The future of Scotland in the UK: does the remarkable popularity of the SNP make independence inevitable?

This post provides notes for my lecture on the 15th June at the Centre for British Studies, Humboldt University of Berlin, as part of the Ringvorlesung: The Future of the UK: Between Internal and External Divisions.

The advertised abstract reads:

The vote to remain in the UK, in the Scottish independence referendum in 2014, did not settle the matter. Nor did it harm the fortunes of the pro-independence party, the Scottish National Party. Instead, its popularity has risen remarkably, and major constitutional change remains high on the agenda, particularly during the run up to a referendum on the UK’s exit from the EU. This continued fascination with the constitution overshadows the day-to-day business of Scottish politics. Cairney highlights one aspect in particular: the tendency for limited public and parliamentary scrutiny of substantive policy issues when they are viewed through a constitutional (rather than substantive policy) lens, producing an image of weak accountability.

I’ll begin my talk by apologising passive-aggressively for not being a specialist in this field. I know some things about Scottish politics, but specialise in public policy rather than elections, referendums, social attitudes, and the future. On that basis, I’ll:

• Explain why the Scottish National Party’s popularity is remarkable
• Note that none of us have predicted it – or indeed much of the short history of devolution – too well, and use this point as a cautionary tale
• Describe why independence is not inevitable (even though it often seems likely)
• Shoehorn in some analysis of the links between our fascination with the constitution and the more humdrum world of actual policy.

The remarkable popularity of the SNP

The SNP’s popularity is remarkable in two main ways:

1. In 1999, the main party was Scottish Labour

Scottish Labour had dominated Westminster and local elections in Scotland for decades before the first Scottish Parliament election in 1999 (it also won a plurality of European Parliament seats, but with far lower margins):

• Westminster (plurality electoral system). Labour won most Scottish seats in every election from 1959-2010. In 1997, it won 46% of the vote and 56 (78%) of 72 Scottish Westminster seats (Cairney and McGarvey, 2013: 45). The SNP won 22% of the vote and 6 (8%) seats. A similar pattern continued until 2010: Labour dominated Scottish Westminster seats even when the SNP began to win Holyrood elections.
• Local elections (plurality until 2003, single transferable vote from 2007). In 1995, its 44% of the vote translated into 613 (53%) of 1155 seats and it remained the largest party until 2007 (Cairney and McGarvey, 2013: 51).

This dominance produced an expectation that Scottish Labour would become the largest party in the Scottish Parliament for the foreseeable future. In that context, the fortunes of Labour and the SNP changed remarkably quickly (see table 3). In 1999 and 2003, the main limit to Labour dominance was the electoral system: it won the majority of constituency seats comfortably but few regional seats (it also won most constituency seats in 2007). By 2011, this position had reversed and, by 2016, the regional list was the only thing standing between Scottish Labour and electoral oblivion.

In contrast, by 2011 the SNP achieved a majority of Scottish Parliament seats because the regional element of the mixed-member proportional system (56 of 129 seats) was not large enough to offset SNP dominance of constituency seats. This is a remarkable outcome if we accept the well-shared story that Holyrood’s electoral system was ‘chosen by Labour to stop the SNP ever the getting the majority it needed to push hard on the independence agenda’ (Cairney, 2011: 28).

2. The SNP’s popularity did not dip after the 2014 referendum

You could be forgiven for thinking that a No vote in the referendum on Scottish independence would damage the SNP. If it is a single issue party, and most voters rejected its position on the issue, wouldn’t you expect it to suffer? Yet, here is what happened instead:

• Its membership rocketed, from 25000 to 75000 in two weeks, then to 115000 by 2016
• It won 56 of 59 Westminster seats in Scotland (2015) on the back of 50% of the vote
• It won a third Holyrood election in a row, only missing out narrowly on a second majority in a row (in an MMP system).

It’s not so remarkable if you know that the SNP is not a single issue party. Instead, it is a highly professional organisation which has won elections on the back of valence politics as well as identity.

The SNP did well in 2007, and very well in 2011, because ‘most voters thought that the party would do a better job in office than its rivals’. People vote for a party when they respect its leader, its vision for the future, and have a high expectation of its competence while in office – and the SNP has benefited from being a party that looks highly professional (although one’s belief in the competence of the SNP may be linked strongly to one’s national identify and support for independence).

So, (a) it’s worth noting that the SNP is doing well partly because 45% of the vote will not win you a referendum, but it (plus a bit more) will do very nicely in a not-super-proportional election system, but (b) there is far more to the SNP’s story than a translation of national identity into support for independence into support for the SNP.

None of us predicted it well: a cautionary tale

You’ll always find someone who claims that they predicted these developments correctly, but that’s because of the immense number and range of hyperbolic predictions – from the claim that devolution provided a ‘stepping stone’ to independence, to the claim that it would kill nationalism ‘stone dead’ – rather than the predictability of politics.

So, for example, in retrospect we can say that devolution provided an important new platform for the SNP, but at the time we did not know that it would use this platform so effectively from the mid-2000s.

Similarly, maybe some people in the future will look back to argue that Scottish independence was inevitable, but without being able to predict the detailed mechanisms of decisions and events.

Scottish independence is not inevitable (even though it often seems likely)

I’ll try to sell you the idea that 10 years is the magic figure between Scottish referendums: a short enough distance to keep pro-independence actors content, and long enough to hope that enough people have changed their minds. We now have 5-yearly elections, so it would be a commitment in the 2021 Holyrood election to hold it in 2024.

