The election hasn’t officially started yet, but we’ve already had the first social media storm of the campaign, thanks to various people and organisations launching tactical voting sites, the recommendations of which have appeared to be somewhat…well, let’s be polite and say “idiosyncratic”, shall we?
I could use this post to explain the methodology of tactical voting sites, look at the rationales behind them, explain the polling that drives them, but I’m not going to do that, because thinking about these sites today has led me to a simple thesis:
Organised tactical voting in the 2019 election is going to be generally useless. For the last twenty years or so, we’ve had an assumption that there will be tactical voting in elections, and so an infrastructure has grown up around that assumption without stopping to think whether it’s actually useful or necessary.
To explain this more, we need to look more at the theory and history of tactical voting, so we can see why it (or at least, the nationally-organised and promoted form of it) is not going to be useful in this election.
There are lots of reasons why people vote (both why they vote at all and why they vote in the way they do when they get to the polling station) but for the purposes of this we can break them down into two broad groups: expressive and instrumental. An expressive voter is someone voting to, well, express something about themselves and represent their beliefs. They’re someone who’s not concerned with the result of the election but in making sure their particular view is expressed in it. An instrumental voter, on the other hand, is someone seeking to bring about a certain end and wants to use their vote to achieve that as best they can. These aren’t two mutually exclusive perspectives — think of them more as two ends of the spectrum where there may be a few people who are entirely one or the other but most are more those who have a tendency to one rather than the other.
Expressive voters are “I am X, therefore I shall vote for Y” types, where X and Y can be “working class” and “Labour”, “environmentalist” and “Green”, “Scottish” and “SNP” or a million other combinations, depending on the individual — and it’s entirely possible for people with the same X to vote for different Ys. Instrumental voters, on the other hand, are “I want X, therefore I will vote Y as they are the best way to achieve it” where the X and Y can be “lower taxes” and “Conservative”, “to stop Brexit” and “Liberal Democrat”, “independence for Wales” and “Plaid Cymru” or many other decisions — and again, two people with the same X can have different outcomes for Y. (Decision making is a complex process, and this is just a simplification)
So, where does tactical voting come into this? Tactical voting is about getting someone to vote for a party other than their preferred choice in order to achieve a bigger electoral goal, most commonly in order to defeat one party or candidate in particular. So, if you have a situation in a constituency where party A have 45% of the vote, party B has 40% of the vote, party C has the remaining 15% and supporters of party C tend to dislike party A more than they dislike party B then it makes sense for party B to encourage some party C supporters to back them in order to achieve the goal of defeating party A. And who are the voters most likely to respond to a call like that? The instrumental ones, who can be persuaded that “stopping A winning” is a worthwhile goal to vote for.
In this example, with one election happening in one seat, it’s a lot easier to make the case for tactical voting, and that’s why we’ve seen it much more often in by-elections than we have in general elections, because the arguments are much easier to make because there’s a lot less information to process before coming to a voting decision. By-election voters don’t have to worry about what results elsewhere might be, because they’re the only ones voting and they don’t really have to consider which party might be forming the government after the by-election because they usually don’t have the power to bring about a change in that. Voters in by-elections get an effective free hit in deciding what kind of message they might want to send to the government, and voters find it easier to switch parties when the consequences are lower so governments will tend to lose seats at by-elections. Voters who switch from Party C to party B generally don’t have to worry about harming Party C’s post-election chances or boosting party B’s at their expense.
However, if you’re voting in a general election, a whole lot of other considerations come into play, most importantly that you don’t know what the state of play is going to be everywhere else after you’ve voted. Before switching, a party C voter not only has to consider if they might be able to help Party B win in their constituency, but what effect that might have on the bigger picture. If party B is a large party, are they happy with giving them an extra seat that might put them closer to power? If they’re more minor, what might party B do in a hung Parliament? The more parties there are in the competition and the more issues of importance there are, the more complex it all becomes to decide how to best use your vote if you want to use it for a specific instrumental end.
