The macroeconomic record of the coalition government.
Wren-Lewis, Simon
This paper examines the outcomes for changes introduced by the UK
Coalition government in 2010. The Office for Budget Responsibility (OBR)
is generally regarded as a success, and should become a permanent part
of fiscal policymaking. The form of the primary fiscal mandate,
involving a five-year rolling target, appears to be a sensible way to
shape fiscal decisions when monetary policy is able to stabilise the
economy. Unfortunately it was introduced, along with a five-year
programme of severe fiscal consolidation (austerity), while the economy
was in a liquidity trap. The OBR estimates austerity reduced GDP growth
by I per cent in both 2010-11 and 201 1-12, and monetary policy was
unable to offset this. For the Liberal Democrats a misreading of the
Eurozone crisis may have been responsible for this mistake, but for the
Conservatives this mistake appears to derive from an unconventional view
that the liquidity trap is unimportant.
Keywords: UK Coalition government; fiscal; monetary; liquidity
trap; zero lower bound; fiscal council; fiscal rules JEL
Classifications: E02; E62; E65
1. Introduction
The simple, but potentially misleading, way to evaluate the
macroeconomic record of any government is to look at outcomes. On this
basis, as I discuss in section 3.3, the record of this government looks
pretty bad. It gained power at the turning point of a deep depression.
The experience of the previous few decades would suggest that this
turning point would be followed by a robust recovery, with above trend
GDP growth. The actual outcome has been much more disappointing. GDP per
head appears to have moved further away from the pre-crisis trend,
rather than moving back towards it.
The problem with that evaluation approach is that it overstates the
role of the government in influencing macroeconomic outcomes. The poor
performance of the UK economy since 2010 could reflect the influence of
the global economy, the lingering effects of a financial crisis, or
other factors beyond the government's control.
This paper takes a different approach. We begin, in section 2, by
outlining the key innovations and changes that the Coalition brought to
the conduct of macroeconomic policy. Then, in section 3, we discuss the
extent to which those innovations evolved, and only then do we discuss
how these interventions influenced macroeconomic outcomes. In other
words, we try and focus on the impact of changes in government policy on
macroeconomic outcomes.
The Coalition brought in three main changes to the conduct of
macroeconomic policy in 2010, all of which involved fiscal policy. The
first, outlined in section 2.1, established an independent body to
undertake fiscal forecasts (the Office for Budget Responsibility, OBR).
This can be judged a success for a variety of reasons (section 3.1)
including the fact that discussion now involves extending its remit. The
second (section 2.2) involved the form of the primary fiscal mandate
introduced to replace the previous government's fiscal rules. I
argue that the structure of this rule represented an improvement over
the previous administration's rules, as long as monetary policy was
capable of successfully controlling demand and inflation. Unfortunately
it was introduced while monetary policy was constrained because nominal
interest rates were at their lower bound.
The third major change introduced by the Coalition government, and
by far the most controversial, was greater fiscal austerity (section
2.3). Fiscal contraction is estimated by the OBR to have reduced GDP
growth by 1 per cent in both 2010-11 and 2011-12, and section 3.3
suggests that there are good reasons for thinking this is a conservative
estimate. The main issue is whether the impact of this tighter fiscal
policy might have been offset by a looser monetary policy, which is
discussed in section 3.4. Section 3.5 briefly considers two innovations
brought in midway through the Coalition's period of office, which
involved encouraging lending and subsidising borrowing. Section 4 draws
some overall conclusions about the impact of the Coalition's
macroeconomic policy, and the rationale behind it.
2. The macroeconomic policy framework
When the 2010 Coalition government took power it introduced
macroeconomic changes mainly in the area of fiscal policy. They can be
split into three: the establishment of the OBR, the form of the fiscal
rules which the OBR was tasked with monitoring, and the particular
targets that were fed into those fiscal rules.
2.1 The Office for Budget Responsibility
George Osborne first proposed establishing a 'fiscal
council' or 'independent fiscal institution' for the UK
in 2008. At the time a few countries had similar bodies, but both the
IMF and OECD were strongly encouraging their formation elsewhere. What
these fiscal councils did varied significantly (Calmfors and Wren-Lewis,
2011): for example in Sweden it had a wide ranging role to act as a kind
of macroeconomic watchdog, with no forecasting responsibility. In
contrast the OBR had a very specific remit, to provide the fiscal
forecasts on which budget decisions would be based. It was not allowed
to examine the implications of alternative fiscal policies.
It is tempting to draw parallels between the establishment of the
OBR and granting independence to the Bank of England that took place at
the start of Labour's period of office. (The limits to such
comparisons are discussed in Wren-Lewis, 2013a.) In both cases
responsibility for a key aspect of macroeconomic analysis was devolved
away from government. Academic economists saw independent central banks
as a way of reducing what they call 'inflation bias'. While
there was far less academic analysis of fiscal councils (for an early
proposal for the UK, see Wren-Lewis, 1996), what there was saw these
councils as a means of countering 'deficit bias': the tendency
for government debt to rise over time (Calmfors and Wren-Lewis, 2011).
However, while deficit bias had been clear among OECD countries as a
whole in the 30 years before the 2008 recession (where debt to GDP
almost doubled), this problem appears less acute for the UK. By the late
1970s UK government debt to GDP had fallen from very high levels
following WWII to less than 50 per cent, and since then the pattern has
been largely cyclical, with if anything a downward trend until the
financial crisis. A counter argument might be that this reduction was
partly achieved through relatively high inflation, which might have been
lower in the absence of deficit bias.
One reason why the UK might have been less subject to deficit bias
than other countries is the strong role of the UK Treasury within
government. Nevertheless, the Conservative opposition argued, with some
justification (Wren-Lewis, 2013b), that the fiscal forecasts of the then
Labour government had been overoptimistic since the early 2000s, and
that an independent institution would be less prone to fiscal
over-optimism. (Frankel and Schreger, 2013, have recently discussed the
problem of overoptimistic fiscal forecasts by governments.) This helps
explain why the OBR's task was quite specific and limited to the
production of fiscal forecasts. In effect, what the new Coalition
government did was contract out the macroeconomic forecast on which
budget decisions would be based. This raised two interesting questions.
