Readings-Topics
for Article Review
Intermediate
Macroeconomics ECOS2002
Semester
2, 2013
You must choose one of the following four readings to
write an article review for
the assignment. Your review should focus on the topic which the article is
primarily concerned with. Remember the word limit for the review (not counting
bibliography) is 800 words so you need to be concise and focus on what you
consider to be most important. Harvard style referencing and always reference
if you utilise other sources.
Article Review
The purpose of an article review is to provide a summary and evaluation of a piece of
writing. When a lecturer reads an article review written by a student, they want to see
evidence that the student has not only understood the topic of the article, but is able to
evaluate the article in relation to their own knowledge of the topic and other relevant
knowledge in the field.
ARTICLE:
The Debate on Expansionary Fiscal Consolidation: How
Robust is the Evidence?
Headnote
Abstract
This
paper critically examines the key empirical evidence used to support the fiscal
consolidation argument, complemented by a brief assessment of the limitations
of the analytical foundation of the growth promoting benefits of the fiscal
consolidation thesis. It also reviews the evidence on the debt-growth
relationship at some length. It finds that the negative relationship between
debt and GDP growth is influenced by outliers or exceptionally high debt-GDP
ratios. It also points out that the composition of public debt matters.
Additionally, the debt-GDP relationship appears to be non-linear – positive
first and turning to negative, but there is considerable variation in the
estimated turning or ‘tipping’ point, which is not helpful as a policy guide.
Historical evidence does not lend support to the concerns that the current
situation is likely to cause rapid upward spiraling of public indebtedness.
Finally, the argument that fiscal consolidation is possible without adversely
affecting growth is not based on robust empirical evidence. This conclusion is
reinforced by a succinct overview of some country-specific experiences
(Denmark, Ireland and United States).
JEL Codes: N14, H63, E31, E62, E63, E65
Keywords
Debt;
Europe; fiscal consolidation; growth.
Introduction
The
2007-2009 global economic and financial crisis led to a sharp increase in
public debt across various parts of the world, especially in advanced
countries. This has heightened concerns about fiscal sustainability and its
broader economic and financial market consequences. In particular, many believe
that public debt is starting to hit levels at which it might slow down economic
growth. These concerns have already triggered further downgrading of the
sovereign debt ratings of several developed economies, including Greece,
Ireland, Japan, Portugal and Spain. One rating agency, Standard & Poor’s,
has recently presented a negative outlook for the debt burden of France, and
downgraded its triple A credit rating, following its downgrading of the United
States in 2011. On 13 February, 2012, Moody’s Investor Service downgraded its
credit ratings for Italy, Portugal and Spain, while France, Britain and Austria
kept their top ratings but had their outlooks dropped to ‘negative’ from
‘stable’.
The consensus
among the major countries represented in the Group of 20 (G20) in the wake of
the current global economic and financial crisis that propelled them to
announce coordinated stimulus packages quickly disappeared with the first signs
of a tepid recovery. Following the 2010 Toronto Summit of the G20 leaders, many
countries, especially in Europe, have committed to fiscal consolidation in
response to these concerns in the form of time-bound and targeted reductions in
the structural budget deficit. For example, fiscal austerity measures in the
United Kingdom aim to cut the deficit by an estimated 8 percentage points of
GDP by 2015.1 In France, the medium-term fiscal plan is to reduce the deficit
to 3 per cent of GDP by 2013 through a mix of revenue and expenditure measures
as well as structural reforms. Fiscal consolidation in Germany started in 2011.
The Stability and Growth Pact (SGP) of Germany requires lowering the general
government deficit to 3 per cent of GDP by 2013; and the constitutional rule
mandates a 0.35 structural deficit at the federal level by 2016 (to be achieved
in roughly equal annual steps) and balanced structural budgets at the state
level by 2020.2 Canada’s fiscal consolidation plan projects a return to a small
deficit of -0.1 per cent of GDP by 2014-15. The winding-down of the Action Plan
would cut the federal budget deficit in half by 2011-12.
The
policy discourse, most notably in the rich nations, is that governments must
now engage in fiscal consolidation and bring back public finances to
sustainable levels. As the Economist (2010a) observes: Across much of the rich
world an era of budgetary austerity beckons’. But signs of budgetary austerity
also seem to be emerging in a sizeable number of low and middle-income
countries. One study by UNICEF (2010) finds that, in a sample of 86 low and
middle income countries, about 40 per cent are engaging in reductions in public
expenditure in 2010-201 1 relative to 2008-2009.The average projected spending
cuts are around 2.6 per cent of GDP. An OXFAM study (Kyrili and Matthew 2010)
shows that the global economic and financial crisis of 2007-2009 has created a
huge ‘fiscal hole’ in the 56 low-income countries (LICs) by reducing their
budget revenues by $65bn over the 2009-2010 period. As a result of the fiscal
hole, most LICs are cutting or at least restraining public expenditure,
especially on education and social protection.
The
International Monetary Fund’s Fiscal Monitor (IMF 2010c) highlighted the need
for major fiscal consolidation over the years ahead. The monitor stated that
though the increase in budget deficits played a key role in staving off an
economic catastrophe, as economic conditions improve, the attention of
policymakers should now turn to ensuring that doubts about fiscal solvency do
not become the cause of a new loss of confidence. Moreover, an equally
important risk to be averted is that the accumulated public debt, even if does
not result in overt debt crises, becomes a burden that slows down long-term
potential growth.
This
paper critically examines the key empirical evidence that is assembled to
support the fiscal consolidation argument. This examination is complemented by
a brief assessment of the limitations of the analytical foundation of the
growth promoting benefits of the fiscal consolidation thesis. We point out that
the composition of public debt matters, especially if it is used to finance
growthpromoting public investment. We also review the evidence on the
debt-growth relationship at some length. We find that the negative relationship
between debt and GDP growth is influenced by outliers or exceptionally high
debt-GDP ratios. Furthermore, when there is a relationship, it appears to be
non-linear – positive first and turning to negative at some point, but there is
considerable variation in the estimated turning or ‘tipping’ point, which is
not helpful as a policy guide. Historical evidence does not lend support to the
concerns that the current situation is likely to cause rapid upward spiraling
of public indebtedness that will push up interest rates (as well as risk
premium) on government securities, thereby putting greater pressure on deficits
to widen and on public debt to increase. Finally, we also find that the
argument that fiscal consolidation is possible without adversely affecting
growth is not based on robust empirical evidence. This conclusion is reinforced
by a succinct overview of some country-specific experiences (Denmark, Ireland
and the United States).
