The agreement publicly reached between David Cameron and Alex Salmond for the holding of a Scottish referendum on independence in 2014 marks the end of a long, and unduly protracted, process. (There’s an account of the latter stages of that by Alan Cochrane of the Telegraph here which strikes me as well-informed if incomplete.) The agreement itself (with the draft section 30 order at the end) is here. The news story about it from Number 10 is here, and that from the Scottish Government is here.
The deal itself is a good and necessary one, if not particularly surprising in its content given the various leaks and rumours about it over the last few weeks. It is also one which delivers each government its key requirements, so in that sense it is a good deal for both sides. And, of course, it confirms that a referendum will indeed happen.
How we got here
It’s worth remembering how we got to this point. The SNP fought the 2007 election on a manifesto commitment to hold an independence referendum if elected, and to publish a white paper on independence before then. That commitment meant that a vote for the SNP would not necessarily be a vote for independence as such, which helped boost support for the SNP so it was able narrowly to win a plurality of votes and seats at that poll, because the election turned into one about ‘valence’ and competence not high-level ideology. In other words, the referendum pledge was a decisive condition precedent in enabling the SNP to make its electoral breakthrough in 2007. While the first SNP government framed a constitutional debate, in the form of the ‘National Conversation’ and delivered a white paper on independence (and other constitutional options), it failed to introduce a referendum bill – though it did publish a draft bill for consultation. That draft bill proposed a cumbersome and probably incomprehensible question, in an attempt to frame a question that might reasonably be argued to fall within the Scottish Parliament’s powers. Of course, with no majority at Holyrood, either for the SNP or for a referendum, any bill that was introduced would have been voted down. Nonetheless, and rather surprisingly, the Unionist parties made little of the failure in the 2011 election, and the SNP fought that vote with a more limited commitment only to introduce a referendum bill into Holyrood. The fact it won a majority meant that promise turned into something much harder – actually to hold a referendum.
With that, the problem of legal powers became a serious practical one. The Scotland Act 1998 clearly reserves to the UK the matters of the Union of Scotland and England, and the UK Parliament. Claims that Holyrood had the power on its own to hold an independence referendum either meant the question had to be framed so that its focus was on non-reserved matters by saying it related to the powers of the Scottish Parliament (as the question proposed in the 2010 consultation paper did), or required an argument that it was consultative and therefore not binding. At best, these arguments were dubious, and in practice would have doomed a Holyrood-called referendum to be the subject of messy and probably protracted litigation. At the same time, there was clearly a mandate for a referendum to be held – and for it to be called by Holyrood not Westminster (though some hardline Unionists seemed to believe it would be proper for Westminster to do so).
The obvious solution – which I was the first person to propose publicly, in June 2011 – was to use section 30 of the Scotland Act 1998 to confer powers on Holyrood to call the referendum. This was already being considered within each of the two governments, though it’s not clear whether they were in fact discussing it with each other at that point (my information was that they were not).
The Commons Scottish Affairs Committee also decided to get involved, launching its two-part inquiry into the referendum, looking both at the mechanics of a referendum and the implications of ‘separation’. As, due to the conduct of the chairman, the sole SNP member of that committee had withdrawn, that inquiry took on a highly partisan character.
The negotiations over the section 30 order gained further intensity following David Cameron’s intervention in January this year, which included publication of a draft section 30 order by the Scotland Office as part of the UK Government’s referendum consultation. That order substantially resembles what has now been agreed; a single referendum question, a poll regulated by the Electoral Commission, and using the local government franchise also used for Holyrood. (The main differences between this and the Westminster franchise are that this allows members of the House of Lords to vote, and also EU citizens living in Scotland.) The issue of when the poll should be held has been resolved more to the SNP’s liking than the Unionist parties’, with a closing date specified for the referendum of 31 December 2014. (That was left open in the January 2012 draft order.)
Both Scottish and UK Government ran their own consultation exercises during the late winter and spring, with the UK Government’s closing in March and attracting some 3,000 responses, and the Scottish Government’s closing in May and receiving some 26,000. While the UK Government hastened to publish its findings of the responses, that has not been the case for the Scottish Government which has still not published its findings despite committing to do so ‘by the end of the summer’. That is not just interesting in itself (though 26,000 responses are a lot to consider), but shows that the Scottish Government has not waited for a detailed analysis of the responses before pressing on with resolving the section 30 order. The process of negotiating the order’s provisions has evidently been protracted and exhausting for those involved, with plenty of stories and counter-rumours going around since the spring and particularly since August. It is unclear how much of that time has been taken by the Scottish Government pushing the idea of a ‘second question’ setting out an option for enhanced devolution, which they appeared to have abandoned by early September at the latest. Serious pursuit of that option would have been the most likely show-stopper for a section 30 order, as the Unionist parties were adamant from very early on that they would not concede such an option.