First, was it possible to contract out this task while much of the
specific forecasting expertise remained within government, and yet for
the OBR to remain independent. Second, was it right to prevent the OBR
examining the impact of alternative fiscal policies to those of the
current government? Both questions are discussed in section 3.1.
2.2 The form of the fiscal rules
The primary fiscal mandate adopted by the Coalition government was
to achieve a cyclically adjusted current deficit (the overall deficit
less government investment) of zero within the next five years. We
discuss the zero number below, focusing here on the form of this fiscal
rule. It replaced the formulation adopted by the previous government,
where one of the rules (before the recession) was to achieve current
balance over the course of an economic cycle.
Both the new and old rules attempted to correct for the influence
of the economic cycle on the deficit. One of the drawbacks of the Labour
government's method of correcting for the cycle is that it made the
rule backward looking, so that favourable shocks during the early part
of the cycle could be used to justify a lax policy later on in the
cycle. As George Osborne put it in his June 2010 Budget speech:
"Past prudence was an excuse for future irresponsibility."
(The extent of this problem in practice, and the danger that the dating
of the cycle might become political, is discussed in Wren-Lewis, 2013b.)
Cyclical correction avoids this, but is arguably less transparent.
Ironically, in the case of the Coalition's main mandate it is
not obvious why any cyclical correction is required. This is because a
forecast normally assumes that in five years time the monetary
authorities will have closed any existing output gap, in which case the
expected cyclically adjusted and actual deficits will be the same. The
only exception might be if monetary policy is unable to close the output
gap because interest rates are at their lower bound but, for reasons
that will become clear, it seems unlikely that the Coalition was trying
to anticipate this problem.
A feature shared with one of Labour's rules was to focus on
the current balance, rather than the overall deficit, the difference
being public investment. This leaves public investment potentially
uncontrolled. In the case of Labour's policy this meant adding a
second rule, which in their case placed a limit on the overall debt to
GDP ratio. For perhaps the same reasons, the Coalition supplemented
their main mandate with a second rule, which was that debt should be
falling as a share of national income in 2015-16. Having a target that
debt to GDP should be falling in just one particular year seems odd
because it is open to manipulation through the timing of discretionary
expenditure.
An important contrast between the primary and supplementary targets
related to the time period over which the target had to be achieved. The
primary mandate involved a 'rolling' target: in 2010 the
target was to achieve a particular balance by 2015, but in 2012 the
target would be shifted to 2017. In contrast, the supplementary target
was for a specific date. This distinction proved important in practice.
As the rationale for a rolling target seems counterintuitive, it is
worth elaborating on it here.
Having a deficit target to be achieved within the next five years,
where that five-year period remains as time moves on (a rolling target)
might seem too easy. There is never a date by which we can unambiguously
say that the target has been achieved or not. It would seem much better
to have a target for a fixed date, as with the Coalition's
supplementary target. The problem with this logic comes when we approach
the target date, and some unexpected shock occurs. Rather than adjusting
to that shock gradually over the next five years, as a rolling target
will allow, a fixed date target can either require adjustment to be very
rapid, or for the rule to be ignored. The former breaks an important
principle of good fiscal policy, which is that the deficit should be a
shock absorber, not a rigid target. (This principle is explained in
Portes and Wren-Lewis, 2014.)
Rolling targets are familiar from monetary policy. This attempts to
achieve the inflation target within the next two years or so. The reason
often given for this is that it takes some time for changes in interest
rates to have their full influence on prices, but this is only part of
the story. Interest rates have some fairly immediate impact on prices,
so it would in principle be possible to try and meet an inflation target
within a shorter time horizon. This is not attempted because it would
lead to damaging variability in interest rates and output. The rolling
target for inflation adopted by many central banks does not stop central
banks being accountable for their actions, but it does prevent damaging
volatility.
The same logic applies to fiscal policy. It is true that rolling
targets do give the fiscal authority the possibility to cheat. If the
government has in the past always cheated and there is no institutional
arrangement to stop this happening, then fixed date targets may be an
unfortunate necessity. However, in the past, UK governments have proved
to be quite capable of taking the actions required to meet fiscal rules:
what has typically derailed them has been unexpected shocks like
recessions. In addition, with the OBR we have an effective fiscal
council which in this respect acts as a watchdog. As a result, Portes
and Wren-Lewis (2014) suggest that the Coalition's main mandate
involving a 5-year rolling target is a good way of managing fiscal
policy in normal times. (1) Unfortunately 2010 was not a normal time in
one key respect.
2.3 Austerity
The Conservative party had criticised the Labour government for
using fiscal policy to support the economy in 2008 and 2009. At the time
their argument appeared to be running against the international
consensus, which was that fiscal stimulus was required to assist
monetary policy in mitigating the impact of the financial crisis and
recession. As George Osborne said in a speech given at the RSA in April
2009:
"The crisis has also exposed two fundamental arguments. The
first is whether, when you are already borrowing too much, you should
deliberately try and borrow your way out of debt. David Cameron and I
have consistently argued against this irresponsible course of action. To
begin with we were almost alone in making this argument, but we held our
nerve and stuck to our principles. And now informed opinion has turned
in our direction...."
Consistent with this view, they fought the election arguing that
Labour's plans for bringing the deficit down were too slow.
Although the Liberal Democrats had in the 2010 election presented fiscal
plans which appeared closer to Labour than the Conservatives, as part of
the Coalition agreement they endorsed George Osborne's policy. In
his budget of June 2010, George Osborne announced a number of measures
that would help achieve their target of eliminating the current deficit
by 2015. In particular spending was to be cut by an additional 32
billion [pounds sterling] by 2015, and the VAT rate was to be increased
from 17.5 per cent to 20 per cent at the beginning of 2011. The burden
of reducing the deficit was planned to fall mainly (80 per cent) on cuts
to spending, with the remainder (20 per cent) accounted for by higher
taxes. (In the event a larger percentage of the actual reduction in the
deficit has been achieved through spending cuts.)