Debt-Growth
Relationship
The IMF’s
Fiscal Monitor (IMF 2010c) that strongly advocates fiscal consolidation
acknowledges that to date, there are only a few studies that assess the
magnitude and significance of potentially adverse effects of high public debt
on growth. Thus, the IMF attempted to fill this gap by undertaking empirical
analysis of the relationship between initial government debt and subsequent
economic growth in a panel of advanced and emerging economies for the period
1970-2007.3 It involved examination of nonlinearities and threshold levels beyond
which debt begins to have an adverse effect on growth. It also did a growth
accounting exercise to explore the channels through which government debt may
influence growth. The analysis paid particular attention to a variety of
estimation issues – such as ‘reverse causality’ or the presence of a third
variable affecting both growth and debt – that can have an important bearing on
the estimation. The study also undertook various robustness checks. Therefore,
the IMF study seeks to provide a solid empirical foundation for fiscal
consolidation.
Yet a
closer look at the scatter plot (IMF 2010c: 63 Figure 1) reveals that the claim
of a negative relationship between growth and initial debt-GDP ratio is
influenced by a few outliers, characterised by either a debt-GDP ratio well
above 100 per cent or very high per capita GDP growth exceeding 8 per cent.
This is not surprising, given the insignificant existence of any relationship
between the debt-GDP ratio and macroeconomic instability, as a closer look at
Figure 4 (IMF 2010c: 67) exposes.
As part
of our critical examination of the pertinent evidence on public debt and
growth, we construct a simple scatter plot between initial debt-to-GDP ratios
and subsequent years of growth for the 1981-2010 period using a large sample of
countries. As can be seen from Figure 1, the slope of the debt-growth ‘line of
best fit’ is rather shallow as it was found to be in the IMF’s Fiscal Monitor.
To reinforce this point we construct bar diagrams with median debt-GDP ratios
and median growth rates for different sub-periods in Figure 2. As can be seen,
despite the major differences in median debt-GDP ratio, the differences in
median growth rates are not significant or pronounced. This is also evident
from the coefficient of variation (-0.34) for median growth versus 0.71 for
median debt-GDP ratio. Interestingly, the median growth rate of countries with
a debtGDP ratio between 90 and 120 per cent increased from around 2 per cent
during 1981-1985 to around 4 per cent during 2006-2009. After a median growth
rate of about 2 per cent until 1995, the median growth continued to rise
despite a high median debt-GDP ratio.
It seems
that the claimed negative debt-growth relationship is due to extreme values or
outliers. To substantiate this point, we use this sample to identify countries
that have the highest median debt-to-GDP ratio (approximately 171 per cent) for
the 1981-2009 period and contrast them with countries that have the lowest
median debt-to-GDP ratio (approximately 19 per cent) by focusing on median
growth rates for the two groups. As can be seen in Figure 3, a more than
nine-fold increase in the median debt-to-GDP ratios is associated with a 2.1
percentage point decline in the growth rate.4 This is, of course, an extreme
scenario and unlikely to represent the norm in the global evolution of public
indebtedness. Indeed, the dominant trend since the mid-1990s is that the
majority of countries lie in the ‘moderate’ (30 to 60 per cent) and ‘low’ debt
categories (less than 30 per cent) – see Figure 4. This trend is worth bearing
in mind as this can be easily forgotten in the current alarmist discourse over
rising public indebtedness in a few countries.
The
absence of a clear negative debt-growth relationship raises considerable doubts
about the nature of the theoretical and empirical foundations of those who
espouse the cause of fiscal austerity. The consensus’ view of mainstream
macroeconomics – derived from synthesis of ‘new classical’ and ‘new Keynesian’
approaches – is that there is no role for counter-cyclical fiscal policy. The
latter should focus primarily on debt and deficit management, while monetary
policy deals with ‘business cycle stabilisation and inflation control’
(Kirsanova et al. 2009: 482). Even its proponents argue that this works well in
the case of modest demand shocks, but when monetary policy is constrained (such
as a liquidity trap – entailing the so-called ‘zero bound’ on nominal interest
rate – or where there is a currency union) in the face of large demand shocks
(which characterizes the Great Recession of 2008-2009), then focusing on this
consensus assignment of macroeconomic policy instruments can turn out to be
welfare-reducing (Blanchard et al. 2010; Wren-Smith 2012).
Concerns
about the effectiveness of fiscal policy- and the need to focus on debt/deficit
management – can also be rationalised in terms of three views: (a) crowding
out’ (b) ‘Ricardian equivalence’ and (c) ‘market confidence’. The crowding out
hypothesis, i.e., government borrowing drives up interest rates and adversely
affects private investment, ignores the consequences (e.g. low profitability,
bankruptcies etc.) of a depressed economy in the absence of increased
government spending. It also ignores the productivity enhancing impact of
government spending on infrastructure, education, and research and development
(R & D) – a point taken up further later. In any case, even in a simple
IS-LM model, ‘full’ crowding out only takes place in the special case when the
LM curve is vertical. As long IS and LM curves have normal shapes, there will
be ‘partial’ crowding out which means that a fiscal expansion will raise
aggregate demand and output and facilitate the move towards ‘full employment’.5
The
Ricardian equivalence argument that individuals have to save more in
anticipation of higher future tax to redeem public debt is spurious as revenues
rise with an expanding economy.6 What needs to be noted is that Ricardian
equivalence applies to tax cuts not to changes in government expenditure. As
Wren-Smith (201 1: R8) has pointed out, in a Ricardian economy, ‘ . .. fiscal
expansion that involves cutting lump sum taxes would have no impact on demand,
because the tax cut would be saved. However, exactly the same model implies
that a temporary increase in government spending will increase demand.’
