WPS7724
Policy Research Working Paper 7724
Do Fiscal Multipliers Depend
on Fiscal Positions?
Raju Huidrom
M. Ayhan Kose
Jamus J. Lim
Franziska L. Ohnsorge
Development Economics
Development Prospects Group
June 2016
Policy Research Working Paper 7724
Abstract
This paper analyzes the relationship between fiscal multipli- fiscal position is strong, while it can be negative when the
ers and fiscal positions of governments using an Interactive fiscal position is weak. Second, these effects are separate
Panel Vector Auto Regression model and a large data-set and distinct from the impact of the business cycle on
of advanced and developing economies. The methodol- the fiscal multiplier. Third, the state-dependent effects
ogy permits tracing the endogenous relationship between of the fiscal position on multipliers is attributable to two
fiscal multipliers and fiscal positions while maintaining factors: an interest rate channel through which higher
enough degrees of freedom to draw sharp inferences. The borrowing costs, due to investors’ increased perception of
paper reports three major results. First, the fiscal multipli- credit risks when stimulus is implemented from a weak
ers depend on fiscal positions: the multipliers tend to be initial fiscal position, crowd out private investment; and
larger when fiscal positions are strong (i.e. when govern- a Ricardian channel through which households reduce
ment debt and deficits are low) than weak. For instance, consumption in anticipation of future fiscal adjustments.
the long-run multiplier can be as large as unity when the
This paper is a product of the Development Prospects Group, Development Economics Vice Presidency. It is part of a larger
effort by the World Bank to provide open access to its research and make a contribution to development policy discussions
around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors
may be contacted at rhuidrom@worldbank.org, akose@worldbank.org, jlim@worldbank.org, fohnsorge@worldbank.org.
The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development
issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the
names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those
of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and
its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.
Produced by the Research Support Team
Do Fiscal Multipliers Depend on Fiscal Positions?
Raju Huidrom, M. Ayhan Kose, Jamus J. Lim, and Franziska L. Ohnsorge
Key Words: Fiscal multipliers, …scal position, state-dependency, Ricardian channel, interest
rate channel, business cycle
JEL Codes: E62, H50, H60
Huidrom: World Bank, Development Prospects Group; rhuidrom@worldbank.org. Kose: World Bank, Devel-
opment Prospects Group; Brookings Institution; CAMA; CEPR; akose@worldbank.org. Lim: World Bank, De-
velopment Prospects Group; jlim@worldbank.org; Ohnsorge: World Bank, Development Prospects Group; fohn-
sorge@worldbank.org. Earlier drafts bene…ted from comments by Syud Amer Ahmed, Jean-Louis Arcand, Raphael
Espinoza, Alejandro Izquierdo, Aart Kraay, Sergio Kurlat, Jaime Marquez, Rahul Mukherjee, Ugo Panizza, Assaf
Razin, Luis Serven, Cedric Tille, Carlos Vegh, Guillermo Vuletin, Sebastian Weber, Kei-Mu Yi, Hakan Yilmazkuday,
Charles Wyplosz, and seminar participants at the Graduate Institute-Geneva, the 2015 Midwest Macro Conference,
and the 2016 Conference on Fiscal Policy and the Macroeconomy at Johns Hopkins SAIS. We thank Kiwako Sakamoto
for excellent research assistance. Ethan Ilzetski kindly shared data.
1 Introduction
During the Great Recession of 2008-09, many countries around the world - both advanced and
developing - deployed …scal policy to support activity. As a result, government debt and de…cits
increased in many countries, and they remain elevated (Huidrom, Kose, and Ohnsorge 2016).
Against this backdrop of weak …scal positions, there has been a revival of interest in …scal policy as
a macroeconomic stabilization tool. Yet, there is scant evidence regarding the extent to which …scal
policy is e¤ective in stimulating the economy during times of weak …scal position. The objective of
this paper is to …ll this gap in the literature. In particular, we ask: do …scal multipliers depend on
…scal positions?
The question we study follows the …nding in recent literature that an “average” …scal multi-
plier which is assumed to apply universally is irrelevant and that multipliers depend on speci…c
macroeconomic conditions. For instance, beginning with the work of Auerbach and Gorodnichenko
(2012b), recent papers have established that multipliers tend to be larger during recessions than
during expansions (Bachmann and Sims 2012; Candelon and Lieb 2013; Owyang, Ramey, and
Zubairy 2013). The notion that …scal multipliers depend on the state of the business cycle is well
grounded in theory. During recessions, the multiplier e¤ect from government spending can rise
due to slack in labor markets, larger frictions in …nancial markets, and an increase in liquidity
constrained agents.1 The literature has, thus far, o¤ered a convincing case that the phase of the
business cycle should be regarded as a key conditioning state that may in‡ uence the e¢ cacy of
…scal policy.
Economic theory, however, does not limit the conditioning state to the phase of the business
cycle alone. In fact, theory suggests that …scal position of the government, as distinct from the
business cycle, can be another important determining factor for the size of …scal multipliers. This
state-dependency of multipliers on …scal position can operate via two channels. First, a Ricardian
channel: when a government with a weak …scal position implements a …scal stimulus, households
expect tax increases sooner than in an economy with strong …scal position (Sutherland 1997; Perroti
1999). The perceived negative wealth e¤ect encourages households to cut consumption and save,
thereby weakening the impact of the policy on output. Thus, the net e¤ects of …scal policy on
output, the size of the …scal multiplier, may be negligible or even negative. Second, an interest
rate channel: when the …scal position is weak, …scal stimulus can increase lenders’ perceptions
of sovereign credit risk. This raises sovereign bond yields and hence, borrowing costs across the
whole economy. This, in turn, crowds out private investment and consumption, reducing the size
of the multiplier. Therefore, both channels suggest that …scal policy is less e¤ective when the …scal
stimulus is implemented from a weak initial …scal position.2
1
These e¤ects may be further ampli…ed in the special case where monetary policy is also constrained by the
zero lower bound (Christiano, Eichenbaum, and Rebelo 2011; Denes, Eggertsson, and Gilbukh 2013; Erceg and
Linde 2014). In advanced economies, …scal policy has received much attention given the crisis-induced zero lower
bound environment that has constrained conventional monetary policy (Blanchard et al. 2010 and 2013; Delong and
Summers 2012).
