The constitutional implications of a second referendum
The idea of a second EU referendum of some kind, though highly controversial, is now a prominent part of public debate in the UK. Whatever view one takes of the first vote, either of its merits as an exercise in democratic decision-making or of the result it produced, it is hard to argue that the UK political system, or indeed the country as a whole, was fully prepared for its consequences. At present, since the government is firmly opposed to a further referendum, the Civil Service is unable properly to prepare for the eventuality of a further popular vote. Parliament too, whatever the private views may be of majorities in both the House of Commons and the House of Lords, is not seriously examining what it might mean in practice. Part of the problem is that the debate over a second referendum is a proxy for the continuing division on the fundamental issue of EU membership. Even to countenance the possibility of a further vote could be depicted as a concession to the ‘remain’ side and its efforts to frustrate the popular will. Yet to rule out another referendum at this point would be as irresponsible as it was to assume that there would be no ‘leave’ majority on 23 June 2016. It is possible to conceive of circumstances in which the force for a second public vote becomes irresistible. If it does, implementation of this change of course will be far from straightforward. In particular, a series of constitutional issues arise, that it is proper to consider now.
A first matter to consider is the precise circumstance in which a decision to hold a second referendum might take place, if it does. When in the exit process would it take place? Before or after a final agreement was arrived at? Perhaps after a breakdown in talks, if one occurs? Might it follow evidence of a decisive swing in public opinion against Brexit, or the particular course that negotiations had taken (perhaps not even leading to a deal), or at least in favour of a second direct consultation with the public? Could some wider public emergency or external event act as a trigger? Might it come through an amendment to a bill secured because of a rebellion on the government side in the Commons? Or because a government decided to hold a referendum? Would it accompany the resignation of the present Prime Minister, and the forming of another Conservative administration, or one of another party or combination of parties, or combination of MPs from different parties? Might yet another General Election be involved? All of these possibilities would have important implications for the precise context in which the referendum, were it held, took place. They would be crucial to shaping the nature of the decision that the electorate faced, and the way in which it was perceived.
Second is the precise choice that might be on offer to the public. The possible permutations are numerous. If a binary referendum, would it be between a deal (were a divorce and free trade agreement arrangement reached) and remaining; or between no-deal (if talks do not succeed) and remaining; or between a deal or exit without a deal? Might one of the options on offer be to return to negotiations, perhaps to seek more palatable terms (or if no agreement had been reached, to give the government a new, more flexible, mandate)? If multiple options were presented to the public – presumably remaining, leaving with a deal, or leaving without a deal, and perhaps returning to negotiations – how would the result be calculated?
Third, attention should be given to who would vote and how the result would be interpreted. What would the franchise be – would it include 16 and 17 year olds, for instance? Or would there be immense pressure to use the same electorate as in 2016? Might there be a supermajority requirement or threshold of some kind? Would there be special consideration given to particular results in England, Northern Ireland, Scotland and Wales? Again, it might be difficult to justify rules that differed from those applied in the previous vote, particularly if it appeared broadly to be a restaging of the previous contest, offering a binary ‘remain’ or ‘leave’ option. Furthermore, were a supermajority required, to which option or options would it apply?
Fourth involves the management of the referendum. If the government made a recommendation to vote in a particular direction, how, precisely, would Civil Service impartiality during the pre-referendum period be enforced? Would an opt-out from collective responsibility for dissenting ministers be available, and if so how would it be managed? When, exactly, would the referendum take place? Given emerging suspicions, still shrouded in uncertainty, about campaign irregularities during the last referendum, how could the authorities, including the Electoral Commission, ensure the integrity of the next contest?
A fifth set of issues relates to the result and its outcome. Would the government possess the necessary legal authority to act upon the vote (such as it transpired was lacking after June 2016, when the executive ultimately needed to seek new statutory power from Parliament to trigger Article 50 of the Treaty on European Union)? Might we anticipate legal complications similar to those that manifested themselves with the Miller case? Would the legislation providing for the referendum also seek to make the result legally binding, or would it be technically advisory, as was the previous referendum? If the logic of a result was that the UK should seek to revoke Article 50, would it be able to do so unilaterally? Might an opinion on this subject need to be sought from the Court of Justice of the European Union in advance of a second referendum?
The strict legal issues will not be wholly decisive in themselves. Ultimately Parliament cannot bind itself anyway, and would probably only give the referendum result the force of law if the political environment was judged conducive to the passing of such a provision. After all, in theory, a UK government, backed by Parliament, could reverse the Brexit policy without any further referendum taking place. Yet opponents of leaving do not advocate this course. Their tactical decision to press for a second referendum arises because of a widely held judgement among political protagonists (and seemingly shared by a significant portion of the wider public). They believe that, whether they like it or not, and regardless of its formal status being only advisory, they are obliged to implement the ‘leave’ result, in some way.
Clearly, departure from the EU could potentially take place in a wide range of different ways. One possibility would be to maximise continuity, remaining within one or both of the Single Market (or parts of it) and the Customs Union, but exiting institutions such as the European Parliament, European Commission, European Court of Justice and so on. Some would regard this outcome as a travesty of the referendum result (notwithstanding the question on the ballot paper being silent on what ‘leave’ meant), but it would still arise from a basic desire to be seen to fulfil – albeit in a minimalist sense – the referendum result of June 2016. There remains a powerful political assumption that this referendum overrides any other type of democratic mandate. It is for this reason that supporters of leave are hostile to the idea of a second vote – the most serious potential threat to their project – while advocates of remain see it at present as the only plausible means of achieving their goal (assuming ‘remain’ is on the ballot paper in some way).
To hold a second EU referendum might be seen as a tacit acknowledgement not only that the first has failed to settle the issue of membership, but has created new problems. Given such an assessment of the use of this political mechanism, how can we be sure that a repetition of this exercise will improve matters, rather than making no difference, or perhaps even aggravating the position? A further vote that favoured leaving would presumably make this outcome inevitable, but might not end the associated political turmoil that has arisen, or settle the issue of membership forever.
If ‘remain’ were not made available as a possibility in a referendum, many would challenge its democratic credibility. Yet if to seek continued membership were offered to voters and won, opponents of this outcome – who already portray the second referendum idea as improper – might deny the legitimacy of the result, depicting it as the outcome of elite manipulation. Moreover, as we now know, even if a referendum choice is binary, there are many different kinds of result. Different divisions: between age cohorts, social groups and, perhaps most importantly, territories of the UK, all have an impact on the way in which the overall vote is perceived. So too do the overall margin of victory and the turnout. What would be the response, for instance, if ‘remain’ won by a larger percentage than had ‘leave’ in 2016, but with a smaller absolute vote than either ‘leave’ or ‘remain’ had received on this previous occasion? Any of these variables might have an influence on the potential for a second referendum, if held, not only to reverse (or confirm) the previous vote, but also to provide some kind of political resolution (though the policy complications that will follow in any eventuality would be great). The decisions made in relation to the various constitutional issues raised above, therefore, could matter greatly. It is advisable to begin closely considering the options now.
This publication presents the personal views of the author and not those of The Constitution Society, which publishes it as a contribution to debate on this important subject.