It is
also a myth that government expenditure must always be financed by raising tax,
so that government budget remains in balance. As Abba Lerner pointed out more
than half a century ago, ‘taxing is never to be undertaken merely because the
government needs to make money payment’ (Lerner 1943: 40, emphasis in
original). Just as government expenditure, taxes must also be judged from their
impact on the economy.
It is
also not necessary that governments always have to borrow from the public. It
is much easier for governments to increase spending capacity by printing money
(borrowing from the central bank). The immediate financial implications of
expansionary fiscal policy action when the central bank uses interest rate – in
a world of endogenous money’ – is to add to the cash reserves of the private
sector banks in which the government checks are deposited. This in turn
increases (net) liquidity in the cash market where the central bank discount
rate is established and defended, assuming that the central bank did not
implement offsetting market operations (buying and selling its own or
government short-term securities, or associated derivatives such as re-purchase
agreements). Under these circumstances the actual central bank discount rate
should tend to decrease, causing downward pressure on retail interest rates.7
This conclusion is contrary to the conventional belief, and therefore should
encourage instead of crowding out private investment.
The
immediate objection to the option of borrowing from the central bank comes from
the supposed link between printing money and inflation. However, this fear
again arises from the lack of appreciation of the root cause of the problem,
i.e. insufficient spending. Printing money in a depressed economy should not
cause inflation. The US Federal Reserve Bank has gone into top gear, called
quantitative easing, since the onset of the Great Recession. Instead of
accelerating, the inflation rate remains well below the ‘desired’ rate of 2-3
per cent. Japan, too, has been facing deflationary pressure despite significant
easing of monetary policy since 1995. After reviewing a mass of evidence, Bruno
and Easterly (1998: 3) arrived at the following answer to the question ‘Is
inflation harmful to growth? The ratio of fervent beliefs to tangible evidence
seems unusually high on this topic’. Likewise, there is scant evidence
supporting the negative impact of fiscal expansion either due to crowding out
or Ricardian equivalence. After surveying a large body of empirical literature
an IMF study, Hemming, Kell and Mahfouz (2002: 36) concluded that estimates of
fiscal multipliers are overwhelmingly positive . . . there is little evidence
of direct crowding out through interest rates and exchange rate. Nor does full
Ricardian equivalence or a significant partial Ricardian offset get much
support from the evidence’.
Finally,
the least theoretically grounded but most influential view is that fiscal
austerity now is necessary because it will instill ‘market confidence’ that
lies at the core of private sector spending decisions. As Jean-Claude Tritchet,
the former President of the European Central Bank put it during a media
interview: ‘It is an error to think that fiscal austerity is a threat to growth
and job creation. At present, a major problem is the lack of confidence on the
part of households, firms, savers and investors who feel that fiscal policies
are not sound and sustainable’ (ECB 2010). As Nobel Laureate Paul Krugman has
often lamented, this represents an undue faith in the confidence fairy’ to spur
growth. Borrowing costs are at a historic low for advanced countries, such as
UK, USA and Japan, despite high public debt. This probably reflects the fact
that there is a flight to safe assets issued by advanced country governments
who ‘still own their currency’ (Krugman 2012).
One
should also note that the advocates of the ‘market confidence’ thesis overlook
the fact that rating agencies typically induct growth variables in their
assessment of sovereign risk analysis. More importantly, studies have shown
that the impact of a growth contraction on measures of sovereign risks is
higher than the impact of debts and deficits on such risks. Hence, when fiscal
consolidation leads to growth contractions they reduce rather than raise market
confidence. It is perhaps not surprising to learn that in the vast majority of
cases (at least for advanced countries) successful fiscal adjustment took place
during booms and normal growth periods (Cotarelli and Jamarillo 2012; Islam and
Hengge 2012).
Composition
Matters
The
problem with the simple debt-growth analyses is that they ignore other factors
that affect growth. Kumar and Woo (2010) include variables that are commonly
used in growth regressions, but their reporting is partial. For example, in
their BE (Between Estimator) model, the coefficient of the initial years of
schooling is 4.2, which is significant at 1 per cent level. This is
significantly larger than the coefficient of the initial debt-GDP ratio.
However, when they claim that ‘a 10 percentage point increase in the initial
debt-to-GDP ratio is associated with a slowdown in annual real per capita GDP
growth of around 0.2 percentage points per year’ (2010: 4), they do not
consider how the initial debt was spent. If the initial debt was incurred to
improve the initial years of schooling, certainly the likely negative impact of
high initial debt would be more than offset, leaving a large net impact of 42
percentage points. Likewise, the initial high debt should affect the size of
government which has a coefficient of 0.1 (significant at 5 per cent level).
This too would lead to a net positive impact of increased public indebtedness
on growth. The story does not change much between different estimation
techniques employed and in the robustness check using the parsimonious
specification.
The
importance of composition of public expenditure is also revealed in the
econometric exercise reported in the IMF’s World Economic Outlook 20 10 (IMF
2010b). It finds that the estimated impact on output of fiscal consolidation
based on cuts to government transfers are relatively benign, whereas for
adjustments based mainly on cuts to government consumption or investment, the
output costs are larger.
What the
above discussion shows is that even if there is a downward sloping debt-growth
relationship curve, when we consider the growth effect of government s
productive investment expenditure and consequent crowding-in effect of private
investment, the curve will shift to the right, producing a positive
relationship between public debt and growth. This can be illustrated by using a
simple diagram.
