Constitution or Con-substitution? What Scottish voters want from the next General Election

by Ailsa Henderson

Before Nicola Sturgeon’s resignation as Scottish National Party leader and First Minister of Scotland, the biggest debate inside the SNP surrounded the idea of treating the next UK General Election as a “de facto referendum”. This contemplated a simple electoral path to a future independence: if a majority of Scottish MPs supported independence (in practice, if they returned a majority of SNP MPs) then this would count as a mandate for Scottish independence and, in the party’s eyes, clear the way to negotiations. 

Treating an election as a referendum on a single issue is a risky strategy but the risks are of uneven sizes. We know from previous research that a nontrivial share of nominally anti-independence Scots backed the SNP at past elections, particularly at the devolved election in 2011 when roughly one third of SNP support came from pro-union voters. These voters backed the SNP for competence reasons, because they believed the party the better able to “stand up for Scotland” and better able to deliver policy achievements than their rivals.  

This kind of “valence politics” has declined in Scotland in recent years in large part because the 2014 independence referendum led to very strong alignment between people’s constitutional preferences and their party vote choice. In short, the SNP consolidated the support of pro-independence Scots and in the process shed its No-supporting valence voters. If one risk of a de facto referendum UK election is that it repels valence voters, then there are fewer of them now than in previous years – and this particular risk is not as large as it once would have been. 

Another risk is that treating any election as a referendum on independence in an effort to drum up support for one side of the constitutional debate could equally kickstart a process of strategic voting on the pro-union side of the electorate. If the goal of independence supporters is to maximize indy-supporting MPs, then pro-union voters might well interpret that their role is to seek to minimize them, and thus back whichever pro-union candidate stands the best chance of winning a constituency contest. While a degree of strategic voting has long been part of the Scottish electoral landscape, it was typically used to lock Conservatives out of seats. But the 2021 Scottish election also saw a degree of anti-SNP tactical voting.  

Treating the next UK General Election as a de facto referendum would likely exacerbate this, as our survey data demonstrates. In our February 2023 Scottish Opinion Monitor (Scoop) survey we asked how people would vote in the next UK general election, and then, much later in the survey, asked how they would vote if the election were run as a “de facto referendum” on independence. The results in Table 1 show little change in SNP support (since most Yes-supporters already vote SNP), but 14% of those intending to vote Labour would back the SNP in such a contest. On the pro-union side, 6% of Conservatives, 5% of Labour voters and 22% of Lib Dems would cast a ballot for a different pro-union party in a de facto referendum general election. It is worth pointing out that the figures might look different if a poll was taken today, but it provides an indication of how such a contest might affect different parties. 

Table 1 – Normal Westminster vote intention versus de facto vote intention (February 2023)

	General Election vote intention
“De facto” vote intention	Con	Lab	Lib Dem	SNP
Con	93	3	13	<1
Lab	6	71	9	1
Lib Dem	0	2	74	0
SNP	0	14	0	92
Other	0	3	2	2
Don’t know	1	8	2	4

Each of these risks would be present in any de facto contest but a third risk relates to the wider context of the next UK General Election. With the UK Labour party polling higher than before previous recent elections, Scottish voters of either constitutional disposition, might see in this a greater prospect of removing the Conservatives from office. Given longstanding antipathy to the Conservatives among a large portion of the Scottish electorate, as the prospect of unseating them increases, more Scots will face the choice of voting along constitutional lines or voting along partisan lines.  

To understand what might happen in such a context, we asked respondents in the last two Scoop surveys (February and June) and in an extra survey in May what their priorities were for the next UK General Election. The four main options were maximising pro-indy MPs, maximising pro-union MPs, keeping the Conservatives in office and removing the Conservatives from office. Figure 1 shows results from the whole sample in each of the surveys from earlier this year. 

Figure 1

If we look just at the constitutional preferences summarised in Figure 1, there is some evidence that a de facto referendum strategy would be at least as likely to animate indy opponents as it is adherents: a similar proportion want to maximize the presence of pro-union MPs as want to maximize pro-indy MPs. This is dwarfed, however, by the proportion of the Scottish electorate that wants to get rid of the Conservatives. A plurality of all respondents – and indeed a majority of those with a view – prioritise getting rid of the Conservative government above other considerations.  

This general question asked of everyone is useful but also can’t tell us how voters would react in the face of direct tradeoffs, the nature of which varies depending on which constitutional camp and party someone backs. We therefore asked additional questions posing these direct tradeoffs to Yes supporters and No supporters, the latter of whom we divided by party. 