What lay behind the view that additional austerity was required in
2010? If the government's budget deficit is viewed as analogous to
that of an individual, the answer is obvious. The deficit was very
large, so the ratio of debt to GDP was rising rapidly, which suggested
that the sooner this was brought under control the better. However the
government is not like an individual, for reasons that have been
understood by macroeconomists since Keynes. This was why governments,
supported by the IMF, had taken measures that increased their deficits
in 2009 in an effort to tackle the recession that followed the financial
crisis. From a Keynesian perspective, large government deficits were
simply the counterpart to a large increase in private sector saving, and
the increased government debt was providing the financial assets that
the additional private sector saving required.
In 2010 fiscal policy switched from supporting demand in a
Keynesian fashion to focus on deficit reduction. This major change in
focus was not limited to the UK, and at the global level was motivated
by a concern that the emerging Eurozone crisis could spread to other
countries. In a recent evaluation (IMF, 2014), the IMF has concluded
that this strongly influenced their own switch from advocating fiscal
stimulus in 2009 to advocating austerity in 2010. This may also have led
the Liberal Democrats to change their position as they became part of
the government. However that IMF evaluation also concludes that this
switch was a mistake, because the Eurozone debt crisis occurred because
individual Eurozone countries could not print their own currency, and
the ECB was not prepared (until the OMT programme of September 2012) to
act as a sovereign lender of last resort. The key reasons why a debt
funding crisis is highly unlikely to occur if the government has
borrowed in its own currency issued by its own central bank under a
floating exchange regime are described in Krugman (2014).
To be fair, as 2010 evolved that point was not clearly understood
(until De Grauwe, 2011, in particular), and the Coalition government
certainly used this fear of a Greek-style funding crisis as part of
their justification for additional austerity. For example, in his June
2010 Budget speech, Osborne said: "Questions that were asked about
the liquidity and solvency of banking systems are now being asked of the
liquidity and solvency of some of the governments that stand behind
those banks. I do not want those questions ever to be asked of this
country."
Another possible reason for advocating additional fiscal austerity
in a recession might be a belief that this would in itself encourage
sufficient private sector demand that it would become expansionary. Such
a view goes against basic macroeconomic models and teaching, but it did
briefly become popular following empirical work by Alesina and Ardagna
(2009), before other studies emerged that were more consistent with
standard theory. How important a belief in 'expansionary
austerity' was in motivating Coalition policy is unclear. In a
Financial Times article (2) with Jeffrey Sachs, George Osborne did
write: "There are many well-studied examples of 'negative
fiscal multipliers', in which credible fiscal retrenchments in fact
stimulated the economy, via greater consumer and investor outlays, by
reducing borrowing costs and spurring confidence." However, a more
typical line taken by the Coalition was to sidestep the question of what
impact austerity might have on output, rather than trying to argue that
it would directly assist recovery by expanding demand.
George Osborne and David Cameron did argue against Labour's
fiscal expansion in 2008/9, which was well before the Eurozone debt
crisis. In a speech delivered to the RSA in April 2009, George Osborne
gave a short account of some history of macroeconomic thought, which
included the following section:
"[New Keynesian] Models of this kind underpin our whole
macroeconomic policy framework--in particular the idea that by using
monetary policy to manage demand and control inflation you can keep
unemployment low and stable. And they underpinned the argument David
Cameron and I advanced last autumn--that monetary policy should bear the
strain of stimulating demand--an argument echoed by the Governor of the
Bank of England last month when he said that "monetary policy
should bear the brunt of dealing with the ups and downs of the
economy". We now appear to be winning that argument hands
down."
The idea here is that although austerity in itself would reduce
demand (as is typically the case for cuts in government consumption or
investment in New Keynesian models), active monetary policy in the
context of an inflation target would move to fully offset that impact.
As George Osborne put it in a speech in September 2009:
"Monetary activism to keep interest rates low and stimulate
the economy. Fiscal responsibility to restore confidence and rebuild our
battered public finances."
I have argued elsewhere (Kirsanova et al., 2009) that this view did
indeed reflect the academic consensus that emerged during the period
before the financial crisis. Monetary policy should be assigned the task
of stabilising demand and inflation, and fiscal policy should focus on
achieving targets for government deficits or debts. However, that view
was predicated on analysis that assumed that monetary policy was
unconstrained, and in particular that nominal interest rates were free
to move in either direction. At about the time that George Osborne gave
the speeches quoted above, UK interest rates were reduced to what has
become their lower bound of 0.5 per cent.
When interest rates hit their Zero Lower Bound (ZLB), many had
argued (following Keynes) that monetary policy needed to be supported by
an expansionary fiscal policy. This was the advice that the IMF had
given during the worst of the recession. An alternative that a number of
macroeconomists had also explored as a result of Japan hitting the ZLB
in the 1990s was to move monetary policy away from unchanging inflation
targets (Krugman, 1998; Eggertsson and Woodford, 2003). A third
possibility was that the Quantitative Easing (QE) policy already adopted
could completely substitute for the inability to reduce nominal rates
below zero. By enacting a more severe fiscal contraction, and keeping
the inflation target unchanged, the government was effectively taking
this third position. I shall argue in section 3.3 that this was a major
mistake.
3. Macroeconomic outcomes
I will consider outcomes in the same order as innovations, thereby
saving the most important (what happened to the economy and the role of
monetary and fiscal policy in influencing this) to last.
3.1 The Office for Budget Responsibility
There is now a political consensus that the OBR was a successful
innovation. Around the world a growing number of similar organisations
have recently been established. This consensus is, quite rightly, not
based on the accuracy of the OBR's forecasts. Forecasting success
in any particular year is largely down to luck, so any sound judgement
on forecasting ability would require decades of experience (and given
institutional change might not be very meaningful as a result). Instead
the perceived success of the OBR reflects the potential advantages in
contracting out the forecasting process to an independent body. Most
recently the shadow Chancellor, Ed Balls, asked for the OBR to review
its own fiscal plans for 2015 onwards, which indicates a high degree of
cross-party support for the OBR.
