As we face another day of parliamentary chaos over Brexit, talk of a further referendum has subsided. Without Labour throwing its full weight behind it, there are nowhere near enough MPs to give the idea traction. And yet parliament is so split it offers one of the few ways forward if the country is to avoid a no-deal Brexit. So it is still worth giving the idea a bit more thought.
The first question is what should be put to voters? Never mind the exact wording for now, but what would voters be asked to choose between? Referendums are almost always binary affairs (indeed I can’t think of an example of a three way one). Official Lib Dem policy is for the choice to be between the government’s deal, and calling Brexit off altogether. It is hard to see that getting through. Many Brexit supporters won't vote for the government's deal because they want a stronger break with the union (i.e. a hard Brexit), and this may well be the view of most of those that voted for Brexit first time round. Excluding this way forward from proceedings would seem like a denial of democracy and make the idea even more poisonous amongst the public than it already is. However a two way referendum between a hard and a soft Brexit would not get the support of Remainers- which would be needed to get the proposal through. A three way referendum between hard, soft and no Brexit looks the best compromise. A more complex set of choices could be devised, but Parliament is more broken than even I think it is if it cannot limit the options to three.
But what would the hard Brexit option be? The obvious one is leaving without a deal. This idea is gaining traction amongst the public at large, but many people think it is completely irresponsible to present it as a serious option. Even hard Brexit MPs like Boris Johnson advocate it as a negotiating tactic rather than as a coherent policy in itself. But these MPs also see the Irish backstop as an insuperable obstacle, and there looks little chance of the EU side taking it off the table - so no deal may be the only practical way of achieving hard Brexit in the shorter term. I find it very strange that hard Brexiteers on the one hand say that there are viable solutions to the Irish border issue outside the Single Market and/or customs union, and yet fear that the Irish backstop could make a customs union permanent. Still, they are calling the shots on this.
And what of what the of the soft Brexit option? If no-deal is to be put on the ballot paper, then the obvious candidate is the current deal on offer. Both options would then be capable of being implemented quickly (if implementation is the right word for a no-deal crash), and so we could resolve the whole thing by the summer, after which any delays cause a mounting political mess. Otherwise two yet to be finally negotiated options might be presented. Hard Brexit could be based on a Canada type long-term arrangement or no-deal if an acceptable deal could not be negotiated. Soft Brexit could be Norway plus or the current deal as fall back. This would clearly take time to sort out. It is exactly this sort of difficulty that has made many baulk at a referendum in the first place. But the easier ways out have no parliamentary majority.
So after that has been decided the next question is how to resolve a three-way choice. The first option, used to choose our MPs, is misleadingly referred to as first past the post (FPTP) (there is in fact no winning post: it is all relative). Voters have a single vote, plump for one of the three choices and the one with most votes wins. But that means the winning option may command barely a third of voters, and may be thoroughly disliked by a large majority. It would probably favour the Remain option, as the Brexit vote would be split. So Remain could win while a substantial majority, say over 60%, picked one of the Brexits. The problem with that is obvious, though the British public seems quite happy with it as a way of choosing MPs.
The obvious alternative is some variant of the Alternative Vote (AV). Voters would rank the options one, two and three (or actually one or two would be fine). If one option did not get 50% of first preferences, the votes of the third placed alternative would be redistributed to second preference. The problem with this method is that it disadvantages the compromise options that might be the least divisive. The compromise option, in this case Soft Brexit, is likely to come third. These voters would be forced to choose between two options each unpalatable to a majority. This method would give Hard Brexit its best chance.
There is another way forward, which is known as the Condorcet method. Voters have to make three choices between the three pairs of options: Remain/Hard Brexit; Remain/Soft Brexit; Soft Brexit/Hard Brexit. The option picking up the most cumulative votes wins. In this case a form of weighted preferential voting would amount nearly the same thing. Voters choose first and second preferences, as per AV, but the first preference would get 2 votes and the second 1. If voters refuse to make a second preference then their choice only gets a single vote. This isn’t quite the same as Condorcet, where you can vote once for each of the three options, but surely much more accessible. This approach gives compromise options, which attract lots of second preference votes, a much better chance than AV or FPTP. I’m not aware of it ever being used in referendums or elections, though it is familiar in other contexts. It is especially useful for ranking several options in order of preference.
To get an idea of the effects of the different methods, consider the case of an electorate of 100 voters. 45 want Remain, with Soft Brexit as second preference. 40 want Hard Brexit with Soft Brexit as second preference. And 15 want Soft Brexit with Hard Brexit as second choice. Under FPTP, Remain wins with 45 votes compared to Hard Brexit’s 40 and Soft’s 15. Under AV Soft Brexit’s 15 votes are redistributed to Hard Brexit, bringing them up to be winners with 55 votes. With the Condorcet/weighted preference system Remain gets 90 votes (2 from each of the Remainers, and none from elsewhere). Hard Brexit gets 80 votes from first preferences and another 15 from seconds: making 95. Soft Brexit gets just 30 from first preferences, but 45 second preference votes from Remainers and 40 from Hard Brexiteers, giving it a winning total of 115. Everybody has voted for it as first or second preference, making it a good compromise proposal.
Given the divisive nature of the Brexit debate, the weighted preference system looks the most appropriate. Alas AV would be much more likely. Traditionalists who might favour FPTP tend to be Brexiteers and they will not agree to anything that might favour Remain. Remainers tend to dislike FPTP anyway, and recognise that a victory would look illegitimate if it failed to secure 50%. The weighted preference system, on the other hand, looks way too innovative for us Brits.
The closer you look at the referendum idea, the messier it gets. But the strongest case for it is that it is a last resort, with parliament rejecting a no-deal, but not even close to agreeing any alternative. What a mess we are in!