However, it’s no more than an idea because nothing about this process is inevitable:

Even a second referendum is not inevitable

In the short term, the only event that matters is the ‘Brexit’ vote this month. If most UK voters choose to leave the European Union, and most voters in Scotland vote to remain, we will have a constitutional crisis. The SNP and its allies will push for a second referendum on Scottish independence, it will have the votes to pass a bill to that effect in the Scottish Parliament, and the main obstacle will be a UK Government led by the party that just used a referendum to justify major constitutional change! It is difficult to see why the Conservatives would bother to oppose a referendum under those circumstances.

In the absence of this constitutional crisis, is difficult to see how the SNP could justify – and, more importantly, expect to win – another referendum within five years of the first. This problem is reflected in the SNP’s manifesto and Sturgeon’s defence of its vague position. It appears to want to keep independence on the agenda for the long term without proposing a referendum within five years. So, its idea is that, in the absence of a Brexit crisis, the only other prompt is a major and sustained upswing in support for independence:

‘the Scottish Parliament should have the right to hold another referendum if there is clear and sustained evidence that independence has become the preferred option of a majority of the Scottish people’ (Scottish National Party, 2016: 23).

Sturgeon confirmed that this measure would be from opinion polls – ‘We would have to see, in a range of polls over a period of time, that independence had become the preferred option of the majority’ – but without stating how many polls, what level of support, or how sustained (BBC News, 2016b).

The unsatisfactory nature of this position seems reinforced by the SNP’s electoral position in 2016: the last referendum was fairly recent, it lacks a strong statement of intent in its manifesto, it now relies on the Scottish Greens (2016: 35) to produce a pro-independence majority in the Scottish Parliament, and the Greens’ trigger for a second referendum – a petition by maybe 100,000 voters – seems equally vague and problematic.

Don’t assume that we can predict the future (revisited)

These limitations are not problematic in the longer term: the SNP can afford to wait until the time is right for a second referendum. Yet, think about some of the other conditions that need to be met before a second vote is worthwhile:

1. There is a sustained rise in support for Scottish independence. From opinion polls, the SNP would be looking for levels of sustained support (high 50s, low 60s?) that are not present even in single surveys.
2. The SNP sustains a Holyrood majority, or it has enough pro-independence allies in Holyrood. At the heart of our short-term discussion is the assumption that the SNP continues to do remarkably well in elections. Yet, Scottish Labour provides a cautionary tale, or the evidence of how quickly a party’s support can disappear. Further, the SNP wouldn’t even have to lose much support to give the sense that it lost ‘momentum’.
3. The SNP can recreate 2011. Instead, it would just have to lose the sense of moral victory that it secured in 2011: the appearance of its Holyrood ‘avalanche’ election victory made a referendum difficult to oppose; its opposition lost ground, and the UK Government struggled to explain why it would not support a referendum. This feeling is difficult to recreate even if (as in 2016) it secures a similar proportion of votes.

Dissatisfaction with devolution is not the same as support for independence

It is possible that Scottish devolution will never seem like a ‘settlement’. Instead, we have had a routine process in which: (a) there is a proposed devolution settlement, (b) it sticks for a while, (c) there is a rise in support for independence or further devolution, (d) there is another settlement.

So far, this has happened in 1999 (the first modern settlement), from the SNP’s first Holyrood win in 2007 (producing the Scotland Act 2011), and during the referendum itself (producing the Scotland Act 2016).

The difference this time is the sense – often generated by supporters and opponents of independence – that the 2016 Act is the final offer. If so, we have two key scenarios:

1. This offer proves to be too unpopular to maintain support for devolution, there is a further referendum, and no-one can offer more devolution in exchange for a No vote.
2. The 2016 Act finally helps address the idea of a ‘democratic deficit’ in which (a) most people in Scotland vote for one party in a UK General election (usually Labour, now SNP) but get another (often Conservative), and (b) this problem helps produce the sense that the UK Government is imposing unpopular policies on Scotland. For the new Act to work, you would need to generate the widespread sense, among the public, that a Scottish Government could choose to mitigate the effects of a UK Government (perhaps without raising taxes).

What happens in the meantime? The humdrum world of scrutiny and policymaking

In the meantime, Scottish politics exhibits an unusual twist on the usual tale of Westminster politics:

1. We have the familiar disconnect between two understandings of politics, in which (a) we use elections and some parliamentary scrutiny to praise or blame governments, but also (b) recognise the limits to central control, which undermine a meaningful sense of accountability.
2. This confusion is complicated by devolution and ‘multi-level governance’ in which we are not always sure about which level of government is responsible for which policy.
3. It is complicated further by the 2016 Act, in which there are many new shared responsibilities between the Scottish and UK Governments.
4. So, politicians tell very different stories about what the Scottish Government can do, who is in charge, and who should take the blame for policy outcomes.
5. And the Scottish Parliament continues to struggle to know how best to try to hold the Scottish Government to account (and it might soon struggle a bit more).

Perhaps one possible exception is the new debate on educational attainment. First Minister Nicola Sturgeon staked a large part of her reputation on reducing the gap in attainment between students in the most and least deprived areas of Scotland. Before the election, she promised to ‘close the attainment gap completely’.

Although the SNP manifesto in 2016 presents more equivocal language (reflecting the sense that it does not know how much it can reduce the gap), it remains significant: as an issue in which there are constitutional complications (the Scottish Government does not control fully the economic and social security ‘levers’ affecting levels of deprivation), but the SNP is not using them to qualify its aims.

This example supplements several ongoing debates of high party political importance, in which there is not a constitutional element (on, for example, the Scottish Government’s ‘named person’ policy and legislation on ‘offensive behaviour’ in relation to football).

So, maybe such cases suggest that, for at least the next few years, we will pretend that there is a Scottish devolution settlement, and that we are not just killing time until the next referendum.