What does all this mean in practice? That’s where the history comes in, especially the 1997 election and the build up to it. There’d been lots of tactical voting at by-elections before then, but those seats had a tendency to then be lost at the next general election when voters returned to their original voting patterns. The main reason for this is that in the 80s and early 90 the SDP-Liberal Alliance/Liberal Democrats had a policy of equidistance between the two main parties, not saying which one they’d back in a hung Parliament situation. This was partly tactical, but also a reflection of the party’s voters and supporters who tended to split evenly between Conservatives and Labour when asked which one they preferred. This meant the Alliance could make a good pitch to Labour voters in by-elections — “vote for us to give the Tories a bloody nose” — but in general elections, they couldn’t guarantee that similar voting wouldn’t lead to them supporting a Tory government so the potential tactical voters stayed with Labour.
So what changed? First Paddy Ashdown declared shortly after the 1992 election that the Liberal Democrats were committed to removing the Conservatives from power, thus abandoning equidistance and then Tony Blair became Labour leader in 1994, ushering in an era of strong co-operation between the two parties. By the time the 1997 election came around, it was clear that the two parties would work together if there was a hung Parliament, but also that the parties and leaders were mutually friendly. This gave a quite strong signal to Lib Dem and Labour voters that it was OK to tactically vote for the other party if they were better placed to win and defeat the Tories in their constituency. And in a lot of seats they did, especially where the challenger party was clear — there were several seats where one of the two parties did the bare minimum of campaigning, giving the other the opportunity to demonstrate it was the clear anti-Tory choice.
There’s a longer story to be told about how tactical voting develops and unwinds after that, but the important thing to remember is that while large scale tactical voting at general elections is possible, it needs more than just a few websites to make it happen. In 1997 there was also extensive co-operation between the parties and much signalling from leaders to voters that switching their votes was all right and that they would work together after the election. Even then, while the number of votes that shifted were critical in some constituencies, they were still small. What’s more, it was happening in a time when the political system was much more stable, with fewer electorally relevant parties in most seats, a less volatile electorate, and, perhaps crucially, a lot fewer sources of information.
There are two main problems with trying to bring about tactical voting in the traditional way in 2019. The first is at the party level, where there’s effectively no trust and no coordination between Labour and Lib Dems, so there won’t be any messaging, either implicit or explicit, from elites to voters that tactical switching would be OK. (There might be a limited Lib Dem/Green/Plaid Cymru electoral pact, but that sounds like it will actually involve parties stepping down in favour of each other, which is a step beyond tactical voting but would only be in a few seats) This is important, because why would you switch from your preferred party to another if the messaging from your first choice is that the other is full of antisemites/yellow Tories/secret Brexiters/neoliberals?
The second problem is that voters aren’t the same as they were in 1997 because the political situation isn’t the same. Then, the situation was simple and the election could be boiled down to “do you want the Tories or not?” Now, while the question is still important, it’s been joined by a barrage of others, some political like “how do you want Brexit resolved?”, “how do we deal with the climate emergency?” or “what’s the future of our economy?” but also others based on trust in parties and leaders. In 1997, tactical voting was based around a simple equation of if you want the Tories out, then you need to vote like this but in 2019 there’s no simple question that you can ask of everyone and give them a simple answer in response.
The wider problem this reveals with tactical voting is that it’s expecting lots of people to act like instrumental voters when they’re actually expressive ones and giving a process answer to a values question. Voters might not be attached to parties any more, but they are attached to causes and ideas, not just the ends that can come from voting. Yes, some are more likely to switch than others, but they’re switching because of messages and meaning, not because of levels of perceived effectiveness in their voting. If you want to persuade large numbers of people to switch their votes, you’re going to need to have an emotional message to persuade them, not a technical one. You could always start out by pointing out that if someone wants to get rid of rigged systems that hold the country back, the electoral system that could give Boris Johnson a clear majority with less than 40% of the vote would be a good place to start…