A key requirement for this to work is that the OBR is in fact
independent. A concern expressed when the OBR was formed was that this
would be very difficult, because with a relatively small staff (around
25) the OBR would still rely on parts of the government for some of the
detailed fiscal forecasts. The OBR has responded to that concern by
being highly transparent in its dealings with government, and in
discussing in detail the source of its forecast errors. It is early
days, but there is no obvious area where the OBR seems to have been
deliberately misled by a government department. Its forecasts have
suffered from over-optimism, but it has not been unusual in that
respect.
We also have additional evidence of independence. During the first
two years of its programme, the government was fond of suggesting that
OBR forecasts and analysis supported its policy position. In a general
sense this could not be the case, because the OBR was specifically
precluded from examining alternative policies. When the Chancellor
claimed support he was careful not to make false statements about what
the OBR was actually saying. So, for example, he would quote the OBR as
reporting that the poor outturn of the economy compared to forecast was
not in its view due to fiscal austerity, and leave it to others to miss
the key phrase 'compared to forecast'. On the one occasion the
Prime Minister in a speech omitted that qualification, OBR head Robert
Chote wrote to the Prime Minister pointing out his error (see Wren-Lewis
Mainly Macro blogpost (C)). As we shall see, the OBR has documented the
extent to which it believes austerity has reduced output. In addition,
the OBR has not been shy of drawing attention to the unprecedented
extent of the reduction in public spending planned by the Coalition
after 2015.
The Treasury under the previous Labour government had greatly
increased the transparency of its fiscal projections, including
producing for the first time long-term forecasts of the government
accounts. The OBR has continued and extended this practice.
One question that was raised earlier, and which remains for the
future, is whether the OBR should be allowed to examine alternative
aggregate fiscal policies to those chosen by the government. (It does
this in its long-term projections, because the government's fiscal
policy is not defined that far ahead.) The intellectual case for doing
so seems strong. It is natural for the body doing the forecast also to
examine alternative policies, and such analysis would greatly enhance
the public debate. It could be argued that by only allowing the OBR to
analyse its own plans, a government can help smother alternative views.
The potential political costs to the government in allowing the OBR
to examine alternative policies are also clear if the government's
policy is not optimal. It could be argued that in its formative years it
was better to avoid the controversy that would undoubtedly flow from
such analysis, but those years have probably now passed. A similar issue
has arisen as a result of Labour's request that the OBR should cost
its fiscal proposals, a request which the government has refused. The
Netherlands provides one example of an independent institution that, on
request, costs the fiscal programmes of all major parties before an
election.
3.2 Outturns for fiscal measures
Table 1 shows the OBR's projections in 2010 for the
government's current balance, and compares it to the outturn as of
December 2014. The original plan was to eliminate the current deficit by
2015, whereas the latest projection involves a current deficit of 2.6
per cent of GDP by then. Since around 2012, deficit reduction has been
significantly slower than originally intended. As the table shows, this
is equally true if we look at the cyclically adjusted figures.
There has been some debate about whether the slowdown in deficit
reduction around 2012 means that 'Plan A' has changed to
'Plan B'--in other words, has there been a deliberate easing
off in the pace of austerity? The government argues that because there
has been no major change in discretionary fiscal policy, Plan A remains.
The slowdown in the pace of fiscal consolidation instead reflects
unexpected (in 2010) developments: in particular lower tax receipts not
accounted for by cyclical movements. However, it is much more
conventional to look at fiscal outturns in judging the intent of policy.
In the years before the financial crisis, fiscal outturns were also
disappointing relative to forecasts, and the government was (rightly)
criticised for not responding quickly enough to those changes.
This is why the form of the government's primary fiscal
mandate is so important. If its primary fiscal rule had been to achieve
current balance by 2015-16, then policy would have been forced to become
more contractionary to meet that target. However, because the target
involved a rolling 5-year horizon, the government could afford to slow
down the pace of budget consolidation and still remain within its
mandate. As any additional fiscal contraction would have reduced output
(unless monetary policy could have offset its impact) this flexibility
has turned out to be important and useful.
In contrast, the government's secondary target was that debt
should be falling as a share of national income in 2015-16. This is now
not expected to happen until 2016-17, and so this secondary target has
not been met. A possible motivation for the secondary target is that the
main mandate excluded public investment. The path of public net
investment is shown in table 1. It rose to over 3 per cent of GDP in
2008-9 and 2009-10, but was then cut back to an estimated current level
of 1.5 per cent. There is a general consensus that the multiplier (the
total impact on GDP) associated with public investment can be
particularly high, because of beneficial supply as well as demand side
effects.
The actual form of the secondary target, which is that the debt to
GDP ratio should fall in 2015-16, cannot explain why public investment
was cut back so sharply before then. As reduced public investment would
have no direct impact on the primary fiscal mandate (which involves the
current balance), the reason for this decision --which was clearly
harmful in macroeconomic terms must lie elsewhere.
Overall the government's fiscal consolidation was less than
planned, and became 'front-loaded'. Arguments that austerity
was beneficial because the economy grew substantially in 2013--while
pretty ludicrous on a priori grounds--become doubly so if austerity had
slowed significantly by then. It would be more accurate to claim that
abandoning the original pace of fiscal consolidation helped the recovery
in 2013. Although the government's rolling target allowed this
reduction in the pace of fiscal consolidation, it is interesting to ask
why the government chose to allow this to happen. Why didn't the
government cut spending further, or raise taxes, in an effort to meet
its original deficit reduction plans?
The natural answer is that this further tightening would have
damaged a weak economy, but it is difficult for the government to use
this argument while also insisting that the policy implemented in
2010-11 did not delay the recovery. The government could argue that by
2012 panic over financing the deficit had passed. Interest rates on
government debt, which fell during the recession, never picked up
significantly after that date, and by 2012 were lower than during the
recession. It did not make this argument. Thus the pause in the pace of
austerity remains both denied and unexplained.