Let us
assume that the public debt-growth relationship is negative as in Figure 5. Let
us suppose an economy is growing at gl and the corresponding debt-to-GDP ratio
is DGR1. Now, suppose that there is a secular increase in the public debtto-GDP
ratio from DGR1 to DGR2. If we simply focus on a movement along a given public
debt-growth curve (from A to B), then the prediction is that the growth rate
will decline from g1 to g2. Indeed, this will be the typical conclusion of any
study that focuses only on the partial impact of debt on growth. On the other
hand, the composition of public debt matters a great deal.8 If we assume that the
increased public debt will finance productivity enhancing investments in
infrastructure and in enhancing public service delivery, then this would be
represented as a rightward shift of the downward sloping public debt-growth
curve and thus create the possibility of a movement from ? to C. Incorporating
this ‘shift’ effect thus means a net positive impact of higher debt on growth.
Empirical studies that ignore ‘shift’ effects will thus arrive at misleading
conclusions on the debt-growth relationship.
In sum,
both the size and composition of public debt matter, as pointed out by Domar
more than half century ago. Domar ( 1 944) in the mid- 1 940s showed that
economic growth is not neutral with respect to the level and composition of
government expenditure. He notes ‘that deficit financing may have some effect
on income . . . has received a different treatment. Opponents of deficit
financing often disregard it completely, or imply, without any proof, that
income will not rise as fast as the debt …. There is something inherently odd
about any economy with a continuous stream of investment expenditures and a
stationary national income’ (Domar 1944: 801, 804).
Tipping
Point
Reinhart
and Rogoff (2010) summarised evidence from 44 developed and developing
economies, and found a threshold of 90 per cent for central government debt to
GDP, after which the real growth rate declines. Although this study received
considerable attention in the press, which has referred to it as a ‘tipping
point’ (Pozen 2010), it uses histograms to describe some stylised facts, and
hence cannot ascertain causality with any statistical significance. As growth
declines, debt-GDP ratio rises, so the causality may run from low growth to
high debt-GDP ratios.
An
examination of various studies that provide an econometric investigation of the
so-called ‘tipping point’ enables one to come to the following conclusions.
First, the estimated thresholds vary widely – from as low as 15-30 per cent to
64 per cent of GDP for developing countries; for developed and emerging
economies, the threshold ranges from 60 per cent to 90 per cent. It is
difficult to advise policy-makers based on these estimates. More specifically,
it becomes difficult to sustain the view that one can use reliable prudential
targets to monitor the sustainability of public debt.
Second,
at least one study (Caner, Grennes and Koehler-Geib 2010) notes that the
coefficient below the threshold (64 per cent of GDP) is much larger than that
above the threshold (+0.065 vs. -0.0174), implying that crossing the threshold
is costly in terms of lost growth, but pushing below the threshold is even more
costly. Furthermore, the high cost of lost output (reduction in growth rate)
highlighted in that study is due to one extreme case (Nicaragua) and is also
based on the strong assumption that a high degree of public debt will persist
for nearly thirty years. For most countries, the estimated annual percentage
point loss in real GDP growth is small.
Finally,
other studies draw attention to the notion of ‘debt irrelevance’. In other
words these studies (Cordello et al. 2005; Presbitero 2010) show that the
public debt-growth relationship flattens beyond the threshold, implying that
rising public debt has no statistically significant impact on growth after a
certain point, suggesting that countries perhaps learn to live with debts.
Risk of
Debt Spiralling
The IMF’s
Fiscal Monitor 2010 (IMF 2010c) warns that without progress in addressing
fiscal sustainability concerns, high levels of public indebtedness could weigh
on economic growth for years. If governments fail to signal a credible
commitment to reduce debt ratios, the resulting increase in interest rates (and
decline in growth rates) could put greater pressure on deficits to widen and on
public debt to increase. These kinds of dynamics have clearly affected Greece,
Ireland, Portugal, Spain and several economies in Eastern Europe, countries
that still have a relatively limited tax capacity, making the vicious forces at
work more powerful. However, despite these experiences, there is scant evidence
supporting strong dynamics between public indebtedness and the cost of
servicing the debt in developed countries. Mounting public debt in the United
States, the major economies of the Euro area and Japan has not pushed real
interest rates up; the real interest rates have even declined in some countries
– as was noted above.
The
historical evidence also does not support the claim for such dynamics to emerge
under all circumstances. For example, interest rates have remained low since
the late 1980s in Japan, where public debt soared to 200 per cent of GDP during
two decades of deflation. The higher debt-to-GDP ratio in Japan is partly due
to very low inflation. A higher, but still moderate, inflation rate will raise
nominal GDP and lower the public debt-to-GDP ratio unless there is an actual
increase in the government’s gross liabilities. In 1988, Belgium had the
highest public debt. Italy’s debt also shot above 100 per cent of GDP during
this period (OECD 2009). None of these countries experienced spiraling
inflation or very high interest rates as is commonly feared when government
fiscal deficits rise. Japan is facing just the opposite – deflationary pressure
and a zero interest rate.
At the
end of the Second World War, the level of public debt in the United States had
increased to over 100 per cent of GDP, but it did not cause any major increase
in interest rates. Several studies on public finances in the United States
found no significant relationship between debt-to-GDP ratios and inflation or
interest rates over the period 1946-2008 (Aizenman and Marion 2009; Missale and
Blanchard 1994). As long as there is spare capacity in the economy or
unemployment, higher fiscal deficits add to purchasing power and do not exert
much upward pressure on interest rates or inflation, nor do they cause large
current account deficits.
Following
De Grauwe (2011), one can also argue that the adverse impact on real interest
rates depends on whether the debt is denominated in domestic or foreign
currencies. This is evident in the contrasting experiences of Spain and the UK
during the current episode of rapid public debt build-up. The UK public debt as
a percent of GDP was 17 percentage points higher than the Spanish Government
debt (89 per cent versus 72 per cent) in 2011. Yet, since the beginning of 2010
the yield on Spanish government bonds has increased strongly relative to the
UK, suggesting that international bond markets price in a significantly higher
default risk on Spanish than on UK government bonds. This difference rose to
200 basis points in early 201 1. One of the reasons why the financial markets
have singled out Spain and not the UK for the possibility of getting entangled
in a government debt crisis is that they know Spain, as member of a monetary
union, does not have control over the currency in which its debts are issued,
while the UK public debt is mostly in its own currency.
