An indicative vote on the government’s Brexit deal has been suggested as a means of determining which of the options available to parliament has the best chance of securing the support of the House of Commons. In this post, Albert Weale examines how an indicative vote process would work, and whether or not it offers a workable solution to what appears to be a parliamentary impasse.
Pressure is growing for an indicative vote in the Commons to break the Brexit logjam. Such a vote would allow MPs to vote on a number of alternatives to the government’s ‘deal’, as laid out in the Withdrawal Agreement announced in November. The purpose of such a vote would be to see whether there was significant support in the Commons for each of the specified alternatives. A similar exercise was tried in 2003 when the then Labour government was seeking support for reform of the House of Lords, and in particular what balance of elected or appointed members a reformed upper chamber should contain. It did not work then, but could it work in the case of Brexit? Answering this question depends on three things: how many options are voted on, how the votes are counted, and the extent to which MPs engage in strategic voting. All three elements interact in complex ways.
To understand the basic logic, consider a simplified version of the various options that are likely to be proposed. With no abstentions, a majority on a motion in the Commons requires 320 votes to pass. In Figure 1, I have shown five possible motions that could be put to an indicative vote. Other things being equal, the more alternatives there are, the harder it is to obtain a majority for any one of them. With the guesstimated level of support for each option among MPs, the impossibility of getting a majority for any one of them is clearly illustrated in Figure 1. In the circumstances, estimates clearly have to be very rough, not least because many MPs have an incentive not to reveal their preferences too early. To avoid cluttering up the exposition, I have provided an appendix in which I lay out how I have derived the estimates.
How might an indicative vote work?
Suppose the Commons used the same method that it did in 2003 on House of Lords reform. The indicative voting procedure put eight options up for approval, ranging from outright abolition through different proportions of elected/nominated members to a completely elected chamber. When the Commons voted, each option gained only minority support (see here for a more detailed discussion).
Approval voting of the type used in 2003 works as follows. Any MP can vote for or against any option that is put to the House. Abstention is also allowed. Suppose first that there are no abstentions and that all MPs vote for the option that is closest to their ideal and against all other options. Then the votes in favour would be the same as in Figure 1. As Table 1 shows, there are substantial majorities against each of the options. These are so large that it is hard to argue that any one of the options forms a focal point for agreement. By way of comparison, the smallest majority against any of the Lords reform proposals in 2003 was three (for an 80% elected chamber).
But the assumption that MPs will vote straightforwardly for a single option is unlikely to be correct. MPs might vote for two or more alternatives: some will be genuinely indifferent as to which of their preferred options prevails, whilst others will vote for two or three in the hope of tactically supporting their preferred alternative. Approval voting allows each MP to vote for any alternative they favour. For example, an MP who is in favour of the government’s deal, but also views the Norway or Canada options as acceptable alternatives, could vote for all three. Similarly, someone who favoured a second referendum but could live with the Norway option would be free to vote for both.
Does the possibility of voting for more than one option offer the prospect of breaking the Commons logjam?
Given that a Commons majority requires 320 votes, this could be secured if enough of those in favour of either a second referendum or the Norway option voted in favour of the government’s deal. However, this is highly unlikely, as Labour MPs make up the vast majority of both blocs, and have a strategic incentive to see the government defeated in the hope of bringing about a general election. Moreover, within the second referendum group there are clearly many people (both Labour and otherwise) who think that Brexit ought to be reversed, and they have no incentive to indicate support for the deal. In short, the government’s deal is beyond the limits of approval for the vast majority of these two groups.
Is there an incentive for some members of these two blocs to vote tactically for each other’s preferred option?
Suppose that all the members of each bloc vote for a second referendum and for Norway. Then there would be a near majority (318 votes) for each of these alternatives. That might seem to suggest the value of an indicative vote. However, even if the estimates of first preferences are correct – and they have to be taken as guesstimates – there are enough MPs in each bloc who see a reason of principle for not supporting anything other than their first preference.
Do those on the hard Brexit side have a tactical incentive to vote for the government’s deal?
No. Hard Brexiteers have every incentive to vote against. Approval implicitly asks those voting to rate an alternative against a default option. Normally the default option is the status quo. In 2003, for example, this was an unreformed House of Lords. However, in the case of Brexit, the situation is quite different. The default is not the status quo (Remain) but No Deal – a position that hard Brexiteers either favour or think can be easily lived with.