3.3 Policy's contribution to economic outturns
Figure 1 plots the log of UK per capita GDP. Per capita rather than
total GDP gives a simple measure of average prosperity. The long time
period helps illustrate the unprecedented nature of what happened
following the financial crisis. The recessions of the early 1980s and
the early 1990s were followed by recoveries that eventually made up the
ground lost. I have included a trend line that increases at about 0.55
per cent a quarter, and GDP has in the past always caught up with that
trend. Not only was the recession caused by the financial crisis much
larger, but in the UK there has been no sign of any catch-up back to
that trend. In fact the divergence from that trend increased even when
output began to rise again, and the recovery since 2013 has done nothing
to close the gap. From 2010Q3 to 2014Q3, the average quarterly growth in
GDP per capita has been about 0.25 per cent, which is well below the
pre-recession trend.
[FIGURE 1 OMITTED]
Another way of making the same point is to compare the total growth
in output per capita from 2010 to 2013 of just under 2 per cent to the
two previous recoveries from recessions (both under Conservative
administrations): in both cases, 1981 to 1984 and 1992 to 1995, growth
was over 8 per cent. In short, the performance of the UK economy over
the period of the Coalition government has been a disaster.
Is there any way of coming to a different conclusion about UK
economic performance since 2010? One possibility is to argue that the
underlying trend in output per head began to slow down well before the
Great Recession. This is exactly what the cyclical adjustment methods of
the OECD and IMF (but not the OBR) assume ex post. If that were done
aggressively, you could bend the trend line so that growth since 2010
did not look so bad. However a consequence would be to recast 2006/7 as
the peak of a very substantial boom. This seems completely inconsistent
with other data (as well as assessments at the time), including
unemployment and inflation (see Wren-Lewis, Mainly Macro blogpost (B)
and Broadbent, 2012). There is a tendency to assume that the
pre-recession period must have involved overheating because house prices
rose rapidly, but by that token 2014 would also suggest an overheating
economy, which is clearly nonsense.
Table 2 compares GDP per capita in the UK with the experience in
the US, Japan and the Euro area. Numbers are normalised to be 100 in
2007. The recession had a similar impact in all four areas, but since
then there have been important differences. In 2010 and 2011 recovery
was slowest in the UK. The US and Japan have continued to recover more
rapidly, such that GDP per capita is now above 2007 levels (although of
course still below the pre-recession trend). The Euro Area has entered a
second recession, such that alongside the UK output per capita in 2013
was well below 2007 levels.
In growth accounting terms, this failure to make up the ground lost
in the recession was not due to a failure to reutilise
labour--employment has increased rapidly--but a standstill in labour
productivity growth. Both monetary and fiscal policy can influence
labour productivity in the short term, as firms hold on to labour as
demand falls (or are slow to take extra labour on as demand rises), but
rapid growth in employment is not normally a sign of a shortfall in
demand. However it would be a mistake to conclude that therefore the
stagnation in output that occurred before 2013 was a largely a supply
side phenomenon, and not influenced by monetary or fiscal policy. What
may have happened instead is that firms have substituted labour for
capital as a result of the unusual fall in real wages over this period
(Pessoa and Van Reenen, 2013).
It is of course nonsense to assume simply that this poor
performance was the result of government policy. Experience from other
countries in the past suggests recoveries from financial crises can be
slow with permanent losses compared to pre-crisis trends (IMF, 2009).
(3) The obvious question to ask is whether the austerity programme
announced in June 2010 might have contributed to a tepid recovery. The
OBR has assessed the impact of fiscal contraction on output, as part of
their forecast evaluation report. It suggests that austerity in 2010/11
and 2011/12 reduced GDP growth by about 1 per cent in both years (so the
level of GDP was 2 per cent lower in 2011/12 as a result). As we have
already noted, fiscal consolidation slowed down substantially after
that, so the subsequent contributions to GDP growth are minor by
comparison.
There are good reasons to think that these numbers are too
conservative. The first issue relates to the size of fiscal multipliers.
The OBR uses multipliers that come from historical experience. Jorda and
Taylor (2013), among others, find evidence that fiscal multipliers are
larger in a depressed economy compared to a strong economy. (See also
Bagaria et al., 2012.) They too have calculated the impact of UK
austerity, and have larger (and much more long-lived) numbers than the
OBR. In addition, there are also good theoretical reasons for thinking
that multipliers will also be larger when interest rates are at the ZLB,
because monetary policy finds it more difficult to counteract any fiscal
impact. The second important factor is announcement effects. The OBR
analysis is based on when measures are implemented, rather than when
they are announced. It seems reasonable to assume that the credible
announcement of a five-year period of fiscal contraction in 2010 might
have had a larger immediate impact than just those measures implemented
in 2010. Third, the effects of fiscal contraction when output is
depressed may have significant longer-term hysteresis effects, as DeLong
and Summers (2012) discuss.
These numbers do not suggest that austerity can account for all or
even the major part of the weakness in the UK recovery, but it does seem
appropriate to conclude that it played an important part. Even if we
make the conservative assessment that the cumulative output losses as a
result of austerity since 2010 are of the order of 5 per cent of GDP,
this is a very large figure in terms of lost resources. However, there
remains one important issue to address and that is what monetary policy
might have done if fiscal policy had been looser.
3.4 Monetary policy
The monetary policy operated by the Bank of England and the MPC
since independence is often called flexible (or forecast) inflation
targeting. Lags mean that it would be both difficult and costly (in
terms of output variability) to target current inflation, so it makes
more sense to target expected inflation some years ahead. Although the
period ahead has never been laid down formally, members of the MPC and
the Governor have often used the two-year-ahead reference point, which
is also as far as the Bank's published forecast traditionally went.