A study
prepared for the United Nations, World Economic Situation and Prospects 2011
(UN 2011) traced the flow cost of servicing the public debt in developed
countries in the present-day context. It found that so far the cost of public
debt in the United States and the major economies of the Euro area has remained
very low. Interestingly, for most countries, the flow cost of servicing the
debt is below 2 per cent of GDP, except for Greece, Italy and Finland. For most
of the developed countries, including the United States, it finds that the
projected expected public debt burden is zero or negative. Most countries with
low projected debt ratios have lower uncertainty in future debt burdens, and
this uncertainty does not increase monotonically with the size of the projected
debt. Thus, WESP (UN 201 1: 26) notes, ‘From this perspective, one could
conclude that, insofar as future growth depends on short-term stabilization
during or in the aftermath of a financial crash and a deep recession, the
additional debt incurred for such stabilization may not translate into
excessively high medium-term flow costs of public debt for an important part of
the developed countries’. Therefore, the risk of triggering vicious public debt
dynamics depends critically on the growth scenarios. On the other hand,
premature fiscal consolidation may slow or delay economic recovery, and could
well trigger such a vicious circle.
Fallacy
of ‘Expansionary Fiscal Contractions’9
In the
preceding sections, we have highlighted the weak and doubtful empirical bases
for a public debt-growth link based on cross-country data. Here, we examine
more closely the claim that it is possible to construct expansionary fiscal
contractions’ and even growth-based’ fiscal consolidation. We consider both
cross-country studies and country-specific experiences. Based on this
evaluation, we are reminded of Domar’s (1944, 1993) key message that fiscal
sustainability crucially hinges on economic growth. In a situation of
uncertainty about the global economy and when the private sector is repairing
its balance-sheet after a long period of profligacy, fiscal consolidation will
most likely lead to an outcome similar to the ‘paradox of thrift’, whereby
people end up saving less as their income declines due to cuts in spending by
all in a bid to save more.
Fiscal Consolidation:
Theory and Cross-Country Experience
Theory
The
proponents of expansionary fiscal contractions’ argue that even the supposedly
short-run damage of fiscal austerity would be limited or not arise at all. If
consolidations are credible, decisive, and of the right kind, then recovery
should follow rapidly. This view rests, as noted above, on the ‘market
confidence’ argument. The best articulation of this idea can be found in the
‘Stability and Growth Pact’ of the European Union:
Upholding
trust in the soundness of public finances enhances confidence among all
economic agents and thereby contributes to sustainable growth in consumption
and investment. Stability and growth are thus not conflicting objectives, but
rather reinforce each other – a fact which is very well captured in the title
of the fiscal framework called the ‘Stability and Growth Pact’. (ECB 2003: 6)
The same
message has been reflected in the recent G20 communiqué. It accepts the
stabilising role of fiscal deficits, but only in exceptional situations. In the
contemporary debate on fiscal consolidation, some commentators have suggested a
‘forward-looking’ interpretation of ‘market confidence’. This implies that
governments have to be proactive and anticipate how markets might react in the
future by adopting a ‘big bang’ approach to fiscal consolidation. Thus: ‘Given
that the current levels of debt are high by historical standards and that they
are very high in many advanced economies, it might be that markets will soon
ask for a strong signal of commitment and, in its absence, risk premia on
government bonds will increase. To avoid an increasing cost of rolling over the
debt, governments could be better off with a strong early adjustment’ (emphasis
added) (Fatás 2010). Therefore, consolidation should take priority over
stabilisation, and discretionary fiscal stimulus measures should be switched
off as soon as possible to avoid any damage to credibility’. This is supposed
to inspire the confidence’ of bond investors to offset any contractionary
impact of public expenditure cuts or increased taxes. This is especially
important for countries facing acute debt problems, with very high debt ratios
together with the prospect of soaring debt service burdens, threatening
crowding out and adverse confidence effects.
The key
link here is between debt costs and bond market confidence. It is an additional
argument that reinforces the neo-Ricardian equivalence idea or hopes for
positive ‘supply-side effects’ from shrinking public spending. Therefore,
fiscal consolidation could be expansionary since cuts in government spending
should strengthen market confidence, and hence lower borrowing costs due to
reduced perceptions of country risk and spur private investment. In the most
favourable case of fiscal consolidation, non-Keynesian effects (from greater
credibility and investor confidence) exceed contractionary Keynesian effects of
reduced public spending, resulting in higher growth.
Fiscal
discipline is also seen as a safeguard protecting monetary policy from
political pressures. Complementing central bankers’ ‘independence’, a prudent
fiscal framework is expected to help maintain price stability. In sum,
deliberate reversal of fiscal trends, brought about by means of redesigned
macroeconomic policies and institutions, is believed to have a positive impact
on business expectations and investment to deliver economic growth and
employment. Thus, important links are believed to exist between fiscal
consolidation, fiscal and monetary institutions, and economic growth and
employment.
Empirical
Evidence
A number
of cross-country studies have sought to demonstrate using historical data that
fiscal consolidation exercises are accompanied by growth and declines in
unemployment. Two cases from this genre can be highlighted. First, an IMF study
(Dermott and Wescott: 1996) focused on 74 cases of fiscal consolidation in 20
industrialized countries over the 1970-1995 periods. It concludes that 14 cases
were ‘successful’ in the sense that they were marked by sustainable reduction
(by about three percentage points over a period of three years) in the
debt-to-GDP ratio as well as an increase in growth and employment creation.
Second, a study by Alesina and Ardagna (2010) assembled 107 episodes of fiscal
consolidation in all OECD countries for the 1970 to 2007 period (see also
Alesina 2010). It concludes that 27 could be classified as cases that combined
fiscal consolidation with growth. Such results are underwhelming. The
historical experience thus suggests that the probability of a successful fiscal
consolidation is between 19 per cent (or 14 out of 74 as in the IMF study) and
25 per cent (or 27 out of 107 as in the Alesina- Ardagna study).
