in this context.  It is a tool through which the Scottish Parliament can engage in a debate within the Union.  It enables the Scottish Parliament to ascertain the strength of feeling amongst the people of Scotland on an issue, and, by expressing that strength of feeling, place pressure on Westminster.  So, an advisory referendum on the legalisation of cannabis that showed support from the people of Scotland for a change in the law would provide a powerful card that could later be played in negotiations with Westminster.

Why should the legal structure of devolution allow an advisory referendum whilst denying the Scottish Parliament to alter the law in these areas?  Perhaps because they are areas in which any change – even a change that applied to Scotland but not the rest of the United Kingdom – would have implications beyond the borders of Scotland.  If Scotland legalised cannabis it would affect the availability of cannabis in England.  If Scotland banned nuclear weapons on its soil, it would change the configuration of the defence of the United Kingdom.  There are, then, good reasons why Scotland should be able to undertake ‘advisory’ referendums on these matters, but not enabled to decide them unilaterally.  The outcome of the subsequent negotiations with Westminster might not track the wishes expressed in the Scottish vote: the interests of all the peoples of the Union need to be considered.  Perhaps the end result of a referendum on cannabis would not be full decriminalisation.  Perhaps the product of the subsequent negotiation with Westminster would be that the drug could be prescribed in Scotland by doctors, or sold only by state bodies – or any one of a number of other compromise positions.

What about the arguments against giving the Scottish Parliament this power?  A case could be made that this is a waste of public resources, that referendums are expensive and time-consuming, and that the Scottish Parliament has other, cheaper, ways of making its feelings felt.  But these are considerations for the Scottish people, not for judges.  Using the constitutional device of a referendum brings with it two risks for the Scottish Parliament.  First, there is always the danger that the referendum will be lost, or that turn out will be so small that the point of the referendum is defeated.  A vote on the legalisation of cannabis in which a majority voted to keep the current prohibition in place would end the debate on the topic, at least for a time.  And if only a tiny portion of the electorate participated, a positive result would give the Scottish government very little leverage in Westminster: this would be a card of very low value.  Finally, of course, the Scottish Parliament is an elected body.  If the Scottish electorate decides that there are too many referendums being called, it can always vote out those who pushed for them.

Addendum:  Aileen McHarg has told me of an interesting parallel in local government law that provides a good illustration of the utility of advisory referendums. In 1994 Strathclyde Regional Council held a referendum over the privatization of Scottish water authorities.   The question of the privatization of water companies fell beyond the scope of the local authority; this was a decision for the national government.  There was a 71.5% turn out of which 97.2% voted against privatization – and water privatization was abandoned in Scotland.  The local authority was not empowered to decide the issue, but did have the power to use a referendum to articulate the views of its constituents and, as a result, to affect the decision of central government.

Andrew Le Sueur’s important recent post on the widening powers of local authorities might also be worth considering in this context.  The Localism Act 2011 accords local authorities very broad powers: these institutions can do anything that ‘individuals generally may do’.  This would seem to encompass holding referendums.  The restrictions of the Scotland Act do not – of course – apply to local authorities.  If a narrow reading of the Scotland Act were adopted by the Supreme Court, would this mean that Oxfordshire County Council could hold an advisory referendum on Scottish independence, but the Scottish Parliament could not?  This does not appear a very attractive conclusion.

Nick Barber is a Fellow of Trinity College, Oxford.