Figure 2 plots the inflation forecast two years ahead by date of
inflation report. (This is based on its forecast using market
expectations of future short rates. The Bank does not publish its own
forecasts for short rates.) The inflation target became 2 per cent in
2004, and the forecast has been pretty close to 2 per cent from that
date until the recession. This, and experience until the end of 2003
when the target was 2.5 per cent, are certainly consistent with the
targeting of inflation two years ahead.
[FIGURE 2 OMITTED]
Short-term interest rates hit their UK lower bound of 0.5 per cent
in March 2009. In the same month the Bank started its Quantitative
Easing (QE) programme, buying 200 billion [pounds sterling] of private
sector assets (mainly government debt) between then and January 2010.
Although CPI inflation was well above target at 2.9 per cent in March
2009, it was expected to fall rapidly as a result of the recession
(together with the impact of the temporary cut in VAT in December 2008),
which it did until it hit 1.1 per cent in September 2009. From that
point onwards, CPI inflation started rising again, reaching a peak of
5.2 per cent in September 2011.
This increase in inflation posed a difficult dilemma for the MPC.
The recovery had yet to get going, but a combination of the lagged
impact of the 2007/8 depreciation in sterling, higher commodity prices
and the increases in VAT meant that inflation was well above target. By
the spring of 2011 three members of the committee voted to increase
interest rates. However, by October 2011 the prospects that inflation
would decline rapidly seemed more firmly based, and with the economy
still stagnant the MPC voted to increase QE purchases by an additional
75 billion [pounds sterling]. There were further increases of 50 billion
[pounds sterling] in February 2012 and July 2012. That brought the total
amount of QE to 375 billion [pounds sterling], where it has remained.
Against that background, what might have happened if fiscal policy
had been less contractionary after 2010? The Bank's forecast GDP
growth would have been stronger, and so it is likely that its forecast
for inflation would have been higher. Instead of almost raising interest
rates in early 2011, might the MPC have actually raised rates in 2010?
Would this have been enough to offset the expansionary effect of less
fiscal contraction, leaving the economy no better off (but with
additional government debt)?
Figure 2 suggests an answer. We can see that during 2009 and 2010
the Bank was expecting inflation to be below the 2 per cent target even
after two years, which suggests that because of the ZLB--and despite
QE--they were not able to stimulate the economy enough. For a brief
period in 2011 the forecast goes back to 2 per cent, before falling
again in 2012. This is consistent with the operation of the QE
programme, which was expanded during periods in which inflation was
expected to be below 2 per cent even after two years.
This suggests that if the prospects for output growth had been
moderately stronger in 2009, 2010 or 2012, the Bank may have taken no
action, and all that would have happened is that expected inflation
would have been closer to target. Only in 2011 might the MPC have
reacted to higher expected growth and inflation by raising rates. (It
was early 2011 when the MPC actually came close to doing so.) It
therefore seems unlikely that monetary policy would have offset the
impact of any fiscal easing except for a brief period in 2011.
There are additional qualifications that need to be made to the
monetary policy offset argument. First, inflation would have been lower
without the 2010 increase in VAT. Perhaps the MPC simply 'saw
through' this effect, and also assumed that wage setters would do
the same, but maybe not. Second, the offset argument implies either very
good forecasts by the Bank, or an absence of lags. Even if the Coalition
government had tightened less in mid-2010, and interest rates had been
increased sometime in 2011, the impact on growth might not have been
felt until later, by which time a more robust recovery might have been
underway.
Fascinating though such counterfactuals are, I think they miss the
essential point of criticism of government policy. Once you make the
reasonable assumption that QE is not a complete substitute for lower
nominal interest rates, the fiscal contraction of 2010 took a large risk
with the economy. It is not so much that the OBR's forecast was
wrong, and that therefore austerity was a mistake. Forecasts are often
wrong, and good policy allows for these risks. Once you take the ZLB
seriously, good policy does not take risks with the economy by
implementing a substantial fiscal contraction at the ZLB unless
something forces you to do so. What monetary policy might or might not
have done in 2011 is beside the point, because decisions in 2010 cannot
have anticipated subsequent events. Interest rates might have been
increased in 2011 if there had been less fiscal austerity, but if
inflation had not risen so quickly (which was unexpected) rates need not
have increased. Once again, good policy anticipates risks.
The only justification for ignoring the ZLB when undertaking fiscal
austerity and retaining inflation targeting is if QE is a complete
alternative to varying the nominal interest rate. Here we can cut
through a great deal of discussion of very tentative econometric results
based on very little evidence (and which are still hotly debated--see
Woodford, 2012) by talking about uncertainty. Even if it is possible to
duplicate the impact of nominal rate cuts by an appropriate amount of
QE, we have virtually no idea of what that required quantity of QE is.
As a result, using QE to regulate the economy means that the economy is
subject to much more risk than if you were using conventional monetary
(or fiscal) policy. So once again, good policy does not take risks with
the economy by implementing a substantial fiscal contraction when
interest rates are at (or even near to) their ZLB.
The problems associated with the ZLB were well known to
macroeconomists as a result of Japan's 'lost decade'. As
we noted in section 2.3, both Paul Krugman and Michael Woodford had
proposed a means by which the impact of the ZLB could be reduced using a
modification of the inflation target strategy. Essentially the central
bank would promise to let inflation go above target (and the output gap
be positive) sometime in the future, when the ZLB constraint no longer
applied. To the extent that the private sector is forward looking,
expectations of higher future inflation (and lower future short-term
interest rates) would reduce long-term interest rates today, as well as
stimulating inflation, consumption and investment today. As a result,
the impact of the recession today would be reduced.
One difficulty with this 'promise to be irresponsible' in
the future (Krugman's phrase) is that it is time inconsistent. Once
the work of the promise in moderating the recession has been done, the
central bank has an incentive to go back on its promise and keep
inflation at target. If the public understands that, it may not believe
the promise. This difficulty might be reduced if the policy could be
embodied in an alternative to an inflation target, by having a price
level target or a (level of) nominal GDP target for example.