2
Sutherland (1997) formalizes the Ricardian channel by postulating that there exists a debt threshold at which the
government makes …scal adjustments, via increasing taxes, to remain solvent. Thus, households expect higher taxes
to be more eminent when the government conducts an expansionary …scal policy from a high initial level of debt. In
Perotti (1999), such expectations of higher taxes can also result in increased tax distortions which are an additional
source of negative wealth e¤ects. With regard to the interest rate channel, Bi, Shen, and Yang (2014) theoretically
establish that sovereign risk premia can increase nonlinearly as government indebtedness rises. Corsetti et al. (2013)
highlight the interest rate is particularly relevant when monetary policy is constrained, for instance during a zero
lower bound episode.
2
To estimate …scal multipliers that depend on the …scal position, we use an Interacted Panel
Vector Autoregressive (IPVAR) model.3 The model is essentially an extension of an otherwise
standard panel structural VAR (SVAR), with the distinction that the VAR coe¢ cients interact
with (observable) state variables. Consequently, these coe¢ cients become time-varying, and evolve
endogenously according to these states. This results in a framework where the VAR dynamics and
hence, the …scal multipliers are conditional on the state variables which we take to be the …scal
position.
More importantly, since the state-dependency is captured by making use of the full sample,
this nonlinear approach allows us to maintain enough degrees of freedom, thus allowing us to draw
sharper inferences. This feature of the model is particularly useful when conditioning on multiple
states of interest: a feature we exploit when we jointly condition on the …scal position and the
phase of the business cycle. The latter exercise allows us to evaluate whether the …scal position
is a unique state, di¤erent from the phase of the business cycle, which determines the size of the
…scal multipliers.
Applying our empirical methodology to a large dataset that covers 34 countries (19 advanced
and 15 developing), at the quarterly frequency over the period 1980:1 – 2014:1, we empirically
establish that the …scal position is a key conditioning state that determines the size of the …scal
multipliers. In particular, estimated multipliers are systematically smaller when the …scal position
is weak (i.e. government debt is low), and vice versa when it is strong. In addition, we show that
the state-dependency of multipliers on the …scal position is independent of business cycle e¤ects.
That is, while we …nd multipliers to be larger during recessions than expansions (consistent with
Auerbach and Gorodnichenko 2012b), the weaker (stronger) multiplier e¤ect that derives from
a weak (strong) …scal position applies even when the economy is experiencing a recession or an
expansion.
Furthermore, we provide empirical evidence that such state-dependent e¤ects operate through
the two channels highlighted above. When the government conducts expansionary …scal policy
during times of high debt, the private sector scales back on consumption in credible anticipation
of future tax pressures due to the weak state of public …nances (Ricardian channel) and private
investment is suppressed plausibly due to an increase in economy-wide interest rate as perceptions
of heightened sovereign risk become stronger (interest rate channel).
Some recent empirical studies have documented the importance of …scal positions for …scal
multipliers. For instance, Ilzetzki, Mendoza, and Vegh (2013) include measures of …scal fragility in
their analyses of multipliers. However, …scal considerations are not the centerpiece of their analysis,
and so they apply only certain debt thresholds, as opposed to our more general stance that allows
these thresholds to emerge naturally from the data. Using a similar IPVAR approach like ours,
Nickel and Tudyka (2014) provide estimates of multipliers that depend on the …scal position for high-
income European economies. However, they do not distinguish between the state of the business
cycle and …scal position. There is, therefore, an indeterminacy over whether the state-dependency
of the multipliers is uniquely attributable to the latter. Using a di¤erent econometric methodology
than ours, Auerbach and Gorodnichenko (2012a) discuss the joint conditioning exercise and …nd
that large government debt reduces the stimulative e¤ects of expansionary …scal policy even during
3
The model has been used in various areas of empirical macroeconomics: exchange rates (Towbin and Weber
2013); capital ‡ows (Sa, Towbin, and Wieladek 2014); and …scal policy (Nickel and Tudyka 2014).
3
recessions. But their identi…cation strategy requires data on government consumption forecast
errors, which essentially limits their study to only OECD countries.
Our paper makes three contributions. First, by clearly distinguishing between the state of the
business cycle and the …scal position, we establish that the …scal position is a unique state that
determines the size of the …scal multiplier. Second, we show the empirical relevance of the trans-
mission mechanisms that underlay the state-dependent e¤ects due to …scal position: the Ricardian
channel and the interest rate channel. Third, compared to previous studies, our sample includes a
larger set of countries covering advanced and developing economies, thus providing a general result
on the state-dependent e¤ects due to …scal position.
The rest of the paper is organized as follows. Section 2 presents the econometric methodol-
ogy. Here, we discuss the IPVAR model, the identi…cation strategy, and the database. We present
estimates of state-dependent multipliers in Section 3. In Section 4, we discuss the transmission
mechanisms that highlight the Ricardian and the interest rate channels. Section 5 discusses robust-
ness exercises and Section 6 concludes.
2 Empirical Methodology
2.1 Econometric Model
A standard panel structural VAR (SVAR) estimates a single set of parameters which then yields an
“average” or unconditional multiplier. Our objective is to go beyond the unconditional multiplier,
and investigate how multipliers can depend on speci…c macroeconomic conditions, in particular …scal
position of governments. For that, we deploy the Interacted Panel Vector Autoregressive (IPVAR)
model where the main innovation, with respect to a standard panel SVAR, is that the model
coe¢ cients vary deterministically according to conditioning (state) variables. Thus, the IPVAR
results in a framework where model dynamics and hence, estimated multipliers are conditional on
the state variables. By choosing the conditioning variable to be a measure of …scal position in the
IPVAR, we estimate multipliers that depend on …scal position.
The IPVAR model, in its structural form, is represented by:
2 32 3 2 323
11 12 13 14
1 0 0 0 gcit l;it l;it l;it gcit l
l;it
6 21 1 0 0 7 6 gdpit 7 XL 6 6
21 22 2376
7 6 gdpit l 7
24
6 0 ;it 76 7 = 6 l;it l;it l;it
74 l;it 7+Xit F +Uit ;
4 310;it
32
0;it 1 0 5 4 cait 5 l=1 4 31 32 33
5 cait l 5
34
l;it l;it l;it l;it
41 42 43 1 reerit 41 42 43 44
reerit l
0;it 0;it 0;it l;it l;it l;it l;it
(1)
where for a given country i in period t, gc represents real government consumption, gdp real
gross domestic product (GDP), reer the real e¤ective exchange rate, and ca current account balance
(as a share of GDP).