Given the low propensity to vote score that SNP supporters give the Conservative party, a direct tradeoff would most likely involve a choice between backing a pro-indy MP or maximising the chances of removing the Conservatives from office by backing a different party that could form the next UK government – essentially, a choice between voting SNP or Labour. To No-supporting Conservatives, by contrast, the tradeoff would be between backing a pro-union MP from a different party – a Labour or Lib Dem candidate – or supporting the Conservatives candidate to help keep the party in power. To Labour and Lib Dem supporters, meanwhile, we posed a tradeoff between backing a pro-union MP, even if this means a Conservative candidate, and voting for a party best placed to remove the Conservatives from office.  

Figure 2

The results show that SNP-backing Yes supporters lean towards electing more pro-independence MPs than getting rid of the Conservative government. In all three samples, nearly half of Yes/SNP respondents selected this option. The gap between these options narrowed significantly in June, however, to just 3 percentage points. . While this might suggest Yes supporters prioritize UK government formation almost as much as constitutional politics, it is also possible that the two are perceived to be related: a Labour UK administration might be seen as more amendable to constitutional change than the incumbent Conservative government (though perhaps in spite of Labour’s official positions on the matter). 

There is also evidence that the strength of indy attachment influences priorities. Table 2 shows the breakdown on this question for those who identify as Yes supporters. For those who see themselves as “very strongly” Yes (this includes more than half of the Yes supporters in our datasets) there is a clear preference for maximizing pro-indy representation. While smaller in number, it is also clear that “fairly strong” Yes supporters would far prefer to get rid of the Conservatives than treat the 2019 UK General Election as a de facto referendum on independence (and there are too few ‘not very strongly’ Yes respondents to do them justice).   

Table 2 – De facto tradeoff by strength of Yes identity

	Very strongly Yes	Fairly strongly Yes
Maximize pro-indy MPs	53	13
Get rid of Conservatives	35	66
Don’t know	11	20
n	218	106

What about the opposing view? Are supporters of the Conservatives keener to save their party or the union? The results in Figure 2 show that No supporters who back the Conservatives in Scotland (there aren’t any Yes-supporting Conservatives in our samples) are less concerned about ensuring their party stays in office and more focused on maximizing the number of pro-union MPs, even if that means backing a party other than the Conservatives.  The gap was particularly stark in May 2023, when three quarters of those said they would prioritise constitutional preferences over partisan interests, and only nine percent said they would prioritise keeping the Conservatives in power. That low figure had increased substantially by June, to around a quarter, but this was still less than half the number who favoured maximising the representation of pro-union MPs. In this it appears that Conservative supporters are driven by constitutional motivations more than partisan ones. 

Figure 3

What about the other part of the pro-union electorate? Figure 3 provides more evidence that the Scottish electorate is unlike the English electorate: a plurality of Labour and Lib Dem supporters would be willing to maximimize pro-union MPs, even if that meant backing a Conservative candidate, over removing the Conservative government. Constitutional politics in Scotland has made bedfellows of parties that, in other parts of the UK, would consider themselves each other’s chief rivals.  

Figure 4

The figures vary by which party was backed in 2019 at the last general election. In June 2023, over half of pro-union 2019 Lib Dem voters (52%) wanted to maximize pro-union MPs compared to 29% wanting to remove the Conservatives from office. No-supporting 2019 Labour voters had a far narrower gap between the two options, with 35% wanting to prioritise pro-union MPs and 38% wanting to get rid of the Conservatives. The gap is similar but reversed among those who backed Labour at both the 2017 and 2019 elections (with 31% wanting to remove the Conservatives and 40% wanting to maximize pro-union MPs).  

All of this suggests that treating the next UK general election as a de facto referendum is likely to kickstart constitutionally-motivated voting on the pro-union side of the electorate. Now, what might actually happen likely lies somewhere between Table 1 (where the current polling fortunes of different UK parties were not primed) and the artificially blunt trade off put to respondents in Figures 2 to 4. There is also an asymmetry between No and Yes supporters – the latter appear more likely to prioritise removing the Conservatives than their constitutional preference, while the former, particularly Conservative supporters, prioritise their constitutional preferences. 

Taken in the round the data serve as a reminder of two salient features of Scottish voting. While we, rightly, have spent a lot of time discussing the impact of constitutional preferences on vote choice since the 2014 referendum, we should never underestimate the anti-Conservative negative partisanship of the Scottish electorate. There is some sign that constitutional politics introduce a level of noise here – one third of loyal Labour voters would prioritise electing a pro-union MP than unseating their main rivals for UK office – but it is also clear that a similar proportion of very strong Yes supporters would prioritise unseating the Conservatives over a de facto strategy. Whatever the merits or risks of a de facto referendum, it should not be separated from the wider context in which the next UK election will be held, or from long standing aspects of Scottish voting behaviour.