None of these possibilities could have been implemented by the MPC
alone, because they all conflict with their mandate. However the
Coalition appears to have had no appetite for a change in the inflation
target regime, despite the weak recovery up to and including 2012. In
March 2013 the government published a 'Review of the Monetary
Policy Framework'. This did three things. The first was to
re-affirm commitment to the 2 per cent inflation target, and thereby
reject alternatives such as nominal GDP targeting. The second was to say
that government would have no problem with the MPC targeting inflation
three (or more) years ahead, rather than the 2-year-ahead target that it
appeared to have been following. The third was to give the green light
to forward guidance, which the new governor Mark Carney was known to
favour. Here forward guidance (which the US Fed had already implemented)
is quite different from the kind of forward commitment to excess
inflation discussed above. These changes are best regarded as minor
tweaks to an inflation targeting regime rather than any radical shift in
monetary policy.
Was the Coalition too conservative in retaining the inflation
targeting regime, particularly when interest rates were at their lower
bound and fiscal policy was ruled out as a stabilisation tool? Although
the alternative of targeting the level of nominal GDP was gaining
converts among the academic community throughout this period, any final
verdict on this will have to wait until we have some practical
experience with such a scheme.
3.5 Funding for Lending and Help to Buy
The Coalition did introduce two important innovations on the
interface between monetary and fiscal policy in 2012 and 2013: the
Funding for Lending (FFL) scheme introduced in July 2012, and the Help
to Buy (HTB) scheme, the full version of which came into effect in 2013.
They share a common feature, in that they involve providing incentives
to increase the amount of private sector borrowing. FFL involved
providing subsidies to banks to increase lending to both households and
companies. HTB is targeted at house purchase, and in part offers to
guarantee up to 20 per cent of a loan. The economic rationale for both
measures is similar. The financial crisis may have made banks too risk
averse in either the amount of lending they were prepared to do or the
conditions required for loans (such as a loan-to-value ratio for
houses). By providing incentives to lend, or additional insurance, the
government could counteract the negative effect this risk aversion was
having on aggregate demand.
In 2013 the pace of the recovery increased, with successive
quarterly growth rates of between 0.5 per cent and 0.9 per cent. We have
already noted that, according to the OBR's analysis, fiscal policy
as conventionally measured ceased to be a drag on growth by 2012, but
nor was it a significant stimulus. Net trade contributed little to
growth in 2013, so the pickup came from domestic demand. It is beyond
the scope of this paper to assess how much either FFL or HTB may have
contributed to this. It is possible, for example, that the pickup in
consumption, which in turn reflected a decline in the savings ratio,
might have occurred anyway as consumers came to the end of a period of
'balance sheet correction' following the financial crisis.
It is also possible that one or both schemes had a significant
effect on demand. For example, by effectively reducing the amount of
deposit a homebuyer required from 25 per cent to 5 per cent, this would
free up a lot of savings that could now be consumed. However, the HTB
scheme proved highly controversial among economists. A major concern was
that, by linking future fiscal outcomes more closely to house prices,
the scheme provided an incentive for governments to keep house prices
high when the optimal policy might be to take measures that have the
opposite effect. Others questioned why the government felt it was able
to create a contingent liability that might have significant future
fiscal consequences, while at the same time stressing the need to reduce
current government borrowing.
4. Conclusions
The Coalition government introduced two important and successful
innovations into fiscal policymaking. The first was establishing the
OBR, albeit with a fairly restricted mandate. The second was the form of
its main fiscal mandate, which in normal times I would argue is a
sensible way to conduct fiscal policy. The tragedy for the Coalition
government was to implement the rule at the one time that it should have
been cast aside, when interest rates were at their lower bound.
Sometimes governments are unlucky, and get blamed for macroeconomic
outcomes they could do little about. A part of the stagnation in UK
labour productivity since the recession may turn out to be a case in
point. In other times governments take macroeconomic risks that they get
away with, and history is forgiving even though it should not be. The
analysis in this paper suggests that the enhanced UK austerity announced
in 2010 is neither of those: it did take a risk with the UK economy and
this risk materialised. Austerity reduced GDP growth by at least 1
percentage point in both 2010 and 2011.
A natural question is why this mistake was made. Other governments
also undertook austerity at around the same time. In the United States
the fiscal contraction occurred a year later, which may have allowed its
recovery to gather pace. The Eurozone is still awaiting a recovery as it
has pursued austerity with much more vigour. The fact that many
governments made the same mistake, introducing fiscal contraction well
before recovery was complete and while interest rates were stuck at the
ZLB, suggests a common cause.
One obvious common factor was the Eurozone debt crisis of 2010.1
suspect that had an important impact on many policymakers, and it may
well have been influential in persuading the Liberal Democrats to change
their mind on austerity as part of the Coalition agreement. It may have
influenced central bankers in the advice they gave, and led them to be
overoptimistic about the potential effects of QE.
However there are two problems in using the Eurozone crisis as an
explanation for the Coalition's error on fiscal policy. The first
problem is that by 2011 it had become clear that any debt funding crisis
was specific to Eurozone economies, and arose because the ECB was
unwilling to act as a sovereign lender of last resort. In the EIK and
US, interest rates on government debt were falling, and there was no
prospect of a bond market panic. With the prospect of a double-dip
recession in the UK, 2011 was the obvious time for a publicly announced
policy reversal, but instead we had to wait another year for fiscal
consolidation to slow down, and the public line remained that austerity
was continuing. The second problem is Conservative policy before they
entered government.
The Chancellor appears not to have believed that fiscal policy
might need to take account of the state of the economy even at the
height of the recession. The Conservative opposition argued in 2009
against the countercyclical measures undertaken by the Labour
government, and their justification for doing so was in effect to say
that the ZLB was not a problem. This view did not reflect the consensus
among academic macroeconomists at the time. Why the Chancellor chose to
take this unconventional view is beyond the scope of this paper to
assess. He proposes to renew the pace of fiscal consolidation over the
next few years, despite interest rates remaining at or close to their
lower bound, which suggests he and the Prime Minister continue to hold
this unconventional view.