We take government consumption as the …scal instrument and we track the e¤ects of …scal policy
in terms of GDP. Separately, we check the robustness of our results by tracking …scal outcomes
in terms of private consumption and private investment (Section 4). Real e¤ective exchange rate
and the current account are included in the model to account for open economy features that
characterize most of the countries in our sample. The matrix X captures additional controls, which
include the time-invariant country …xed e¤ects, and U is a vector of uncorrelated, i:i:d: (structural)
4
shocks. The shock corresponding to government consumption is the …scal shock. Following Ilzetzki,
Mendoza, and Vegh (2013), we set the lag length as L = 4.4
The impact matrix A0 (matrix of coe¢ cients on the left hand side of Equation (1)) is lower
triangular. This along with the ordering of the variables in the VAR is related to our identi…cation
scheme (discussed in detail in the next section). Both the impact matrix A0 and the coe¢ cient
matrices Al , l = 1; : : : ; L (on the right-hand side of Equation (1)) comprises time-varying model
coe¢ cients that, for any given entry in row j and column k , evolve deterministically according to:
jk jk jk
l;it = 1;l + 2;l f sit ; (2)
where f s refers to the …scal position.5 Our baseline measure of the …scal position is the gov-
ernment debt-to-GDP ratio. While the literature has used a variety of measures in this regard,
our choice is in line with theoretical macro models, where government debt is the modal state
variable.6 Since measures of …scal position are endogenous and move in tandem with the business
cycle, we take lagged moving averages of all our …scal measures to control for business cycle e¤ects.7
Equations (1) and (2) jointly denote the IPVAR system. When the law of motion in Equation (2)
is suppressed, the IPVAR reduces to a standard panel SVAR which we use to estimate the un-
conditional multipliers. The latter serve as a baseline against which we compare the conditional
multipliers from the IPVAR.
The matrices Al , l = 0; : : : ; L determine the e¤ects of structural shocks on the dynamics of
endogenous variables in the VAR system. By conditioning the law of motion of the coe¢ cients in
these matrices on the …scal position, as in Equation (2), we are allowing those e¤ects to depend
on the …scal position. This scheme allows us to calculate impulse responses and hence estimates of
…scal multipliers conditional on a given level of …scal position.8 When estimating the VAR system,
we make use of the full sample. This enables us to circumvent the degrees-of-freedom challenge
that limits the ability of existing empirical models to account for joint conditioning on multiple
states.
As standard in the literature, we compute the cumulative …scal multiplier at horizon T as the
discounted cumulative change in output until horizon T , as the discounted cumulative government
consumption increases by one unit. That is,
4
We use the same lag length of 4 when we report results for speci…c country groups as well. Ilzetzki, Mendoza,
and Vegh (2013) note that the optimal lag length in the VAR varies across country groups. Choosing the same lag
length (that equals 4) ensures that di¤erences in the multipliers are not attributable to the lag structure of the VAR.
5
Including …scal position in the law of motion in Equation (2) is tantamount to having interaction terms with
…scal position in the regressors of Equation (1). For this reason, we do not separately include …scal position as an
endogenous variable in the IPVAR.
6
For instance, while Riera-Crichton, Vegh, and Vuletin (2014) condition multipliers on …scal balances, Auerbach
and Gorodnichenko (2012a), Ilzetzki, Mendoza, and Vegh (2013), and Nickel and Tudyka (2014) condition on gov-
ernment debt. For robustness, we present results when …scal balances are the conditioning variable.
7
In particular, we take the 5-quarter moving average of the …scal position, and then lag it by 2 quarters. Given
the average length of the business cycle, this e¤ectively allows us to abstract from changes in the …scal state that
may potentially be contaminated by cyclical movements. We allay any residual endogeneity concerns by jointly
conditioning on the …scal position and the phase of the business cycle.
8
More precisely, the impulse response calculation assumes that the initial level of …scal position on which the
impulses are conditioned prevails throughout the impulse horizon. In practice, …scal position can also respond to the
…scal shock and its dynamics can have implications for …scal multipliers (see Ramey and Zubairy (2014) for a similar
point). Since …scal position is not an endogenous variable in our IPVAR model, calculating impulse responses while
taking into account the endogenous evolution of …scal position in not possible.
5
XT t
(1 + r) gdpt
t=0
M ultiplier (T ) = XT ; (3)
t
(1 + r) gct
t=0
where r denotes the interest rate. We utilize the median short-term rate in the sample for this
purpose which is about 7.4 percent.
From (3), the impact multiplier is obtained by setting T = 0 and the long-run multiplier by
setting T at an arbitrarily large number, which is taken to be T = 20 (5 years) in our exercise.
At T = 20, impulse responses in our model by and large revert to their unconditional means, and
so we take this to be representative of the long run. In addition, we speci…cally report multipliers
corresponding to one-year (T = 4) and 2-year (T = 8) horizons, when the …scal multipliers typically
peak. To calculate the …scal multiplier using the coe¢ cient estimates from the IPVAR, we …rst
cumulate the discounted impulses of output and government consumption at di¤erent horizons and
compute the ratio of the two impulses. That ratio is then multiplied by the average government
consumption to GDP ratio in the sample to yield multipliers.9
2.2 Identi…cation and Estimation
To identify …scal shocks, we rely on the standard recursive identi…cation scheme of Blanchard
and Perotti (2002). The key timing assumption in this scheme is that discretionary …scal policy
does not respond to macroeconomic conditions within the quarter.10 Such a timing assumption
can be motivated by implementation lags typically associated with discretionary …scal policy. In
the VAR model, this timing assumption is achieved by ordering government consumption …rst
in Equation (1), before GDP. The timing assumption for the remaining variables in the VAR
follows Ilzetzki, Mendoza, and Vegh (2013): the current account is ordered before the real e¤ective
exchange rates. This ordering implies that GDP does not respond to the current account within one
quarter, and that the current account does not respond within one quarter when the real e¤ective
exchange rate moves. The precise ordering of the latter two variables is, however, immaterial for
our main results. Of course, there are alternative identi…cation schemes used in the literature. For
instance, Romer and Romer (2010) use a narrative approach to identify exogenous …scal shocks for
the US. Auerbach and Gorodnichenko (2012a) proxy exogenous …scal shocks by forecast errors of
government consumption for OECD countries. Due to data limitations, neither of these approaches
is feasible for our sample that includes developing countries.11
The IPVAR system, comprising Equations (1) and (2), is estimated with ordinary least squares
(OLS) applied separately to each equation.12 The estimated system yields model coe¢ cients that
9
This step to calculate multipliers from impulse responses follows Ilzetzki, Mendoza, and Vegh (2013). Since
the conditional multipliers are estimated from the panel of countries, they re‡ ect an average estimate across those
countries included in the panel. Thus, we use the average government consumption to GDP ratio in the sample to
calculate the multipliers rather than country-speci…c government consumption to GDP ratios.