However,
Alesina and Ardagnas study as well as a similar earlier study by Alesina et al.
(1995) have been criticized by the IMF (2010a) on methodological grounds.
Specifically, these studies often identify periods of fiscal consolidation
using a statistical concept – the increase in the cyclically adjusted budget
surplus – that is a highly imperfect measure of actual policy actions. ‘This
way of selecting cases of consolidation biases the analysis toward downplaying
contractionary effects and overstating expansionary ones’ (IMF 2010a: 94).
Thus, in its World Economic Outlook the IMF (2010a) uses an alternative method
for identifying periods of fiscal consolidation by focusing on policy actions
intended to reduce the budget deficit. It finds that fiscal consolidation
typically has a contractionary effect on output. A fiscal consolidation equal
to 1 per cent of GDP typically reduces GDP by about 0.5 per cent within two
years and raises the unemployment rate by about 0.3 percentage point. Domestic
demand – consumption and investment – falls by about 1 per cent.10
Even if
one accepts that fiscal consolidation exercises have a reasonable chance of
being accompanied by growth and employment creation, one should not attribute
such an outcome to budgetary austerity alone. There are often complementary
factors at work that might be more important than fiscal actions. They include:
(1) the influence of the global business cycle, (2) monetary policy, (3)
exchange rate policy and (4) structural reforms. Dermott and Wescott (1996: 10)
found that ‘ . . . strong global economic growth helps to achieve a successful
consolidation, and weak global growth reduces the chances that consolidation
will cut the debt-to-GDP ratio’. It is also well-known that fiscal
retrenchments can be combined with loose monetary policy to offset recessionary
consequences. One European Commission study (Posen 2005) finds that, in more
than 50 per cent of the cases examined, fiscal austerity programmes were
accompanied by expansionary monetary policy that enabled growth to be
sustained. Similarly, the idea of combining fiscal retrenchments with
devaluation that boosts net exports to offset the decline in aggregate demand
(so-called expenditure reducing policies’ combined with expenditure switching
policies’) is well known. Furthermore, the expansion that accompanies fiscal
consolidation might well arise from structural reforms that alleviate binding
constraints on growth.
The
importance of these enabling factors needs particular attention in the light of
current circumstances. To start with, the post-crisis economic recovery is
still fragile when judged from an employment and labour market perspective.
Consider, for example, private sector assessments of anticipated employment
trends gauged by The Manpower Employment Outlook for the third quarter of 2012
– a report that has been around for 50 years. It interviewed over 65,000
employers across 41 countries and territories to measure anticipated employment
trends between July and September 2012 (Manpower 2012). Overall, it reports an
overall expected some slowing of hiring activity from the already flat pattern
for the same period in 2011, despite areas of positive hiring activity in 33
countries. Hence, the regional/global business cycle is not conducive enough to
the effective working of fiscal consolidation. The Eurozone economies also do
not have scope for devaluations nor do they have much room to cut interest
rates further through expansionary monetary policy as policy rates are still at
historically low thresholds.
In sum,
the results of historical studies of fiscal consolidation exercises suggest a
relatively high failure rate. Even in the successful cases, there were enabling
factors at play that might have offset the recessionary consequences of fiscal
retrenchments. Furthermore, the usual arguments that are invoked to justify fiscal
consolidation (Ricardian equivalence, crowding out and market confidence) lack
robust empirical substantiation. One study (Broyer and Brunner 2010) has
offered some estimates of the net impact of fiscal consolidation on growth in
eight European economies (Germany, France, United Kingdom, Italy, Spain,
Netherlands, Portugal, and Greece). It suggests that, even by 2016, all
countries bar one will suffer an output contraction as a result of the
transition from fiscal stimulus packages to consolidation.
Fiscal
Consolidation: Some Country-Specific Experiences
The US
case under the Clinton Administration
Fiscal
consolidation during the Clinton era might be considered as one of the best
recent examples of fiscal consolidation that was accompanied by significant and
sustained growth. The strong dollar policy and lower interest rates were the
prime movers behind the Clinton era expansion, which in turn increased revenue
and reduced social security expenditure, and thereby achieved a fiscal
surplus.11 Expansionary monetary policy, in particular, triggered and sustained
growth. As we will argue, fiscal consolidation was a consequence, rather than a
cause, of such growth.
Then
Treasury Under-Secretary Summers argued against a weak dollar policy to promote
growth through exports for at least two reasons. First, there is a time-lag
between exchange rate adjustments and the response of exports to such
adjustments – the so-called I-curve effect. Second, in a world of increased
capital mobility, any perception of a fall in dollar has to be matched by a
corresponding rise in interest rates (due to the arbitrage condition known as
interest parity). The higher interest rates associated with a weaker dollar
thus stifle growth while export growth lags.
The new
Secretary of Treasury, Rubin, endorsed the strong dollar argument. In his
confirmation hearings before the Senate Finance committee, he stated that a
strong dollar was in the best interest of the U.S. economy and warned that the
exchange rate should not be an instrument of U.S. trade policy. At a prime-time
news conference on 19 April 1995, President Clinton stated that the U.S. ‘wants
a strong dollar’ and that it ‘has an interest over the long run in a strong
currency’ This change of exchange rate policy saw the rise of index value of
the US dollar against major currencies from less than 80 in January 1995 to
over 100 by January 2000 (DeLong and Eichengreen 2011).
A strong
US dollar also meant less imported inflation, allowing the Fed to maintain
expansionary monetary policy. The Fed refrained from terminating the boom of
the 1990s by raising interest rates even though unemployment fell markedly
below any previous conventional NAIRU measure. In addition to boosting growth,
low interest rates helped keep bond yields close to the nominal GDP growth, so
that the interest burden stayed under control, as reflected in stability of
primary balances. The US experience illustrates that it was expansionary
monetary policy, in particular, that ignited and sustained growth, allowing
fiscal consolidation. In large part, fiscal consolidation during the Clinton
era was a consequence of the economic boom itself.