The delay in the UK recovery over the first part of the Coalition
government's term is at least in part a result of the
government's fiscal decisions. I have argued that these decisions
were a mistake not just in hindsight but when they were taken. It will
be many years before we can settle on a figure for the total cost of
that mistake, but measured against the scale of how much governments can
influence the welfare of its citizens in peacetime, it is likely to be a
large cost.
REFERENCES
Alesina, A. and Ardagna, S. (2009), 'Large changes in fiscal
policy: taxes versus spending', National Bureau of Economic
Research (NBER), Working Paper No. 15438.
Bagaria, N., Holland, D. and van Reenen, J. (2012) 'Fiscal
consolidation during a depression', National Institute Economic
Review, 221, pp. F42-54.
Broadbent, B. (2012), 'Deleveraging', Speech 15th March,
Bank of England.
Calmfors, L. and Wren-Lewis, S. (2011), 'What should fiscal
councils do?', Economic Policy, 26, pp. 649-95.
De Grauwe, P. (2011), 'The European Central Bank as a lender
of last resort', VoxEU 18/08/11, or (2011), 'The management of
a fragile Eurozone', Working Paper, CEPS, April.
DeLong. J.B. and Summers, L.H. (2012), 'Fiscal policy in a
depressed economy', Brookings Papers on Economic Activity, Spring,
pp. 233-74.
Eggertson, G.B. and Woodford, M. (2003), 'The zero bound on
interest rates and optimal monetary policy', Brookings Papers on
Economic Activity, Economic Studies Program, The Brookings Institution,
34(1), pp. 139-235.
Frankel, J.A. and Schreger, J. (2013), 'Over-optimistic
official forecasts and fiscal rules in the Eurozone', Review of
World Economics, 149, 2, pp. 247-72.
IMF (2009), 'What's the damage? Medium term output
dynamics after financial crises', World Economic Outlook, October,
pp. 121-51.
--Independent Evaluation Office (2014), 'IMF response to the
financial and economic crisis'.
Jorda, O. and Taylor, A.M. (2013), 'The time for austerity:
estimating the average treatment effect of fiscal policy', NBER
Working Paper No. 19414.
Kirsanova, T., Leith, C. and Wren-Lewis, S. (2009), 'Monetary
and fiscal policy interaction: the current consensus assignment in the
light of recent developments', Economic Journal, 119, pp.
F482-F496.
Krugman, P. (1998), 'It's baaack! Japan's slump and
the return of the liquidity trap', BPEA, 2:1998, pp. 137-87.
--(2014), 'Currency regimes, capital flows and crises',
IMF Economic Review, forthcoming.
Pessoa, J.P. and Van Reenen, J. (2013), 'The UK productivity
and jobs puzzle: does the answer lie in labour market
flexibility?', C.E.P. Special Paper No.3l.
Portes, J. and Wren-Lewis, S. (2014), 'Issues in the design of
fiscal policy rules', Oxford Economics Department Discussion Paper
No. 704 and NIESR Discussion Paper No. I 1865.
Woodford, M. (2012), 'Methods of policy accommodation at the
interest-rate lower bound', mimeo.
Wren-Lewis, S. (1996), 'Avoiding fiscal fudge', New
Economy, 3, pp 128-132.
--(2013a), 'Comparing the delegation of monetary and fiscal
policy', in Kopits, G. (ed.), Restoring Public Debt Sustainability,
Oxford University Press.
--(2013b), 'Aggregate fiscal policy under the Labour
government, 1997-2010', Oxford Review of Economic Policy, 29, pp.
25-46.
--Mainly macro blogposts. (A) 10/07/13 (B) 05/08/14 (C) 05/03/13.
NOTES
(1) In December 2014 the Coalition proposed changing this to a
three-year rolling target, a time period that is in danger of being too
short to adjust optimally to persistent shocks.
(2) Osborne, G and Sachs, J. (2010), "A frugal policy is the
better solution", Financial Times, 14 March. My thanks to John
McHale for bringing this to my attention.
(3) This IMF study does however note (p. 123) that "the
evidence suggests that economies that apply countercyclical fiscal and
monetary stimulus in the short run to cushion the downturn after a
crisis tend to have smaller output losses over the medium run."
Simon Wren-Lewis, * Oxford University, Blavatnik School of
Government. E-mail: simon.wren-lewis@bsg.ox.ac.uk. I am grateful to
Giles Wilkes and Simon Kirby for valuable comments, but responsibility
for the end result is entirely mine.
Table 1. Fiscal aggregates, forecasts and outturns
Current balance December 2014 Cyclically adjusted
June 2010 Statement June 2010
Budget Budget
2008-9 -3.5 -3.4 -3.1
2009-10 -7.5 -6.9 -5.3
2010-11 -7.5 -5.9 -4.8
2011-12 -5.7 -5.0 -3.2
2012-13 -4.0 -5.0 -1.9
2013-14 -2.3 -4.2 -0.7
2014-15 -0.9 -3.5 0.3
2015-16 0.0 -2.6 0.8
December 2014 Public net investment
Statement December 2014
Statement
2008-9 -3.4 3.2
2009-10 -4.8 3.3
2010-11 -3.9 2.5
2011-12 -3.1 1.9
2012-13 -3.1 2.1
2013-14 -2.6 1.5
2014-15 -2.7 1.5
2015-16 -2.2 1.4
Sources: OBR Economic and Fiscal Outlooks; June 2010 (Table CI) and
December 2014 (Table 4.47, page 189). Supplementary data from OBR
Public finances databank (Dec 14).
Table 2. GDP per capita (2007=100)
2008 2009 2010 2011 2012 2013
UK 98.9 93.9 95.0 95.7 95.7 96.8
US 98.8 95.2 96.8 97.7 99.2 100.7
Japan 99.0 93.6 98.1 97.4 99.0 100.7
Euro 12 99.8 95.1 96.7 97.9 97.0 96.5
Sources: UK--ONS, Euro 12--Eurostat, US and Japan--FRED.