10
One caveat of the recursive identi…cation scheme is that the …scal shocks identi…ed using this scheme may be
predicted by private forecasts (Ramey, 2011). Ilzetzki, Mendoza, and Vegh (2013), who use a similar identi…cation
scheme and sample of countries like ours, provide evidence that this is unlikely the case.
11
There are alternative identi…cation schemes used in the literature. For instance, Romer and Romer (2010) use
a narrative approach to identify exogenous …scal shocks for the US. Auerbach and Gorodnichenko (2012a) proxy
exogenous …scal shocks by forecast errors of government consumption for OECD countries. Due to data limitations,
these approaches are not feasible for our sample that includes developing countries.
12
Because the error terms are uncorrelated across equations by construction, estimating the IPVAR equation by
equation does not result in loss of e¢ ciency. See Towbin and Weber (2013) for a discussion.
6
depend on the …scal position such that a given level of the …scal position maps out to a set of model
coe¢ cients. For presenting the results, we evaluate model coe¢ cients at speci…c values of the …scal
position which are taken to be the percentiles within the sample. Con…dence bands are calculated
by bootstrapping over 300 iterations. We report median estimates, along with the 16 - 84 percent
con…dence bands.
2.3 Database
Our main database comprises an unbalanced panel that covers 34 countries (19 advanced and
15 developing), at the quarterly frequency over the period 1980:1 – 2014:1.13 Real government
consumption and real GDP are based on the quarterly database in Ilzetzki, Mendoza, and Vegh
(2013) which are extended until 2014:1 by splicing from the OECD and Haver Analytics. Real
e¤ective exchange rates are from the narrow (wherever available) and broad indices of the BIS,
and current account from the IMF’ s WEO database. The short-term rate used for discounting the
multiplier is drawn mainly from the IMF’ s IFS database. For the robustness results, we augment
this database to include quarterly real private consumption and private investment series. These
are drawn from the OECD, Haver Analytics, and Eurostat. Additional details on the sources and
de…nitions of all of these variables are provided in Table A2 in the Appendix.
The government consumption and GDP series (as well as private consumption and private
investment) are converted into logarithmic form, and detrended using a linear quadratic trend as
in Ilzetzki, Mendoza, and Vegh (2013). The exchange rate is transformed into quarter-to-quarter
growth rates, and the current account series is seasonally-adjusted using the X11 routine. All these
series are detrended and demeaned on a country-by-country basis, which e¤ectively controls for
country …xed e¤ects in the regressions.
We also employ another database that is an unbalanced panel with the same cross sectional
and time series coverage as before but at the annual frequency. This includes the conditioning
variables that are not explicitly required for the identi…cation scheme to be valid in the VAR
model but are necessary to estimate the interaction terms. These are government debt and …scal
balances as percentage of GDP which are drawn from the IMF’ s WEO (October 2014) database; and
government consumption-to-GDP ratios which we obtain from the World Bank’ s WDI database.
3 Results
3.1 Unconditional Multipliers
To establish a benchmark, we …rst report estimates of the unconditional multiplier from a standard
panel SVAR. For that, we suppress the law of motion for the coe¢ cients in Equation (2). This
renders the coe¢ cient matrices Al in Equation (1) invariant across countries and time. Figure
1 presents the unconditional multipliers for the select horizons: on impact, 1 year, 2 years, and
long run (5 years). Barring only a few periods in the impulse horizon, the unconditional impulse
responses of output due to a positive …scal shock are either negative or insigni…cant.14 Indeed,
13
The list of countries is presented in Table A1 in the Appendix. Our developing-country coverage comprises
primarily emerging and frontier market economies that have some ability to tap into international …nancial markets,
which renders the …scal solvency risks that underpin our nonlinear crowding-out mechanisms relevant. We exclude
low-income countries not only because of data reliability issues, but also because they primarily rely on concessional
…nance for government expenditure, which would not re‡ ect the crowding-out mechanisms.
14
When we split our sample into advanced and developing economies, our estimates of the unconditional multiplier
are very similar to the ones reported in Ilzetzki, Mendoza, and Vegh (2013). See Figure A2 in the Appendix.
7
across all horizons considered, the uncertainty surrounding these estimates is su¢ ciently large such
that the multiplier is essentially statistically indistinguishable from zero. This echoes the often small
and the wide range in the estimates of the …scal multipliers as reported in previous studies (see
Batini and Weber (2014) for a survey). The unconditional impulse responses presented in Figure
2 corroborate the small and imprecise estimates of the e¤ects that …scal policy has on activity on
an average.
The main message we take away from above is that the unconditional multipliers can mask
important state-dependencies as suggested by theory. The estimates of unconditional multipliers
suggest that …scal policy, on average, has no stimulative e¤ects on the economy. However, as recent
empirical work shows, …scal policy can be stimulative during speci…c times, for instance during
recessions (Auerbach and Gorodnichenko, 2012b). Accordingly, we turn, in the following section,
to our conditional multiplier estimates.
