The economics and politics of the three sides of aggregate fiscal policy


This post is
prompted by the discussion around the Autumn Statement, but also in
particular a post
by Ian Mulheirn. Here is the key diagram from his post.

I think the diagram
is useful because it distinguishes three aspects of aggregate fiscal
policy that can easily become confused. However it is worth talking a
bit about its limitations lest the diagram itself adds to confusion.

Each of the three
circles above represents a non-central point on a spectrum. Take the
‘support short term growth’ for example. At one extreme (near to
my own position [1]) you can believe that in a recession, or before
an expected recession, or during the recovery from a recession, the
need to stimulate the economy should take absolute priority over all
other objectives. The other extreme would be to ignore the
macroeconomic situation entirely, which Osborne/Cameron in 2010.
There are plenty of intermediate positions between those two
extremes, like Balls/Miliband over the same period.

The same is true for
public services. I would go
further and make
a conceptual distinction between what
you might call the ‘scope of the state’ and ‘how well the
existing state is funded’. These both have to be continuums. My own view
on the latter is that we are currently not even at an extreme point,
but one which is not tenable.

What does fiscally
conservative mean? I think the label here is misleading, because
fiscal conservatism could equally be the label for the other circles.
In my view the issue here is about the expected path of public debt, or
public sector net wealth, over the long run. One extreme on this
would be that public debt doesn’t matter at all. The other might be
that debt should be reduced rapidly, and there are plenty of
intermediate positions between these two extremes. Any good medium
term current deficit fiscal rule will have to make a choice about
where on this continuum it wants to go.

I would not include
under this heading debates about the form of fiscal rules.
This debate is more technical in nature, and involves whether it is
better to target the current deficit or the total deficit, or whether
some particular change in the debt/GDP ratio is a good target, and so
on. Discussion often confuses the form of a target with the value it
is set at. For example, a target for the medium term current deficit
is a good rule in my view, but there is still a debate about what
level the target should be at, and that will depend on how quickly
public debt should fall or rise in the long run.

Making these distinctions help us see historical events in a clearer light. 2010 austerity, for example, combined two of these circles: a large contraction of public services, and ignoring the need for fiscal stimulus in a recession. It is also interesting to note that the level of taxes is influenced by at least two of these aspects of fiscal policy: higher public spending implies higher taxes, as does (for given public spending) a more conservative debt policy.  

The correct way to
locate past Chancellors in all three dimensions of fiscal policy would be a three dimensional chart, with the three
axes being ‘how high is public spending’, ‘the priority given
to the macroeconomy during economic downturns’, and ‘how quickly
should debt rise or fall over the long run’. But three dimensional
diagrams are not easy to look at, which is why I suspect Ian uses the
simpler formulation above. But that also creates problems.

While I have no
problem with the classification of Osborne/Cameron, you could
question whether Balls/Miliband were suggesting higher public
spending or just less cuts than Osborne. Equally by putting
Johnson/Sunak in the higher public spending circle, it’s clear that
Ian is taking a conservative definition of ‘higher’. With
Hunt/Sunak a budget that back loaded spending cuts and immediate
energy subsidies could be spun as worrying about the economy in the
short run, but they are still putting up taxes in a recession. I
suspect the real reason that spending cuts have been delayed is not
the coming recession but just that spending cannot be cut further,
and the real reason for energy subsidies is political.

I think it’s
clearer if we try and use one straight line and a 2D chart

The straight line
just looks at the amount of government spending, and I’ve only
included actual Chancellors, not PMs or shadows, and excluded Hammond
and Javid for space. Brown is out in front for creating an NHS that worked,
Sunak (under Johnson) increased spending as a share of GDP in a few
public services but not others. Hunt gets to be to the right of
Osborne mainly because his spending cuts have not happened yet. (I
have put Darling in brackets because his period was dominated by the

The second diagram
plots the remaining two aspects of fiscal policy. Here I have only
included four Chancellors, because Brown didn’t have a recession to
deal with as Chancellor, and the GFC blew up his fiscal rules.
Darling wins on stimulus in a recession because he did exactly that,
while Sunak is next because of furlough. On debt reduction speed
Osborne is clearly on top because of what he did once the deficit was
brought down, and Darling is where he is because of plans he never
got a chance to implement.

I think this is a
more informative way to compare what Chancellors did, but it is a lot
less pretty than Ian’s original.

I want to end by
disagreeing with the main policy point that Ian makes, which is that
Labour should aim for the intersection of his three circles. I have
no problem with ‘higher government spending’ and ‘stimulus in
recessions’. Indeed I think all good governments, from the
starting point we are currently in, should increase public spending,
and every government should follow basic macroeconomics. Where I
differ is on a conservative policy towards public debt.

I understand that in
opposition it may make political sense to match the government’s
rolling target for falling debt to GDP. But I would hope, once in
government, Labour would commission HMT (or the OBR) to do the kind
of in depth analysis they did in 1998. This should show the drawbacks
of that particular target that I outline
. The most pertinent from Labour’s point of view
is that by focusing on debt rather than assets this target
discourages the kind of Green investment that is a centrepiece of
Labour’s economic policy offering. It is no good hoping these two
things don’t conflict based on current forecasts, because at some
point they almost certainly will.

Much more difficult
is the judgement about the desirable path for net public sector
wealth over the long run. Again a good analysis has to look at the
costs of getting the judgement wrong. The costs of underfunding
public services are today painfully clear. The cost of not reducing
the current negative level of net public sector wealth is less clear,
once you move beyond nonsense involving market pressure.

[1] I would not
argue for immediate fiscal stimulus right now because it could be
immediately offset by higher interest rates, but nor would I agree
with tax increases. But I am probably at the extreme in believing
fiscal stimulus should apply until the recovery from a recession is
complete, which is why I
the Biden stimulus was a good idea.