The winner from that protracted process has to be the SNP and YesScotland. The extended nature of the debate has kept the issue alive, and created a narrative of the Scottish Government ‘struggling’ to obtain the UK Government ‘concessions’ to running the referendum on its terms, so it can claim a ‘made in Scotland’ referendum (constitutional autochthony, for scholarly readers). In reality, the concessions divide pretty equally: the SNP on the second question and the role of the Electoral Commission including on the intelligibility of the question, the UK Government on timing and franchise.
Similarly, in reality the argument about the referendum being ‘made in Scotland’ was powerful rhetorically but always impossible legally. If the Scottish Government had sought to hold a poll on the basis of its own claimed powers, there is no doubt that the matter would have gone to the courts and my view, shared by many if not by all lawyers, is that they would have lost. Actually holding a referendum depended on the section 30 order. The main concern on the UK Government side was that the SNP were keen to duck out of its commitment to hold one, and would use the ‘refusal’ of the UK Government to facilitate that it in order to do so. The door to that is now emphatically closed. It has been telling how keen Nicola Sturgeon was to trumpet that what has been agreed delivers a ‘made in Scotland’ referendum; despite the compromise nature of the package, as the Scottish Government’s minister for constitutional matters, she is willing to accept ownership of the deal and the referendum that results. Salmond has also followed suit. That will prevent cries of ‘foul’ about the terms of the referendum if independence is defeated, and means that it will be very hard for the SNP to revive the flag of independence in that eventuality for a long time to come. The SNP have likewise won a symbolic game, by ensuring that Salmond only met Cameron, and leaving more junior ministers in each government to deal with the details.
The referendum deal
It’s important to note how the package is structured. The order itself is quite short and sweet, ‘unreserving’ a referendum on the subject of reserved matters (including the Union and the UK Parliament) provided it is held by the cut-off date of 31 December 2014, and consists of only a single ballot paper with a single question. By barring another poll on the same day, the order prevents the Scottish Parliament putting the ‘third option’ in play by holding a second referendum poll on the same day, on some sort of enhanced devolution versus another option. (Such a poll would, in my view, otherwise be lawful; there is no bar in the Scotland Act 1998 to discussion of further devolution, including by a referendum, so long as the legislation passed by Holyrood ‘relates to’ matters within devolved competence including the powers of the Scottish Parliament or Government. It is the Union and the role of the UK Parliament that are reserved.)
There is also no date before which the referendum must be held, only a closing date for it. Legally, it would be open to the Scottish Parliament to hold the referendum before 2014 if Holyrood wished, though that would not only breach the agreement just reached with the UK Government but also run against everything the SNP have said on the subject since January 2012.
As the order is silent on the question of who can vote, and as the memorandum notes, it will be open to the Scottish Parliament to decide whether to include 16 and 17 year olds in the franchise for the referendum. Clearly there are political attractions in doing so, both because of the assumed preferences of those voters and because of the act of constitutional self-assertion in enabling them to vote. However, the logistical problems of doing so are considerable, and that commitment may prove rather a poisoned chalice.
Some key issues – about the role of the Electoral Commission, and the application of the provisions of the Political Parties, Elections and Referendums Act 2000 about time periods, spending limits and participant organisations in the referendum – are not set out in the order, just in the agreement between governments. These will fall to be included in the referendum bill when that is passed by Holyrood. The guarantee on this point is the Scottish Government’s political commitment, rather than any legal safeguard.
Potentially the most far-reaching provisions of the agreement are those relating to co-operation between the two governments, which ‘have agreed to work together to ensure that a referendum on Scottish independence can take place’. They also agree, in its last paragraph, that
The two governments are committed to continue to work together constructively in the light of the outcome, whatever it is, in the best interests of the people of Scotland and of the rest of the United Kingdom.
What that actually means, and whether it is more than warm words, will become clearer in the months to come. It is certainly unusual, comparatively speaking, for two governments in such a tussle to be able to agree the fundamental ground rules for such a high-stakes vote.