3.2 Fiscal Position-Dependent Multipliers
Figure 3 presents the set of estimated …scal multipliers (on the vertical axis) that depend on
government debt (on the horizontal axis) - our baseline measure of …scal position.15 The four panels
correspond to the four horizons previously selected. The …gure shows that there is a systematic
link between the size of the multiplier and the …scal position: the median value of the multiplier
decreases monotonically in debt, for all horizons reported. That is, the estimated multipliers for
all the horizons are positive and signi…cant for low levels of debt, but turn negative or insigni…cant
when debt levels are high. For instance, the long run multiplier is close to unity when debt is
low (strong …scal position), but is negative for high levels of debt (weak …scal position).16 The
di¤erence in the estimated multipliers for low and high levels of debt is particularly signi…cant at
longer horizons. Our empirical results therefore lend support to the theoretical insights of earlier
studies which show that a weak …scal position can result in stronger crowding-out e¤ects, blunting
the stimulative e¤ects of …scal policy (Sutherland 1997; Perroti 1999; Corsetti et al. 2013; Bi, Shen,
and Yang 2014).17
Compared with the unconditional multipliers (Figure 1), the conditional multipliers paint a more
nuanced picture of the e¤ects of …scal policy. For instance, at the 1-year horizon, the unconditional
multiplier is small and insigni…cant. The estimated conditional multipliers at the same horizon
(Figure 3) highlight that much of those small and insigni…cant average e¤ects actually re‡ ect
episodes when …scal positions are weak. On the other hand, when the …scal position is strong, the
conditional …scal multipliers are not only larger than the unconditional estimates but they are also
statistically di¤erent from zero.
To better grasp the economics underlying these results, it is useful to examine the conditional
impulse responses associated with expansionary …scal policy. For the purpose of illustration, we
consider impulse responses conditional on two levels of debt: one corresponding to the 10th per-
centile in the sample (strong …scal position) and the other corresponds to the 90th percentile (weak
15
Figure A1 in the Appendix provides the distribution of government debt-to-GDP ratio in our sample. Table A3
provides the speci…c percentile values from the sample.
16
Our estimates suggest that the long-run …scal multiplier can be as low as -3 when the …scal position is weak.
One way to reconcile such a magnitude is in terms of the private investment response: private investment declines
signi…cantly in response to a positive …scal shock during times of weak …scal position (Figure 7).
17
The median multipliers for all horizons are presented in Figure A9 in the Appendix. Our headline result remains
robust when we split the sample into advanced and developing economies. See Figure A3 in the Appendix.
8
…scal position). For comparability, the shock size in each case is normalized such that government
consumption rises by 1 percentage point on impact. The conditional impulses are shown in Figure
4.
While output increases on impact and remains signi…cantly positive for around 2 years when
the …scal position is strong, such stimulative e¤ects dissipate after about a year with output falling
signi…cantly below zero through till the end of the projection horizon.18 In the case of government
consumption, the conditional impulses for both strong and weak …scal positions exhibit some per-
sistence in response to the positive …scal shock. However, …scal expansion is more quickly unwound
when the …scal position is strong than weak. In other words, relative to the strong …scal position,
the government in fact spends more, especially during the initial periods, when …scal position is
weak.19 Despite this, it is then quite remarkable that output falls more during times of weak …scal
position. This is a result that reinforces our earlier point that a weak …scal position can blunt the
stimulative e¤ects of expansionary …scal policy.
3.3 Distinguishing between Two States: Business Cycle and Fiscal Position
Recent studies (e.g. Auerbach and Gorodnichenko 2012a) have established that …scal multipliers
depend on the phase of the business cycle: they tend to be larger during recessions than expansions.
To the extent that …scal position is endogenous and varies according to the business cycle, it is
possible that our empirical exercise so far of conditioning only on debt is simply capturing business
cycle e¤ects. Controlling for business cycle e¤ects is therefore important to establish that …scal
position is a unique state that matters for the size of the …scal multipliers. In this section, we
undertake a multi-pronged sequence of empirical exercises designed to demonstrate this.
First, we tabulate a number of descriptive statistics to verify that there is little relationship
between incidences of the two states. The top panel of Table 1 computes the relative frequency
in which countries in our sample experience both a strong or weak …scal position state and a
recession.20 The fact is that the two states rarely coincide: for the pooled sample, the concurrence
of both states occurs around 2 percent of the time. Even for the category with the highest relative
frequency - developing economies with a weak …scal state undergoing a recession - the coincidence
of these states is very infrequent (at most 3 percent of the time).21
Second, we perform a number of formal tests that compares the distribution of …scal position
(debt-to-GDP ratio) during recessions and expansions. These are reported in the bottom panel of
Table 1. It is clear that any di¤erences - to the extent that they exist - are minimal: for instance,
the average debt-to-GDP ratio in the expansionary state is 52 percent, compared to 54 percent
18
To allay any concerns that the choice of the 10th and 90th percentiles merely re‡ ects outliers, we report results for
the 25th and the 75th percentiles as well (Figure A4 in the Appendix). Even though the di¤erences in the conditional
impulse responses are admittedly not as sharp as before, they are statistically signi…cant in the relevant horizons so
that our conclusion remains robust.
19
Nickel and Tudyka (2014) also report similar …ndings, although government consumption in their study is un-
wound at longer horizons during times of high debt.
20
Like before, the strong …scal position corresponds to the 10th percentile of debt-to-GDP ratio in the sample while
the weak …scal position corresponds to the 90th percentile. The recessionary state is de…ned as the period from peak
to trough as determined by the Harding and Pagan (2002) business cycle dating algorithm. We discuss alternative
approaches to date the business cycle in the robustness exercise in Section 5.
21
We check the relative frequencies at the country level as well. There are several countries for which there are no
recessionary episodes either during periods of strong or weak …scal positions. Beyond those countries, incidences of
recessions are generally lower when …scal position is strong. These results are available upon request.
9
during recessions. More formally, the t tests all fail to reject the null hypothesis of no di¤erence in
means at the standard con…dence levels. In e¤ect, there is little evidence that the distributions of
…scal position in our sample di¤er between recessionary and expansionary states.