The cases
of Denmark and Ireland12
Denmark’s
(1983-87) and Ireland’s (1987-89) successful consolidations are widely seen as
demonstrating the dominance of non-Keynesian effects and the possibility of
expansionary fiscal contractions’ However, Denmark had to live through a long
period of sluggish growth from 1987 until 1993, following its supposedly
expansionary fiscal contraction of 1983-87. Domestic demand shrank in six of
seven years, turning a current account deficit of 5.3 per cent of GDP in 1986
to a surplus of 2.8 per cent in 1993. Since 1994, domestic demand recovered as
the real interest rate began declining and converged to the EU25 area rate. The
Danish central bank not only followed all of the interest rate changes made by
the ECB, it also implemented several interest rate cuts independently, thus
supporting GDP growth. This easing of monetary policy was possible due to the
stability of the exchange rate, which remained within the fluctuation bands
defined in ERM II. Denmark embarked on fiscal consolidation in 1996 and, in
2003, only after growth had returned.
Ireland’s
fiscal tightening of 1987-89 was followed by a sharp fall in GDP growth; the
Irish resurgence took off in earnest in 1994 (after a second dip recession in
1993). The move to wider ERM exchange rate bands in August 1993 and the
commitment to EMU were successful in removing the pressure on exchange rates
and its implications for interest rates. The more stable exchange rate
environment allowed interest rates to fall and converge to the German levels,
which helped boost domestic demand and investment (Murphy 2000). It seems that
the contraction of the Irish economy by over 7 per cent in 2009, following its
severe austerity measures, is a replay of its earlier experience. Instead of
being rewarded for its actions, taken long before the Greek drama unfolded,
investors seem to have punished Ireland. The factors that contributed to
Ireland’s exceptional growth performance in the 1990s included interest rate
convergence, massive EU transfers and foreign direct investment, not so-called
investor confidence.
Thus,
like the US story of the Clinton era, the Danish-Irish growth since the
mid-1990s was the result of favourable exchange rate and interest policies,
which, in turn, helped fiscal consolidation. A close scrutiny of the
Danish-Irish cases would also show that both economies were helped by a
favourable world economic environment through buoyant export demand and direct
foreign investment.13 This fundamental point is often missed in conventional
analysis
What are
the policy lessons from these cases? This can be answered by using the familiar
identity among budget deficits (BD), saving (S), investment (I) and current
account deficits (CD): BD = S – I + CD. Thus, there are two policy options:
either adjust the right hand or the left hand variables to achieve the desired
goals. That is:
i)
Discretionary cuts in BD which lead to adjustments in S, I & CD
ii)
Policy adjustments to influence S, I & CD
It seems
all three, the US, Denmark and Ireland, chose the second option. The most
pertinent policy lesson is, there are alternatives and that adjustments of
relevant policy instruments to achieve desired goals seem to yield better
results than discretionary and forced adjustment. Although the discretionary
option may seem attractive as having to deal with only one variable instead of
three and hence can give an impression of more certainty, actual outcome
appears to be different.
Concluding
Remarks
There is
no doubt that many industrialised countries are facing serious fiscal crises
that need to be tackled soon. However, as a recent BIS paper noted, the key
challenge for fiscal authorities is ‘how to do that without seriously
jeopardising the incipient economic recovery’ (Cecchetti, Mohanty and Zampolli
2010: 2). Unfortunately, fiscal authorities in many industrialised countries,
Europe in particular, seemed to have failed. They have embarked upon drastic
fiscal consolidation measures without regard for the underlying causes of
fiscal problems, i.e. how much of this is due to cyclical factors, such as the
ongoing recession and how much is due to structural factors, such as ageing
population and unfunded pension schemes, non-progressivity of the tax system or
bloating of the public sector.
Many
countries had huge public debts when World War II ended. Despite calls then for
drastic expenditure cuts, governments spent a great deal more on economic
reconstruction and social protection measures. Had they caved in to the fiscal
hawks of their time, post-war European recovery would have been delayed and the
Cold War could have been lost. As governments continued with massive
expenditure to rebuild their countries, economies grew all over the world, and
debt burdens diminished quickly with rapid economic growth and fast growing tax
revenues. These experiences show that deficits and surpluses should be adjusted
counter-cyclically over the course of business cycles (Matthews 1968).
There is,
of course, one big difference between then and now. The financial sector is
much more powerful now, with governments often held hostage by financial
markets and the whims of rating agencies. The record of rating agencies before
the 2008 global economic crisis is now widely acknowledged as abysmal, with
even the US Congress seriously debating whether or not they should be
prosecuted.
The claim
that high public debt causes lower growth and that it is possible to have
expansionary fiscal contractions’ cannot be supported by robust crosscountry
evidence as well as by country-specific experiences. Such a claim also ignores
the impacts of other variables, especially of those which may be affected by
public debt itself. For example, high public debt could be used to improve
schooling which is found to have a larger positive impact on growth than the
estimated negative impact of public debt-GDP ratios. Public debt to enhance
government capacity and capabilities (measured by size of government) or to
improve infrastructure can also positively contribute to growth and outweigh
the potential adverse impact of high initial debt-GDP ratios. In other words,
as Domar pointed out, both the size and composition of debt matter. The
growthinhibiting effects of a given percentage increase in debt-to-GDP ratio
can be easily overwhelmed by a given percentage increase in the
growth-promoting variables achieved through public spending.
The
issue, however, is different when it comes to the accumulation of external
liabilities. The question is then not only of being able to repay, but also
whether other countries would be willing to continue to lend. Paradoxically, in
crisis-hit countries with access to private capital markets, fiscal prudence
does not offer any safeguard against the pitfalls and perils of private
sector-led accumulation of external liabilities because they eventually become
the liabilities of the government. This is a lesson that Ireland and other
debt-ridden economies of the Eurozone have painfully discovered today in the
wake of the global recession of 2008-2009 and as Indonesia and Thailand
discovered during the 1997 Asian financial crisis.
