During the Brexit referendum of 2016 prime
minister David Cameron was adamant that this was a one-off event,
not to be repeated. ‘I am absolutely clear a referendum is a referendum,
it’s a once in a generation, once in a lifetime opportunity
and the result determines the outcome . . . You can’t have neverendums,
you have referendums.’ !
So if Cameron ruled out a second referendum, how could another
prime minister allow one? The answer is that under the British
constitution no government can bind its successors. It is always up to
parliament to decide what happens next and new legislation can be
introduced to overturn previous decisions. There are no rules against
broken promises. !
That is also the problem – because there are no rules, there is also no
fixed agreement about how a second referendum would work. There
have been just three UK wide referendums in recent British history – on
continuing European Community membership in 1975, on changing the
voting system to AV in 2011, and on leaving the EU in 2016. In each
case the legislation setting out the terms of the referendum had to be
drafted from scratch. None of these referendums fixed how the next
one should be run. !
Some guidelines are in place. Under the terms of the Political Parties,
Elections and Referendums Act 2000 (PPERA), the Electoral Commission
has a statutory role to assess the question for fairness and
intelligibility. That doesn’t mean it picks the question – this is still the
government’s job. But it can say if it thinks the question is skewed one
way or the other. !
There are also time limits: PPERA specifies that the formal, regulated
period of a referendum must last 10 weeks, including the campaign.
Any referendum requires the passage of primary and secondary
legislation through parliament, setting out its terms. This takes time as
well. The Institute for Government has estimated that it would not be
!1 possible to get a 2nd Brexit referendum done,
from start to finish, in less than 20 weeks.
! Yet even these rules could be speeded up if
parliament were able to agree on new legislation. The difficulty is
that parliament would also have to decide on lots of other things too:
who gets to vote? Does it include 16 and 17 year-olds, as happened in
the Scottish independence referendum of 2014? How many questions on
the ballot paper – yes or no, or multiple choice? Is it a straight 50/50
fight or do the winners need to pass a higher threshold? Decisions,
But there are examples of second referendums from around the world
to draw on. For instance, in New Zealand it has become routine to
have 2nd referendums to confirm the results of initial votes. In a 1992
referendum on changing their voting system, voters were first asked if
they wanted to keep the current system or switch to something else.
Having voted for change, a second referendum in 1993 then asked
them to pick from a range of options, which included the existing
system. In the end, they picked a new system, called mixed-member
proportional representation. !
In 2015, New Zealanders were asked in a referendum which of five
possible designs for a new national flag they preferred. Having picked
one, they were asked in a second referendum if they preferred this
option to the existing flag. They chose to stick with the existing flag by
a margin of 57:43. !
These show how a 2nd Brexit referendum might work: either by giving
options for the future relationship, or by asking whether voters prefer
a particular deal (or no deal) to staying in the EU. But there is a big
difference. In the New Zealand examples the prospect of a second
referendum was factored in at the time of the first referendum.
Voters knew that their answer to the first question would be followed
by a chance to think again. They were under no illusions that the first
vote was a once in a generation chance. !
A second referendum on Brexit would have to face this obstacle:
because the electorate was told the first vote was a one-off, a second
!2 vote under any format might appear to be changing
the rules of the game. That poses another risk, which is that
disgruntled supporters of Brexit might boycott it. If you are going
to change the rules, they might say, we’re not playing any more. A
big majority to remain in the EU on a small turn-out, because Brexit supporters
stayed at home, would not solve anything.
! As things stand, there is a chicken-and-egg
problem. Parliament would need to agree on the terms of a second referendum,
which means working out all the detail about who votes,
when, on what and how. But if parliament were capable of agreeing
on that then we probably wouldn’t need a second referendum, because
parliament would also be able to agree on Brexit itself.
! It would be a lot easier if the UK had a constitution
that specified the rules governing referendums. But it doesn’t.
And the same chickenand- egg problem applies here too: a parliament
that could agree the terms for a new constitution probably wouldn’t
need a 2nd Brexit referendum to get it out of the mess it’s
The 2018 abortion referendum in Ireland, where a citizen’s assembly
helped set the terms of the debate and pick the question, showed that
there are other ways of doing people’s votes. But again the lesson is
clear: it worked in Ireland because it came before the controversial
vote, not after. It secured buy-in because the voters were told in
advance what they were buying. !I
f it’s going to settle anything, a second Brexit referendum would need
another vote first, in which the voters were asked whether they agree
to the new choice they are being offered. Under the UK system, there
is probably only one kind of vote that fits the bill. It’s called a general
election. It is hard to see how there could be a second referendum
unless a new government is elected that promised one in its manifesto.