Our third approach is to estimate …scal multipliers conditional on the …scal position while
explicitly controlling for business cycle e¤ects. For that, we replace Equation (2) by the following
expression that jointly conditions the model coe¢ cients on both the …scal position and the business
cycle state as follows:
jk jk jk jk
l;it = 1;l + 2;l f sit + 2;l bcit ; (4)
where bc is an indicator variable that equals 1 for a recession and 0 for an expansion as de-
termined by the Harding-Pagan (2002) dating algorithm. The IPVAR system now comprises of
Equations (1) and (4).
Figure 5 presents estimates of the multipliers for di¤erent …scal positions during recessions.
Compared with the earlier result (Figure 3), which e¤ectively spans both phases of the business
cycle, the magnitude of the multipliers during recessions (Figure 5) is larger for any given level of
…scal position. For instance, the point estimate of the long-run multiplier for the strongest …scal
position during recessions almost reaches 1.5, while it is less than 1 when conditioned only on
the …scal position. This echoes the empirical literature that has argued that multipliers tend to be
larger during recessions (Auerbach and Gorodnichenko 2012b; Bachmann and Sims 2012; Candelon
and Lieb 2013). Our results show that multipliers remain dependent on …scal position even during
recessions: estimated multipliers decline monotonically in debt for all horizons.22
One important corollary of this result is that the multiplier can be small even during recessions,
if the …scal position is weak. This is especially the case in the longer-run, as the implications
of a heavier debt burden on private demand ultimately play out. Where the …scal position is
especially weak, the multiplier even turns signi…cantly negative. Our central result, therefore,
nuances other …ndings that multipliers are larger during recessions than expansions. Conditioning
our IPVAR only on the phase of the business cycle, we indeed obtain similar results reported in
earlier studies (Figure 6) (e.g. Auerbach and Gorodnichenko 2012b).23 Yet, our results based on
the joint conditioning show that, even during recessions, multipliers can be small and even negative
if …scal position is weak.
4 Why Fiscal Positions Matter - Transmission Channels
The key mechanism that could reduce multipliers when …scal positions are weak, especially in
the long run, rests on private agents’ concerns about …scal sustainability when the government
implements expansionary …scal policy. As mentioned earlier, this can operate via reductions in
private consumption as households anticipate a larger tax burden in the future (the Ricardian
channel), or via reductions in consumption and investment by investors facing an ever-greater
22
The state-dependency of …scal multipliers on the …scal position also holds during expansions (Figure A5). For a
given level of government debt, the estimated multipliers are larger during recessions than expansions.
23
Despite the di¤erences in econometric approaches and sample, the precise magnitude of our multipliers during
recessions and expansions is comparable with Auerbach and Gorodnichenko (2012b). For instance, their point estimate
of the long-run multiplier (when government consumption is the …scal instrument) is around 1.47. The corresponding
number from the IPVAR model is around 1.67.
10
borrowing costs (the interest rate channel). In this section, we attempt to assess the relative
strength of these two channels.
We …rst consider the Ricardian channel by augmenting the IPVAR system with private con-
sumption, with the model coe¢ cients conditioned on …scal position. For this speci…cation, we
order private consumption right after GDP, thus keeping intact the recursive identi…cation scheme
of Blanchard and Perotti (2002). Ordering the current account and exchange rates last preserves
a domestic macroeconomic bloc in the IPVAR. The conditional impulse responses of private con-
sumption and output to the …scal shock, for both the strong and the weak …scal position, are
presented in the left panel of Figure 7. As before, the strong and the weak …scal positions re-
spectively correspond to the 10th and 90th percentile of debt-to-GDP ratio from our sample. We
check the robustness of our results by choosing the 25th and 95th percentiles (Figure A6 in the
Appendix).
The results are unambiguous: when the …scal position is strong, private consumption rises fol-
lowing the impact of the …scal shock, peaking around a year after the shock before returning to its
initial level. On the other hand, when the …scal position is weak, private consumption falls precip-
itously and remains depressed for around three years after the …scal shock. During these horizons,
the di¤erence in the response of private consumption is also statistically signi…cant, judging from
the non-overlapping con…dence bands. The divergence in private consumption responses across
strong and weak …scal positions is consistent with the Ricardian channel outlined earlier where
households reduce consumption in anticipation of more imminent …scal adjustments during times
of high government debt (Sutherland 1997 and Perroti 1999).24
Our result on the divergence of private consumption paths provides a new dimension on the
debate concerning how private consumption responds to …scal stimulus. Perroti (2005) …nds that
private consumption rises in response to a positive …scal shock, while Ramey (2011) shows that
private consumption actually declines –a di¤erence which is attributed to the speci…c identi…cation
scheme used in these studies. Ilzetzki, Mendoza, and Vegh (2013) reconcile these two contrasting
views in terms of monetary policy behavior and argue that once monetary policy is controlled for,
…scal policy has expansionary e¤ects on private consumption. Our results, by explicitly showing
how a weak …scal position undermines and reverses the response of private consumption, suggest
an additional aspect that can help reconcile the con‡icting results found in the literature.
For the interest rate channel, we would ideally introduce a proxy for sovereign risk, such as the
yield spread, directly into our IPVAR system. However, this is precluded by the paucity of credible
long-term rates, especially in developing countries, at the quarterly frequency. We thus proceed
with our second-best option, which is to augment private investment into the IPVAR system. As
in the case of private consumption, private investment is ordered after GDP but before the current
account. Since private investment is particularly sensitive to borrowing costs, a reduction in private
investment during times of weak …scal position is indicative of the interest rate channel. Figure 7
presents the conditional impulse responses of private investment for both weak and strong …scal
positions.25
24
The estimates of the multipliers with this speci…cation are broadly in line with the baseline estimates. More
importantly, our headline result that multipliers depend on …scal position holds when private consumption is included
in the IPVAR. See Figure A7 in the Appendix.
25
The multipliers are presented in Figure A8 in the Appendix. Our main result that multipliers depend on …scal
position generally holds.
11
The contrast between strong and weak …scal positions for the path of private investment is,
again, striking. Investment rises signi…cantly when the …scal position is strong, peaking after
around 6 quarters, but remaining sustained through at least 10 quarters. When the …scal position
is weak, investment drops sharply after about a year, and never fully recovers, failing to revert
even after 5 years. The di¤erence in the impulse responses across strong and weak …scal positions
is also statistically signi…cant (barring the initial few quarters). These responses are qualitatively
similar to those of private consumption but much larger in magnitude. This suggests that investor
concerns about borrowing cost could be an additional channel for dampening the e¤ectiveness of
…scal policy.
