Economy

How 14 years have shown the impossibility of shrinking the UK state

[ad_1]

 

The Budget was
predictable, and predictably boring. Hunt cut taxes, but the tax
burden is still rising because of the tax increases already
programmed in. Furthermore, he was only able to make the tax cuts he
did (i.e. reduce the extent of tax increases) because he had
previously pencilled in assumptions about public spending that were
fantastically low. You can either portray those assumptions as
Austerity 2.0 or just silly – I
did the latter here
.

However, with (I
hope) the not silly assumption that this will be the last
Conservative budget [1] for a while, I thought it might be useful to look
back on the previous 14+ such events since 2010 to see if there are
any general lessons we can draw from them all. One in particular runs
through most of them and really sticks out. From 2010 onwards
Conservative Chancellors have attempted to 
cut what they like to call the ‘tax burden’ by reducing the size of the
state without any major changes in what the state is meant to do, and as the chart below shows (which 
includes the impact of yesterday’s Budget) they have
completely failed to achieve this objective. 

The professed aim of
Austerity 1.0 from 2010 onwards was to reduce the budget deficit, but
it quickly became clear that was not the only aim, because Osborne
started cutting taxes in his budgets as well as reducing spending.
(The initial VAT increase was deliberately designed to give the
impression it was all about the deficit.) Yet despite cuts to
corporation tax and personal tax thresholds, all Osborne could do
was to keep the tax share stable at around 33% of GDP.

Then came Brexit and
Boris Johnson. Johnson understood that trying to make Brexit work
while continuing to shrink the state was politically impossible, so
he undertook a partial and limited (in scope) reversal of Austerity
1.0 by raising spending on the NHS, schools and the police. This
would inevitably mean a large increase in taxes, undertaken by then
Chancellor Sunak for reasons he
clearly set out here
. Even without the intervention of
Covid it is unlikely the additional spending would have been enough
to start bringing NHS waiting lists down, so the government got the
worst of all worlds in political terms: public services were
inadequately funded yet the tax share was going up significantly.

When Johnson was
thrown out of office, what little political sense he had brought on
the size of the state left too. It was replaced by fantasy and
deception, in that order. The fantasy was of course Truss, who had
bought the Laffer curve idea that all you needed to do to get more
revenue was to cut taxes because strong economic growth would surely
follow. Very few people believe this, in large part because it’s
not true. The deception is Jeremy Hunt, who is pretending he can cut
taxes by using make-believe numbers for future public spending
(Austerity 2.0).

Almost 15 years of
trying to reduce taxes, and complete failure. There are many reasons
why, but one for me stands out because it doomed the project to
shrink the state from the start. The chart below shows health
spending as a share of GDP in the UK, France, Germany and Italy since
1980.


Don’t worry about
the details, just note that all four series are trending upwards by
substantial amounts. There are many reasons for this trend, like
people living longer or discovering new ways to help them live
longer, but as yet we have not found anything to counteract health
absorbing a steadily increasing share of national income.

If governments try
to keep the health share constant (aka ‘protecting it’), as the chart clearly shows the UK
government did from 2010 until just before the pandemic, then the
quality of healthcare provided for most of the population will
steadily deteriorate. To avoid that deterioration, which is not
sustainable politically, you have to pay more of national income into
healthcare. If you have the NHS, that means a rising share of taxes
in GDP.

Decades ago this
trend rise in health spending as a share of GDP was offset by the
‘peace dividend’, with defence spending falling because of the
end of the cold war. Those days have long gone, with no obvious
replacement in terms of a major area of public spending where less and
less money is needed.

None of this was
unknown in 2010. The shrinking the state project was doomed from the
start, and anyone familiar with these numbers knew it was doomed from
the start. So why didn’t Conservative politicians realise this, and
why are they still in denial about it? I think in 2010 at least there
was a view among Conservatives that everything in the public sector
was inefficient, and the way to improve efficiency was to squeeze
resources or introduce market mechanisms. [2] Again there were
international comparisons that suggested this wasn’t true, for the
NHS at least, but the story fitted too easily with a neoliberal
viewpoint.

However you have to
ask if any Conservative who had realised the futility of trying to
shrink the state would have been successful as politicians? It was
and continues to be a message that Conservative members, press
barons or donors don’t want to hear. Look at how Sunak’s position has
changed from one recognising realities as Chancellor to a Prime
Minister who has to pretend he can get something for nothing. The way
politics is done in the media doesn’t help either, where basic
numerical facts like an international trend rise in the share of
health spending in GDP seems too much for many political journalists
to remember.

So the chances of
the Conservatives giving up their obsession with tax cuts is close to
zero. In addition the media will remain constantly surprised that UK
tax shares are steadily rising. This is unfortunate, because in
trying to do the impossible (reduce the tax share) the Conservative
party has done a great deal of harm. Obvious harm to the public
services, but also to the economy. 

Austerity 1.0 is a key reason why
the UK’s recovery from the Global Financial Crisis recession was so
weak, and austerity also
played an important part
in influencing the Brexit referendum result. The
damage caused by Truss we all know, while the game played by
Hunt/Sunak is in danger of preventing Labour doing enough when they
gain power. The dire state of the NHS is also directly influencing the economy. As the OBR notes, the number of inactive working age adults has increased substantially since the pandemic, with many citing long-term illness. The OBR now expects no recovery in labour force participation over the next five years, making the UK quite different from other countries where post-pandemic participation rates have recovered. This seems quite consistent with the continuing squeeze on public sector spending. For more details on how poor health has a negative influence on the economy as well as wellbeing, see the reports from the IPPR’s Commission on Health and Prosperity, and Bob Hawkings here.

While there will always be a debate about whether high or
low tax countries grow faster, the UK’s experience over the last 14 years show that trying to cut taxes by shrinking the state when it
is impossible to do so is very damaging indeed. Unfortunately neither the Conservative party nor many political commentators in the media appear willing to recognise the damage these attempts have done to both social wellbeing and the UK economy. 

[1] I fear there will be one more Autumn Statement before the election, and because that will involve another year of nonsense public spending assumptions, it will give the government room within its fiscal rules for further tax cuts.

[2] What they also did was starve the NHS of investment, which was bound to decrease efficiency, and privatize increasing amounts of its provision, which reduced the quality of provision.   

[ad_2]
Source link

Related Articles

Back to top button