A recent press story suggested that supporters of the ‘Out’ side in the upcoming referendum are already planning to argue, in the event of an ‘In’ vote, that a further referendum must be held again within a few years. At first sight, it looks as if at least these ‘Out’ supporters expect to lose the referendum, and are planning to be sore losers at that. I wonder if that is a helpful message for them to send – but then it’s not my role to advise the ‘Out’ side on strategy.
But is it intrinsically outrageous to suggest that there should be a further referendum on the issue? The ‘Out’ side can reasonably argue that they are only copying the strategy of pro-EU politicians, who pushed for fresh votes in Denmark and Ireland after three referenda in those states voted down the Treaties of Maastricht, Nice and Lisbon, and repackaged the Constitutional Treaty after it was defeated in the French and Dutch referenda. In principle, that’s a fair point to make: but if the ‘Out’ side are going to make it, they can no longer criticise the pro-EU side for being ‘undemocratic’ when it pushed for repeat referenda. In fact, the anti-EU side already have form on this issue themselves – since they went back to the Czech Constitutional Court to challenge the Lisbon Treaty a second time, when they didn’t like the first court ruling. The judges weren’t impressed.
Repeated referenda are not unheard of in other contexts or countries: Ireland has had several referenda on divorce and abortion, and Quebec has twice voted on separation. Many Scottish nationalists also aspire to a second independence referendum in the foreseeable future. So I don’t think one can simply argue that referenda should never be repeated. But I do think that they should only be held in certain circumstances.
What circumstances are these? In my view, there are two: a significant change in circumstances, and the conditional nature of the vote. These two criteria may well be present at the same time. The first of these criteria justifies having a new vote on the UK’s membership now, given the changes that have occurred since the last vote in 1975: five major Treaty amendments and five enlargements of the EU, leading to much greater ‘immigration’ of EU citizens to the UK. The second Quebec separation vote was justified on the basis of the second criterion: Quebeckers voted ‘no’ to independence in 1980 on the premise that the Canadian constitution would be amended to address their grievances. Two attempts to agree such amendments then failed, and that was the rationale for having another independence vote in 1995. Some Scots argue similarly that promises of greater devolution to Scotland made in the 2014 independence referendum are being broken; and many Scots believe that if the UK votes to leave the EU while Scotland votes to stay, the first criterion would be satisfied.
How do we apply these criteria to the EU referenda? In all the cases where a second referendum was held, there were intervening changes in circumstances. There were Decisions and Declarations by the EU which directly addressed those concerns of the Irish and Danish voters who had voted ‘No’. The Constitutional Treaty was scrapped as such, and replaced by the Lisbon Treaty, which contained most of the same substantive text, but without the ‘constitutional’ trappings of the failed treaty which had outraged the anti-EU side so much at the time.
Applying these rules to the UK’s planned EU referendum, there would be a case for a new vote if there were intervening changes in circumstances, which directly refuted the basis for voting ‘In’. Equally there would be a case for a fresh vote, if most or all of the case for the ‘In’ side was based on conditional promises of EU reform which then did not take place.
Of course, two can play this game. It must equally follow that in the event of an Out vote, there would be an argument for holding a fresh referendum on the basis of the same two criteria. Article 50 TEU, the ‘withdrawal’ clause, expressly provides that a State can apply to rejoin the EU if it leaves. It’s also arguable that a State can cancel its withdrawal request, although Article 50 is not clear on this point either way. Certainly it’s possible to suspend a withdrawal request de facto, by delaying the withdrawal date indefinitely. (See further my discussion of Article 50 here).
What would this mean in practice? Applying the first criterion, it’s possible that the remaining EU would be upset at the prospect of UK withdrawal so much that it offered a new renegotiation package before it happened. Or imagine that in the longer term, the EU changed profoundly, allowing for more restrictions on the movement of people and a greater number of vetoes for national governments and parliaments. There would then be a good case for holding a vote on rejoining. Applying the second criterion, there would be a case for a second referendum if the conditions set by the Out side were not satisfied in practice: for instance, if there were no satisfactory trade deals with the remaining EU and many third States, or if the UK still had to pay a price in return for trade access (contributions to the EU budget, acceptance of EU regulation, the full free movement of people) which the Out side had claimed that it would not have to pay.
Art credit: M.C. Escher