5 Robustness Exercises
We consider three exercises to check the robustness of our headline …ndings: (a) an alternative
measure of …scal position where we use …scal balances instead of government debt; (b) an alternative
dating scheme of the business cycle similar to Auerbach and Gorodnichenko (2012b) to de…ne
recessions as periods with a signi…cant probability of negative output growth; (c) estimating …scal
position dependent multipliers while controlling for exchange rate regimes.26 For the last exercise,
we estimate the IPVAR model by jointly conditioning the model coe¢ cients on both the …scal
position and an exchange rate regime dummy. The law of motion of the model coe¢ cients then is:
jk jk jk jk
l;it = 1;l + 2;l f sit + 2;l erit ; (5)
where er is an indicator variable that equals 1 for a …xed exchange rate regime and 0 for a
‡exible exchange rate regime.27 The measure of …scal position, f s, is taken to be the government
debt-to-GDP ratio as in the baseline speci…cation.
Table 2 presents the results. The top panel shows the range of estimates of the …scal multipliers
for the strongest and weakest …scal positions which, like before, are taken to be …scal balances
corresponding to the 10th (weak) and 90th (strong) percentiles from the sample. By and large, our
baseline results are qualitatively similar when …scal balances, instead of debt-to-GDP ratios, are
used to measure …scal positions. That is, the multipliers are systematically larger for high …scal
balances (strong …scal position) than low …scal balances (i.e. weak …scal position). This is true
regardless of the horizons considered and when jointly conditioned on the state of the business
cycle. The middle panel of Table 2 presents the multipliers using the alternative de…nition of
recessions. Our headline result – multipliers depend on the …scal position even during recessions
- generally holds, especially at longer horizons. The bottom panel presents the …scal position-
dependent multipliers for ‡ exible and …xed exhange rates. Consistent with the literature (e.g.
Ilzetzki, Mendoza, and Vegh 2013), for a given …scal position, multipliers are larger in …xed than
‡ exible exchange rate regimes. That said, the state dependency on …scal position still holds:
multipliers are larger when …scal position is strong than weak irrespective of the exchange rate
regime.
26 exp( zit )
Following Auerbach and Gorodnichenko (2012b), we de…ne the indicator function, I (zit ) = 1+exp( zit )
, where
zit is taken to be 7 quarter moving averages of quarter-to-quarter growth rates normalized to have a zero mean and a
unit variance. Calibrating as 1:5 > 0, the indicator function pins down the probability of negative output growth.
Recessions are then de…ned as periods where that probability exceeds a threshold, which in our implementation is
taken to be 80 percent.
27
The exchange rate regime classi…cation follows Ilzetzki, Mendoza, and Vegh (2013) which is extended until 2014
using the IMF de-facto classi…cation of exchange rates.
12
6 Conclusion
We document that …scal multipliers tend to be larger when the …scal position is stronger. For
instance, our estimates suggest that the long run multiplier can be as big as unity when the …scal
position is strong but it can turn negative when the …scal position is weak. A weak …scal position
can undermine …scal multipliers even during recessions. Consistent with theoretical predictions,
we provide empirical evidence suggesting that weak …scal positions are associated with smaller
multipliers through both a Ricardian channel and an interest rate channel.
Future work can usefully focus on two issues. First, while data limitations have precluded a
deeper and more direct exploration of the interest rate channel, future research, perhaps with more
comprehensive and representative data on yield spreads, can seek to improve our understanding
of the interest rate channel. Second, …scal-monetary interactions can be studied using a similar
empirical model like ours. In particular, one can evaluate whether monetary policy o¤ers a more
e¤ective stabilization tool during times of weak …scal position.
13
References
[1] Auerbach, A. J. and Y. Gorodnichenko (2012a). “Fiscal Multipliers in Recession and Expan-
sion,” in Fiscal Policy after the Financial Crisis, ed. by A. Alesina and F. Giavazzi, Chicago,
IL: University of Chicago Press, 63-98.
[2] — — –(2012b). “Measuring the Output Responses to Fiscal Policy,”American Economic Jour-
nal: Economic Policy, 4, 1-27.
[3] Bachmann, R. and E. R. Sims (2012). “Con…dence and the Transmission of Government Spend-
ing Shocks,” Journal of Monetary Economics, 59, 235-249.
[4] Batini, N.L. and A. Weber (2014). “A Simple Method to Calculate Fiscal Multipliers,”Working
Paper 14/93, International Monetary Fund, Washington DC.
[5] Benhabib, J. and M. M. Spiegel (2000). “The Role of Financial Development in Growth and
Investment,” Journal of Economic Growth, 5, 341-360.
[6] Bi, H., W. Shen, and S.C. S. Yang (2014). “Fiscal Limits, External Debt, and Fiscal Policy in
Developing Countries,” IMF Working Paper 14/49.
[7] Blanchard, O., G. Dellariccia, and P. Mauro (2010): “Rethinking Macroeconomic Policy,”
Journal of Money, Credit and Banking, 42, 199–215.
[8] Blanchard, O., G. Dellariccia, and P. Mauro (2013): “Rethinking Macroeconomic Policy II:
Getting Granular,” IMF Sta¤ Discussion Note 13/03.
[9] Blanchard, O. and R. Perotti (2002). “An Empirical Characterization of the Dynamic E¤ects
of Changes in Government Spending and Taxes on Output,”Quarterly Journal of Economics,
117, 1329-1368.
[10] Booth, L., V. A. Aivazian, A. Demiruc-Kunt, and V. Maksimovic (2001). “Capital Structures
in Developing Countries,” Journal of Finance, 56, 87-130.
[11] Candelon, B. and L. Lieb (2013). “Fiscal Policy in Good and Bad Times,”Journal of Economic
Dynamics and Control, 37, 2679-2694.
[12] Christiano, L. J., M. S. Eichenbaum, and S. T. Rebelo (2011). “When Is the Government
Spending Multiplier Large?” Journal of Political Economy, 119, 78-121.
