It is now well established that Rishi
Sunak as Chancellor played a significant role in increasing the death
toll from the pandemic on at least two occasions. The first was to
introduce ‘Eat Out to Help Out’ in the summer of 2020, and the
second was to
advise Prime Minister Johnson to ignore the medical
advice from SAGE to impose a lockdown in the early Autumn and subsequently.
In both cases he will argue that, as
Chancellor, his role was to protect the economy. Yet he did no such
thing. As Chancellor, he failed to understand that to protect the
economy you had to control the virus, which means keeping the number
of people infected low. I and other economists argued
this at the time, but in this post I want to set out
the logic in a new way to show why there never was a health/economy
trade-off.
A decade before the pandemic a
group of us published an article on the economic
effects of a pandemic. One of the main findings of the paper was that
a severe pandemic can involve serious economic costs because
consumers will avoid what we called ‘social consumption’. Social
consumption involves anything that brings consumers into contact with
others, so includes eating out, going to pubs or the cinema, using
public transport etc. Social consumption involves a third of total
consumption, so if people significantly reduce their participation in
these activities the impact on the economy will be large [1].
We could call this effect an
‘unofficial lockdown’. Individuals stay at home rather than eat
out or go to the cinema because they want to avoid catching the
virus, not because they have been told to by the government. The key
point is that if the government does nothing, individual actions
attempting to avoid catching a potentially deadly virus will lead to
a substantial economic slowdown. Swedish GDP fell by 7.6% in 2020Q2,
even though no official lockdown was imposed.
This is why reducing the number of
people infected also helps the economy recover. There is no
health/economy trade-off in this kind of pandemic. If economic policy
encourages people to put themselves at greater risk of getting
infected, as Eat Out to Help Out (EOTHO) did, then any boost to the
economy would have been limited to when the scheme operated, and
thereafter there would only be economic damage as infections
increased. The only situation where this might not happen is if R
(the average number of people infected by one person) was
sufficiently less than one and it remained below one despite EOTHO,
but we know this wasn’t the case and Sunak made a point of not
asking SAGE about it.
While EOTHO played some part in the
second wave that grew during the Autumn of 2020, just as serious a
failure was Sunak arguing against the SAGE proposal for a second
lockdown in September. It is the case that an official lockdown has a
bigger immediate negative impact on the economy than an unofficial
lockdown. This is because, for example, in an unofficial lockdown
-
Many people will not be well
informed, and will not reduce their social consumption much if at
all -
Some people will be well
informed, but decide the risk to themselves is small so they will
not reduce their social consumption, and discount the risk of them
infecting the more vulnerable. -
Employers may force workers to
continue to travel work, even though both the work environment and
travelling to it may risk infection.
Yet for the same reasons, an unofficial
lockdown has less of an effect in reducing R than an official one.
[2] This is what the UK experienced in the Autumn of 2020, even with
the addition of some regionally based restrictions imposed by the
government. With R>1, not only are more people being infected,
with some dying or getting Long Covid, but the economic damage
persists as individuals try to protect themselves by withdrawing
from social consumption.
The UK and other countries experience
of full official lockdowns is that they reduce R to less than one, so
with a short lag infections start falling. This was the case for the
lockdown at the end of March, the one month lockdown in November and
the lockdown in January 2021. Because R<1, the number of
infections fall and then the economic damage caused by individuals
avoiding social consumption dissipates.
My focus on what happens to R is
crucial, because there is a world of difference between R<1 and
R>1. In the former the pandemic is being controlled, so that when
lockdown ends the situation is manageable, and the hit to the economy
from reduced social consumption will be relatively small. If R>1
the damage to the economy just keeps getting larger.
So while an official lockdown might do
more damage to the economy than an unofficial one while it lasts, the
official one deals with the problem, so reduces the time that Covid
damages the economy. In contrast doing nothing, or taking measures
that fall short of a full lockdown, allows infection numbers to
increase and so allows damage to the economy to persist.
This is exactly what we saw in the
Autumn of 2020. Thanks in part to pressure from Sunak, the government
rejected advice from the experts to impose a full lockdown, and so
infection numbers grew and consumption remained over 10% below its
end-2019 level. When a sustained lockdown came in 2021Q1 consumption
was only a few percentage points lower than 2020Q3 (GDP was actually
higher), but that lockdown brought cases right down, and vaccines
then removed the need for further lockdowns.
It is really difficult to rationalise
what Sunak did during the summer and autumn of 2020. By deliberately
not asking SAGE about the impact of EOTHO, he must have known this
would increase infection rates. Did he really think the economy would
be largely unaffected by a second wave? Unlikely, as in enacting
EOTHO he was aware of people reducing social consumption because of
the pandemic! Perhaps his actions were guided by perceived political
advantage rather than economic or health impacts.
Gross incompetence is a strong term,
but I fear it clearly applies to Sunak in these two cases. His
thinking appears not to have got beyond the level of a right wing
newspaper column, despite having the resources of the Treasury at his
disposal. [3] His actions not only led to many people dying, but his
actions also damaged the economy when he was the minister in charge
of protecting it.
[1] This response modelled in our paper
involves individuals trying to avoid catching the virus. It was not
coordinated by governments in any way. In the paper we didn’t look
at government imposed lockdowns beyond school closures.
[2] Obviously this judgement is country
dependent. In countries where people and employers are better
informed and more socially minded, unofficial lockdowns may come
closer to replicating official lockdowns. This is why comparisons
between countries that did lockdown and Sweden are potentially
misleading, and why comparisons between Sweden and other Scandinavian
countries are much more informative.
[3] Reporting on the Covid inquiry has
naturally focused on political culpability rather than the advice
politicians were being given. In this particular case it is
inconceivable that the Treasury was unaware of the analysis I outline
here. What happened to that analysis, and how far up the civil
service hierarchy it got, are interesting questions we do not know
the answer to. Until we know, we can only wonder whether senior Treasury officials’ concern about higher government borrowing in lockdowns mattered more than the health of the economy.