What happens next
The section 30 will need formally to be laid by the Secretary of State as a draft before the UK and Scottish Parliaments. It needs to be approved by in both the Commons and Lords at Westminster, and at Holyrood, seemingly (so far as Westminster is concerned) within 40 (calendar) days of the order being laid. That will create the legal basis for Holyrood to legislate for the referendum. The suggestion is that will happen by February, but it could be a good deal sooner than that. Both Holyrood and Westminster are currently sitting, and will be for some time – laying an order over the summer or early autumn would have run into the problem of one parliament sitting but not the other, affecting the working of the statutory time limits for its approval.
The steps after that are the preparation of the ballot paper and testing of the question, and the passing of a referendum bill at Holyrood. According to the Scottish Government’s proposed timeline, preparing the ballot paper is to take place during the autumn and winter of 2012, with the bill to be introduced in ‘early’ 2013 and to be passed next October.
It’s worth noting that the agreement outlines in some detail what will need to be in the referendum bill. If Holyrood were to enact legislation that failed to comply with those provisions, there could be serious problems. That in turn limits the scope for the legislature to amend legislation before it, because the legislation has to conform to the intergovernmental agreement; in many key respects, the referendum bill will be un-amendable. This is familiar from similar cases elsewhere; the ‘lending’ of legislative power actually limits the power of the legislature to which power is lent. (This is a well-known problem in Australia, with the working of section 51 (xxxvii) of the Australian constitution by which state parliaments can refer matters within their exclusive competence to the federal parliament.) Given the SNP’s Holyrood majority and internal discipline, this is unlikely to be a practical problem, but it is an important issue of principle that can have practical implications when parliaments seek to assert their prerogatives.
The big loser: the Scottish people
While both the Scottish and UK Governments can claim to have won from the overall referendum deal, there is one big loser: the Scottish public. There is abundant evidence (such as that set out in the recent 29th British Social Attitudes report, available here) that the Scottish public’s preferred choice is greater self-government within the Union, rather than independence or the status quo, even with the Scotland Act 2012. The debate has been portrayed as about a ‘devo max’ option as some sort of consolation for the SNP, or about the absence of an ‘enhanced devolution’ option as meaning that the referendum cannot reflect the true wishes of Scottish voters. That is all true. While all three unionist parties are now willing to consider further devolution, no-knows what that might involve, with a proliferation of different plans. The Lib Dem Home Rule Commission will report later this week. The Devo Plus group has published reports on finance and social welfare, with a third due shortly. However, Labour only just formally established its own commission, and it remains unclear whether the Conservative party will have a formal review of its own. The clearest indication of the Conservative position appears to be Ruth Davidson’s suggestion (apparently endorsed by David Cameron) of a full-scale constitutional convention if there is a No vote to independence. Such a convention would be a vast undertaking, hugely problematic if not carefully prepared – and potentially a source of delay in delivering any meaningful change. There was no clear ‘enhanced devolution’ option to put to the voters at a referendum in 2014, even if there were the political will – and there never was that either.
But the Unionist parties’ strategy – in effect, an excluded-middle strategy – is a high-risk one (as I argued in June 2011 HERE, and also in this talk from March). In fact, there are two risks. One is that voters seeking their desired choice of ‘greater devolution’ conclude that independence – especially the ‘lite’ version likely to be proposed by the Scottish Government – is closer to what they want than the status quo that appears on proposed by the Unionist parties. To an extent, that appears already to be happening. The other is that, by failing to deliver what Scottish voters want, the Unionist parties undermine the legitimacy of government in Scotland more broadly. (That, I think, is what James Mitchell from Strathclyde is getting at on the BBC here, though his is an odd way of putting it.) Proposals like Cameron’s may be even more dangerous, as at best they may act as causes of delay and at worst will be seen as warning of a repetition of obstacles put in the way of self-government by both Conservative and Labour parties in past.
The Unionist parties are clearly reluctant to set out any sort of agreed ‘enhanced devolution’ scheme at this stage, partly because they think it would muddy the waters of their referendum strategy and partly because there is no agreement between them about what it would involve. Keeping the ‘enhanced devolution’ option out of the debate during this period risks causing a serious dis-service to the Scottish people by forcing them to talk about two options which few people want, and not that wanted by a plurality if not a majority. The political tactics of the referendum may get in the way of a democratically justified outcome.
Moreover, it may be bad tactics for supporters of the Union. Faced with a Yes campaign that can tell a positive story, putting that sort of option on the table sooner rather than later would offer significant advantages. As I suggested HERE, it would be in fact be the cautious course to ensuring that the referendum can be won, and the least damage done to the standing of government in Scotland.