The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

Volokh Conspiracy

Why a New Brexit Referendum Would Not be a "Betrayal" of Democracy [updated with response to Ryan Bourne]

If referenda are a legitimate mechanism for making political decisions, then it is also legitimate for them to be overruled by new referenda. Those who live by the referendum sword risk dying by it.


Prime Minister Theresa May's Brexit deal with the European Union has proven so unpopular with both "Leave" and "Remain" supporters that it might well be rejected by Parliament. That has given new life to the "People's Vote" movement, which advocates a new Brexit referendum they hope will reverse the results of the June 2016 vote, in which a narrow 52% majority voted to leave the European Union. A second referendum (like the first) would have to be authorized by Parliament, and its results would not be binding without parliamentary approval (again, like the first). But such a referendum could potentially occur if Parliament rejects the May deal, and there is no parliamentary majority for any other approach to Brexit.

Many Brexit supporters, however, argue that a second referendum would be undemocratic, because it would represent a refusal to accept the results of the first. Prime Minister May has denounced the idea as a "gross betrayal of our democracy." British political commentator Tom Slater offers a more sophisticated version of that argument in an October article in Foreign Policy.

Brexit is a tremendously important issue in itself. And the broader question of when it is justifiable to reverse a referendum result with a new vote has implications that go beyond this specific example. Referenda are used to decide a wide range of policy issues, including many important state government policies right here in the US. If it is wrong to reverse Brexit with a new referendum, perhaps the same theory applies to other policies adopted by referendum, as well. But, in truth, there is no good reason to grant referendum results any special immunity from reversal—especially if that reversal is itself brought about by the same process as the original decision.

There is a longstanding debate on whether referenda should be used to decide important policy issues. Critics argue they should not, because voters often lack the knowledge needed to decide complex questions, or because such issues cannot be boiled down to a "yes or no" referendum question. Ironically, Margaret Thatcher—icon of British Euroskeptics—was one such opponent of referenda, which she denounced as "device[s] for dictators and demagogues." If you take a Thatcherite view of referenda, then Parliament should not only only reverse Brexit, but do so without holding a second referendum. That would set a strong precedent for the principle that referenda should not be used to decide important issues.

In my view, however, it is far from clear referenda really are worse tools for deciding policy issues than ordinary legislation. Voter ignorance is a very serious problem. But whether it is a worse problem with referenda or with ordinary legislation will vary from case to case. Thus, I am not, on principle, opposed to referenda. Whether Brexit was the sort of issue better decided by referendum or by ordinary legislation strikes me as a close case.

But if it was legitimate to use a referendum to decide in favor of Brexit, it is also legitimate to use the same process to reverse it. Indeed, one of the basic principles of legislation is that the same entity that enacted a law also has the power to modify or reverse it. US states with referendum processes regularly hold new referenda on initiatives intended to reverse or modify earlier ones. There is no reason to exempt Brexit from similar potential reversal. Indeed, the 2016 Brexit referendum was itself a reversal of the decision made by the 1975 UK referendum on joining the European Communities (which later became the European Union). If the 2016 Brexit referendum can reverse the result of the 1975 EC referendum, then it in turn can be reversed by yet another referendum. That's how legislation works.

Slater argues that reversing Brexit would be improper because it hasn't been fully implemented yet (Britain is not scheduled to depart the EU until March 29, 2019). But there is no reason why the body that made a legislative decision cannot reverse it prior to implementation. For example, congressional Republicans repeatedly tried to repeal Obamacare, including many times before key parts of it were ever implemented. That may or may not have been a good idea. But no one seriously objected on the grounds that pre-implementation reversal would somehow be an affront to democracy.

Pre-implementation reversal may actually be especially appropriate in cases where new evidence indicates that going ahead with the previous decision is likely to be more costly and dangerous than anticipated. The problems associated with the May deal raise that very prospect: the UK may not be able to leave the EU without either undergoing a dangerous "hard Brexit" (with no agreement on free trade and investment) or having to accept continued subjection to a variety of EU regulations, without having any say in their formulation (as per the May deal).

In Slater's view, a second Brexit vote would also be undemocratic because its main advocates are "elites" whose real goal is to reverse Brexit by any available means, not promote democracy for its own sake. He may well be right about these elites' motives. But so what? The vast majority of referendum initiatives are promoted by people whose main goal is to prevail on a particular policy issue. In most cases, they would be more than happy to achieve the same result by ordinary legislation, a judicial decision, or any other legally permissible means. That was in large part, true of the Brexit referendum itself, which was promoted by "Leave" enthusiasts because they want Britain out of the European Union (and by some Remainers who hoped that a referendum would put an end to "leave" agitation within the Conservative Party). Hard-core Brexiteers would have been happy to see Brexit enacted by Parliament, without a referendum. And had they lost in 2016, they would likely have been happy to see that decision reversed by a future referendum.