With opinion divided over the five options, approval voting creates no incentive to compromise. Instead the incentive is to indicate as strongly as possible which alternative or alternatives you favour, voting against those you disapprove of. It is hard not to think that the Prime Minister understands this logic, at least at an intuitive level, which may be why she has resisted the idea of an indicative vote.
The Condorcet winner
A purely indicative vote provides no one with a reason to compromise and vote for something of which they disapprove because the alternative is even more unpalatable. The government’s strategy has been to try to force compromise by presenting MPs with a choice between the current deal and something that they would find even less attractive. For the Brexiteers, the alternative she offers is no Brexit, whilst Remainers are told it is the government’s deal or a hard Brexit.
In effect, this is to present the deal as what voting theorists refer to as a ‘Condorcet winner’. A Condorcet winner is an option that can beat every other alternative when pitched in a pair-wise contest against each of them. In terms of Figure 1, it is the option in the middle of the spectrum. It may be unloved by the overwhelming majority as a first choice, but it offers the most acceptable compromise when the alternatives are deeply unpalatable to a majority in the Commons.
The only way the government can force enough MPs to adopt the compromise is by making the vote not indicative, but real, which is what it has chosen to do on 15 January. The problem with this strategy is that it requires the government to persuade enough people that they really are faced with a seriously unpalatable alternative if they do not accept its deal. Given the advantage that the hard Brexiteers gain from the default being No Deal, they cannot be so persuaded. So, the government has to turn to those favouring a softer Brexit or no Brexit at all to support it. Given that the party composition of that group is overwhelmingly Labour, they cannot realistically be persuaded to accept the choice the government is seeking to force on them.
It was often said in the referendum campaign that Brexit would mean the restoration of UK parliamentary sovereignty. An indicative vote is the pretend version of that sovereignty. The vote on 15 January will be the real thing. If the government’s deal is voted down, it will show that it is possible to combine parliamentary sovereignty with parliamentary deadlock.
About the author
Appendix: How the Numbers Are Derived
There are 650 MPs but Sinn Fein MPs do not take their seats and so do not vote. There are currently four suspended Labour MPs, but they can be assumed to vote with Labour in any motion. There are four MPs in the Speaker’s Office, one of whom, the Speaker, ceases to be a member of a party, while the others retain their party allegiance. None in the Speaker’s Office vote in Commons divisions, but the one Conservative among them would have voted in the confidence vote. That leaves 639 voting MPs. The four tellers do not vote, but two come from each side, so it is easiest to treat them as part of a voting coalition. So, assume a figure of 320 for a majority with no abstentions.
Consider a situation in which there are five options to be voted on by the Commons:
- Second Referendum.
- Norway Option.
- The government’s deal.
- Canada Option.
- No Deal.
How can we estimate the number of MPs who would support each option? Putting numbers on those favouring each of these five options as their first choice has become a little easier since the vote of confidence in Mrs May’s leadership within the Conservative Party last month. That gives us a starting-point from which to estimate support for the government’s deal.
For the government’s deal: 194.
Assume that any Conservatives supporting the government’s deal come from the side that supported her in the no confidence motion. Then the maximum support for the government’s deal from the Conservatives is 200. However, from this we must subtract those Conservatives who have publicly declared for a second referendum (six at time of writing), plus those who in an indicative vote would approve of the Norway option in the alternative. A conservative estimate would put this at 20, split half and half between a second referendum and Norway. Support for the deal would then come from the two Independents (Frank Field and Lady Hermon) together with one Liberal Democrat and approximately 12 Labour MPs who are pro-Leave.
For No deal or Canada: 127
Assume that all Conservative MPs who voted against Theresa May in the recent confidence vote favour either a Canada model or No Deal, and split this vote half and half. To the 117 Conservatives, add 10 DUP members.
For Second Referendum: 143
Among Conservatives, we can count those who have already declared in favour of this option. I have allowed 10 in total, being half of the ‘pro-EU’ Conservative grouping. There are nearly 30 Labour MP’s who have declared in favour of a second referendum. If the vote is treated as indicative, then we can expect more to join them, particularly given the support of many Labour Party members for this option. At a conservative estimate that means one third of the parliamentary party. Then add members of the other parties.
For the Norway Option: 175
This is made up of the majority of Labour MPs, excluding those in favour of Brexit or a second referendum.
The detailed calculations are set out in Table 1.