Labour’s current
policy on Brexit is designed to help them win power. There is nothing
the government and their press would like more than to suggest
Starmer intends to undo Brexit, and so the policy of “making [hard]
Brexit work” is tailored to remove
any credibility from such a claim. However the moment
Labour wins power other considerations come into play.
In power Labour
becomes responsible for the health of the economy, and Brexit has
undoubtedly brought severe damage to the economy and average incomes.
Starmer knows this, which is perhaps why he added “at this stage”
when saying
joining the Single Market would not help growth. In power he will not
be able to avoid two clear truths. The first is that, Northern
Ireland Protocol apart, the economic effects of being more
cooperative with Europe within the context of Johnson’s hard Brexit
are
small. The second is that the gains from joining the
EU’s customs union and, even more so, their single market are
certainly not small, just as the costs of leaving both have
not been small.
The dangers of
alienating those voters who still identify with Brexit will remain, or more
generally worrying those voters who fear that changing the Brexit deal in a
fundamental way would paralyse the government just as the Brexit
referendum did. However the size of the first group will
diminish over time, in part because
of simple demographics. The size of the second group will also
diminish if Labour manages to handle negotiations with the EU over
more minor matters with little fuss.
In addition, as
Stephen Bush notes,
“a Labour government would free up vast swaths of civic society and
lobbying organisations, many of which think that Brexit is a disaster
but feel they have to pussyfoot around the topic to remain on Downing
Street’s Christmas card list”. I noted two
weeks ago that we had passed a turning point, where
the costs of Brexit had become so evident that the broadcast media
felt compelled to start talking about them. With the focus on a
Labour rather than Conservative government, there will be fewer
voices defending Johnson’s Brexit deal and many more pointing out
its problems.
Indeed, it seems
likely that one of the last groups to change their mind on Brexit
will be the Conservative party. This, combined with this trend in
public opinion away from either believing in Brexit or fearing its
modification, means that at some point support for Johnson’s Brexit
deal (or something even harder) will become an electoral liability
for the Conservatives. As it becomes clearer that Brexit has reduced
living standards and held the economy back, an attachment to the
policy will be associated with a party that wants to keep the country
poorer.
There is therefore a tipping
point in public opinion, when suggesting Johnson’s hard Brexit deal
needs to be thrown in the dustbin of history no longer becomes a
political liability but a political necessity for Labour. Some may
think we have already reached that tipping point, but two key factors
suggest otherwise. The first is that FPTP helps Brexit supporters,
and works against those who want to change Brexit, because the latter
are concentrated in cities. The second is that Brexit regret does not
necessarily imply a wish to change Johnson’s Brexit deal, because
the prospect of reopening the Brexit question reminds many of the
three years after the referendum.
It may seem hard to
imagine passing this tipping point, because before we get to that
tipping point the opposite is true, hence Labour’s current
commitment to making Brexit work. But the big Brexit divide cuts both
ways, so that when enough voters see Brexit as a mistake that needs a
fundamental correction it will become politically advantageous to
argue for that correction, and a political liability to oppose this. Then the costs for the Conservatives of
championing a policy mainly favoured by the elderly that incurs
serious economic costs will finally come home to roost.
Labour are unlikely
to reverse their current Brexit policy immediately on taking office.
They should instead immediately start discussions with the EU about “making
Brexit work”. It is important that such discussions start quietly
and without diverting attention from more immediate concerns. It
would for the same reason be an obvious political mistake to raise
expectations about what such negotiations can achieve, because any
economic gains that follow will be small. But the existence of a
tipping point for public opinion on Brexit means not only that Labour
need to be prepared to switch from making Brexit work to changing
Brexit during their time in government, but also that it is in
Labour’s interests to do anything it can to hasten the arrival of
that tipping point.
The Conservatives
ability to tag Miliband’s Labour party with economic incompetence
was crucial in winning the 2015 election, and Johnson’s Brexit deal
can do the same for a Labour party once it holds power. It is
therefore in a Labour government’s interests, assuming we have one
in 2025, to be as open and honest as possible about both the economic
advantages of the alternative ways of softening a hard Brexit, and
also about what is and is not possible in terms of any deals with the
EU.