The
current preoccupation with public debt and fiscal consolidation has had the
consequence of distracting attention from the crucial role that fiscal policy
plays in promoting growth and development. This point is made forcefully in an
insightful ‘interim report’ that informed the deliberations of the Development
Committee of the IMF and World Bank in April 2006. The authors of the report
note that debts and deficits are useful indicators for:
. . .
controlling the growth of government liabilities, but (they) offer little
indication of longer term effects on government assets or on economic growth.
Conceptually, the long-term impact is better captured by examining the impact
of fiscal policy on government net worth. (Development Committee 2006: i)
The
report argues that ‘ . . . there is clearly a need for fiscal policy to
incorporate, as best as possible, the likely impact of the level and composition
of expenditure and taxation on long-term growth . . . ‘ (ibid.).14 It is time
to resurrect these important ideas as an antidote to the alarmist discourse on
public debt.
There is
also the more fundamental issue of whether the policy-making process should
become hostage to the confidence game’ in which evidence-based policy-making is
replaced by a band of amateur psychologists seeking to read the collective mood
of financial markets. When this happens, fundamental macroeconomic policy
errors are likely to be committed, as the mishandling of the 1997 Asian
financial crisis by international financial institutions has shown.
Acknowledgements
An
earlier version of this paper was presented at a seminar in the Reserve Bank of
Australia, organised by the Economic Society of Australia. The authors are
grateful to Sarah Anwar, Employment Policy Department, ILO, for tapping into
the IMF historical debt database which enabled new evidence to be harnessed to
assess the public debt-growth relationship. They are also grateful to
Professors Geoff Harcourt, John Nevile and Peter Kriesler of the University of
New South Wales and Professor Raja Junankar of the University of Western Sydney
for their insightful comments. We are particularly grateful to two anonymous
referees for their helpful and incisive comments that guided this revised
version. However, the usual disclaimer applies: the authors are solely
responsible for any remaining errors and omissions.
The views
expressed herein do not anyway reflect the views of the UN-ESCAP or ILO or any
of the UN agencies.
Footnote
Notes
1.
‘Fiscal consolidation is a policy intended to reduce deficits and the
accumulation of debt. The term typically refers to a government economic
policy. The terms ‘fiscal consolidation and ‘fiscal austerity’ will be used
interchangeably in this paper.
2. See
Federal Ministry of Finance, Germany: 2012. For other OECD countries, see OECD
2010, Chapter 4. Fiscal consolidation requirements in the OECD countries are
summarized as: ‘Consolidation requirements are substantial; merely to stabilise
debt-GDP ratios by no later than 2025 requires strengthening the underlying
primary balance from the current position by more than 5 per cent of GDP in the
OECD area on average. Tightening by more than 8 per cent of GDP is called for
in the United States and Japan, with the United Kingdom, Portugal, Slovak
Republic, Poland and Ireland all requiring consolidation of 5 to 7 percentage
points of GDP. Consolidation requirements would be much more demanding if the
aim were to return debt-to-GDP ratios to their pre-crisis levels. In addition,
for a typical OECD country, offsets of 3 per cent of GDP will have to be found
over the coming 15 years to meet spending pressures due to ageing, representing
additional cumulative consolidation requirements of about lA per cent of GDP
per year (OECD 2010: 12).
3. The
detailed findings and methodologies employed are reported in Kumar and Woo
(2010).
4. This
does not, of course, imply causality between debt and growth, but the nature of
the association shown here is insightful.
5. Of
course, one should recognise the limits of the IS-LM framework, most notably
the assumptions pertaining to the exogeneity of money and the ability of
governments to control monetary aggregates. This affects the use of the LM
curve as a valid analytical device. See, for example, Romer (2000). The
advocates of the DSCGE or dynamic computable general equilibrium models would
also debunk the use of the IS-LM framework in probing the crowding out’ thesis
on the ground that this framework is not rigorously microfounded’. Yet, DSGCEs
also suffer from various limitations. A detailed discussion of this issue is
beyond the scope of the paper.
6. Chick
and Pettifor (2011), using historical UK data for 1909-2009, show that there
was a very strong negative association between government expenditure and the
government debt, excluding the two outliers for the World Wars. When public
expenditure increased, public debt fell, as the economy expanded and revenue
rose. On the other hand, contrary to the conventional wisdom, fiscal
consolidations did not improve public finances.
7. See
Hart (201 1) for exposition of this.
8. Based
on historical Australian data, Nevile and Kriesler (2011) stress the importance
of the composition of government expenditure.
9. This
Section draws on the authors’ VoxEU commentaries (Chowdhury and Islam 2010a,
2010b; Islam and Chowdhury 2010), and G24 Policy Briefs Nos. 57 & 58 (Islam
and Chowdhury 2010b, 2010c).
10.
Jonathan Portes, Director of the National Institute of Economic and Social
Research, which is Britain’s oldest independent research institute, has pointed
out that the Alesina-Ardagna study has been so discredited that the UK
Treasury, which was briefly influenced by that study and used its findings to
justify the Emergency Budget of 2010, has now retreated from that position. See
Portes (2011). John Quiggin, a leading Australian economist, chastises the
Alesina-Ardagna (AA) paper for making elementary factual errors in the case
study of Australia, advancing it as a clear case of expansionary fiscal
contraction (Quiggin 2012). Others who have been critical of the AA study
include The Economist which considers the study ‘seriously flawed’ (The
Economist 2010b).
11.
President Clinton was also able to fend off pressure to increase spending when
a surplus did emerge in 1998 with the successful slogan, ‘Save Social Security
First.’ The policy was at least ‘let’s save the surpluses until we put social
security on a firm footing’ – no tax cuts, no big spending increases. For the
political economy explanation, see Berglund and Vernengo (2004).
12.
Giavazzi and Pagano (1990) first raised the prospect of expansionary fiscal
consolidation based on their analysis of Denmark and Ireland.
13.
Perroti (201 1) reports similar observations after detailed case studies of
fiscal consolidations in Ireland, Denmark, Finland and Sweden.
14. This
report was attached to the 23 April 2006 World Bank’s Development Committee
meeting.