Finally, Slater and some other Brexit advocates argue that a new vote would be unfair to those who voted for Leave in the first referendum because it would "demoralize" them. That may be true. But, again, so what? The same can be said for any reversal of a policy decision much valued by its supporters. Many Remainers were deeply angered, perhaps even "demoralized," by the 2016 referendum's reversal of the 1975 vote to join the EC. Those who live by the sword of the referendum must risk dying by it, as well.

I don't deny that I myself would be happy to see Brexit reversed by a second referendum. I was opposed to Brexit in 2016, and developments since then have mostly reinforced that view. The issue is one that has divided my fellow libertarians on both sides of the Atlantic. But I tend to agree with those like Jacob Levy, and Johan Norberg, who argue that Brexit will result in more statism, not more liberty and more free market policies. It is increasingly clear that a post-Brexit Britain is likely to be less libertarian than the alternative, not more so. Sam Bowman, former executive director of the Adam Smith Institute, offers a thoughtful libertarian case for a second referendum, which I largely agree with.

Be that as it may, you don't have to agree with me about the merits of Brexit to recognize that there is nothing undemocratic in using one referendum to reverse the results of another. And that's true regardless of how soon the second vote comes after the first, regardless of whether it is promoted by "elites" who objected to the initial decision, and regardless of whether supporters of the earlier vote might be "demoralized" by reopening the question.

UPDATE: Ryan Bourne of the Cato Institute offers a thoughtful response to this post here:

A second referendum so soon would violate the U.K.'s convention of having one-off constitutional referendums, the results of which are respected for a generation. The U.K. has had major referendums in the past on remaining within the European Economic Community (1975), changing the general election voting system (2011), deciding whether Northern Ireland should join the Republic of Ireland (1973), and Scottish Independence (2014). The results of all these constitutional decisions have been implemented without discussion of the need to check again whether people really meant to vote as they did. In the case of the EU, the gap between the EEC vote and 2016 was 41 years.

I don't think a handful of cases like this is enough to create a legally or morally binding "convention." Moreover, Brexit differs from all of these examples, in that new evidence has quickly emerged suggesting that the initial referendum decision was based on flawed assumptions: in this case, that the UK could secure the benefits of free trade and investment flows with the EU without also having to accept various EU rules. The negotiations leading to the May deal strongly suggest that is not the case. Even if a "convention" of the sort Bourne points to does exist and does apply to the unusual situation created by Brexit, failing to follow it would still not be a "betrayal" of democracy, though perhaps it might be objectionable on other grounds. If the 2016 referendum result is overturned by a new referendum, it will simply be a case of a democratic decision being overturned by another decision of exactly the same kind. Doing so relatively quickly may be undesirable (though it might also be wise to overturn a flawed decision before it has had a chance to cause much harm). But it is not undemocratic.

Bourne also argues that libertarian critics of Brexit are wrong to oppose it on the grounds that it will undermine free migration, free trade, and economic liberty, because Brexit is just a constitutional decision on the allocation of power and does not mandate any particular policies on these issues, and a post-Brexit UK Britain will be "as pro- or anti-liberty as its people and politicians decide." That is true. But, for reasons explained in the posts by Johan Norberg, Jacob Levy, and Sam Bowman linked above, a post-Brexit UK is in fact likely to be less free than one that remains in the European Union. Constitutional decisions about the allocation of power should be made with an eye to how that power is likely to be used by its new holders. If they are likely to cause more harm than good, that's a good reason to deny it to them.

In the last part of his post, Bourne (like some other defenders of Brexit) analogizes Britain's vote to leave the EU to the American Revolution. The analogy actually cuts in the other direction. The American Revolution was arguably justified because it made America a freer society than it would have been under the continued rule of the British Empire. The revolutionaries famously defended their actions on the grounds that Britain was violating their rights to "life, liberty, and the pursuit of happiness," not on the theory that an ethnically or territorially defined group had a right secede for any reason they wanted. Had the revolutionaries sought to create a less free society than what existed under British rule, that would have been a good reason to oppose their actions. If libertarian critics of Brexit are correct, that is precisely what will happen if the UK continues on its current course.