That raises the
question as to when Labour should begin to suggest the possibility of
going further, which in practice will almost certainly mean rejoining
the EU’s customs union and/or single market? Should this happen
during Labour’s first term, or its intended second term?
The point at which Labour changes from doing what it can with a hard Brexit to changing Brexit will depend on many things beside the tipping point in public opinion, such as the pressure from domestic events and willingness on the EU
side of the table. Two other factors are worth noting here. The first is the likely lag between any agreement
(which itself will take time) and the economic benefits that would
follow. If one of key reasons for a Labour government to change
Brexit would be to benefit the economy and therefore household
incomes and public services, the sooner this happens the better.
This suggests moving as quickly as is feasible to join the EU’s customs union. While the economic benefits
are probably less than being part of the single market, it would
remove some of the paperwork that small businesses in particular find
is a barrier to exporting to the EU. It would also provide a good
excuse to revise or end the few genuinely new trade deals obtained by
Liz Truss, which seem
lopsided to the detriment of some UK sectors. There
may also be some gains to be made through alignment of standards,
cooperation in research etc.
In assessing the
likelihood of the UK joining the single market we need to consider a
second factor influencing timing, and that is the behaviour of the
Conservative party in opposition. If the Labour government benefits
from the economic recovery that follows from the current recession,
then – as with the last Labour government – the Conservatives will
have to focus on social rather than economic issues. Just as with the
last Labour government, immigration is likely to be their most
effective issue.
Attitudes to
immigration have been changing over the last decade, but it remains
the most potent social issue among potential Conservative voters.
Furthermore, how important those voters feel that immigration is
compared to other issues depends a great deal on how much coverage it
is given in the Conservative press. That in turn will depend on
whether the Conservatives are in power or not. [1] Newspaper stories
about immigration or asylum seekers are therefore likely to begin to
increase under a Labour government, and public concern will rise
along with it.
It is therefore
immigration, rather than attitudes to Brexit per se, that is likely
to be the main barrier to Labour trying to negotiate membership of
the Single Market, because of course that membership requires free
movement of labour within that market. The best way Labour have to
moderate this Conservative weapon is to move to some form of PR for
general elections, for reasons noted above. There may be other means
of making free movement politically acceptable, as Peter Kellner
suggests
here.
None of these
arguments guarantees that Labour will or will not try to negotiate to
join the Single Market, but they do suggest that negotiations to join
the Single Market will be one of the last stages in the long road
back from Brexit.
However this should
not stop Labour being honest about the economic costs of leaving the
EU’s single market, and therefore the benefits of joining. Chris
Grey suggests
an interesting additional idea, which is that Labour suggests that it
would be unwise (or the EU might be unwilling) to consider the UK
being part of the Single Market until there is cross-party agreement
that this should happen. This would have the political benefit, for
Labour, of emphasising the economic costs of the Conservative’s
continuing attachment to a hard Brexit.
To summarise, the
existence and proximity of a tipping point on public opinion about
Brexit means it is important for any Labour government from 2025 to
be open and honest about Brexit’s economic costs and what can and
cannot be achieved in negotiations. It also means that it is quite
possible to see Labour moving beyond ‘making Brexit work’ once
they have power, although rejoining the Single Market remains the
most difficult move in political terms.
Once the tipping
point is passed, believing in a hard Brexit will become a serious
political disadvantage for the Conservatives. Yet what is difficult
to imagine is how they will escape from what will become a serious
drag on their popularity. Johnson and others have made Brexit part of
the Conservative party’s DNA. The conditions that make me
confident in saying that no Conservative government
will change a hard Brexit for at least a decade also make it
difficult imagining them renouncing the policy in opposition. Perhaps
it will take many years in opposition for MPs, but also crucially party members and the owners of the right wing press, to realise what
a liability Brexit support is for them.
[1] The exception
was during the Cameron government, but I don’t think it would be
overly cynical to say that this had a great deal to do with Brexit,
which the press favoured and Cameron did not. The data supports this,
with the number of articles mentioning immigration rising
rapidly from 2013 until 2016.