[13] Corsetti, G., K. Kuester, A. Meier, and G. J. Muller (2013). “Sovereign Risk, Fiscal Policy,
and Macroeconomic Stability,” Economic Journal, 123, F99-F132.
[14] Corsetti, G., A. Meier, and G. J. Muller (2012). “Fiscal Stimulus with Spending Reversals,”
Review of Economics and Statistics, 94, 878-895.
[15] Delong, B. and L.H. Summers (2012). “Fiscal Policy in a Depressed Economy,” Brookings
Papers on Economic Activity.
[16] Denes, M. R., G. B. Eggertsson, and S. Gilbukh (2013). “De…cits, Public Debt Dynamics and
Tax and Spending Multipliers,” Economic Journal, 123, F133-F163.
[17] Erceg, C. J. and J. Linde (2014). “Is There A Fiscal Free Lunch In A Liquidity Trap?”Journal
of the European Economic Association, 12, 73-107.
14
[18] Harding, D. and A. R. Pagan (2002). “Dissecting the Cycle: A Methodological Investigation,”
Journal of Monetary Economics, 49, 365-381.
[19] Huidrom, R., A. Kose, and F. Ohnsorge (2016). “Having Space and Using It: Fiscal Policy
Challenges in Developing Economies,” CEPR Working Paper forthcoming.
[20] Ilzetzki, E. O., E. G. Mendoza, and C. A. Vegh (2013). “How Big (Small?) Are Fiscal Multi-
pliers?” Journal of Monetary Economics, 60, 239-254.
[21] Kraay, A. C. (2012). “How Large Is the Government Spending Multiplier? Evidence from
World Bank Lending,” Quarterly Journal of Economics, 127, 829-887.
[22] Mineshima, A., M. Poplawski-Ribeiro, and A. Weber (2014). “Size of Fiscal Multipliers,” in
Post-Crisis Fiscal Policy, ed. by C. Cottarelli, P. Gerson, and A. Senjadji, Cambridge, MA:
MIT Press, 315-372.
[23] Nickel, C. and A. Tudyka (2014). “Fiscal Stimulus in Times of High Debt: Reconsidering
Multipliers and Twin De…cits,” Journal of Money, Credit, and Banking, 46(7), 1313-1344.
[24] Owyang, M. T., V. A. Ramey, and S. Zubairy (2013). “Are Government Spending Multi-
pliers Greater during Periods of Slack? Evidence from Twentieth-Century Historical Data,”
American Economic Review, 103, 129-134.
[25] Perotti, R. (1999). “Fiscal Policy in Good Times and Bad,” Quarterly Journal of Economics,
114, 1399-1436.
[26] — — –(2005). “Estimating the E¤ects of Fiscal Policy in OECD Countries,”CEPR Discussion
Paper 4842.
[27] Ramey, V.A. (2011). “Identifying Government Spending Shocks: It’s all in the Timing,”Quar-
terly Journal of Economics 126, 1-50.
[28] Ramey, V.A. and S. Zubairy (2014). “Government Spending Multipliers in Good Times and
in Bad: Evidence from U.S. Historical Data,” NBER Working Paper No. 20719
[29] Riera-Crichton, D., C. A. Vegh, and G. J. Vuletin (2014). “Procyclical and Countercyclical
Fiscal Multipliers: Evidence from OECD Countries,” Policy Research Working Paper 6993,
NBER, Washington, DC.
[30] Romer, C.D. and D. H. Romer (2010). “The Macroeconomic E¤ects of Tax Changes: Estimates
Based on a New Measure of Fiscal Shocks,” American Economic Review, 100(3), 763-801.
[31] Sa, F. G., P. Towbin, and T. W. Wieladek (2014). “Capital In‡ows, Financial Structure and
Housing Booms,” Journal of the European Economic Association, 12, 522-546.
[32] Sutherland, A. J. (1997). “Fiscal Crises and Aggregate Demand: Can High Public Debt Reverse
the E¤ects of Fiscal Policy?” Journal of Public Economics, 65, 147-162.
[33] Towbin, P. and S. Weber (2013). “Limits of Floating Exchange Rates: The Role of Foreign
Currency Debt and Import Structure,” Journal of Development Economics, 101, 179-194.
15
Figure 1: Unconditional Multipliers
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
Impact 1 year 2 years Long run
Note: The graph shows the unconditional fiscal multipliers for select horizons. These are based on estimates from the
SVAR model of Ilzetzki, Mendoza, and Vegh (2013) that features with no interaction terms. Bars represent the median,
and error bands are the 16-84 percent confidence bands.
16
Figure 2: Unconditional Impulse Responses
1.20 A. Government Consumption 0.08 B. GDP
1.00 0.06
0.04
0.80
0.02
0.60
0.00
0.40
-0.02
0.20
-0.04
0.00 -0.06
-0.20 -0.08
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Note: The graphs show the unconditional impulse responses (percentage points) to a positive shock to government consumption. These are based on estimates
from the SVAR model of Ilzetzki, Mendoza, and Vegh (2013) that features no interaction terms. Solid lines represent the median, and dotted lines are the 16-84
percent confidence bands.
17
Figure 3: Fiscal Position-Dependent Multipliers
A. On Impact B. 1 Year
0.3 0.7
0.6
0.2 0.5
0.4
0.1
0.3
0.2
0.0
0.1
-0.1 0.0
-0.1
-0.2 -0.2
12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4 12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4
C. 2 Years D. Long Run
1.0 2.0
0.8
0.6 1.0
0.4 0.0
0.2
0.0 -1.0
-0.2
-0.4 -2.0
-0.6 -3.0
-0.8
-1.0 -4.0
12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4 12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4
Note: The graphs show the conditional fiscal multipliers for different levels of fiscal position at select horizons. These are based on estimates from the IPVAR
model, where model coefficients are conditioned only on fiscal position. Government debt as a percentage of GDP is the measure of fiscal position and the values
shown on the x-axis correspond to the 5th to 95th percentiles from the sample. Fiscal position is strong (weak) when government debt is low (high). Solid lines
represent the median, and dotted bands are the 16-84 percent confidence bands.
18
Figure 4: Conditional Impulse Responses
1.20 A. Government Consumption 0.15 B. GDP
1.00 0.10
0.80 0.05
0.00
0.60
-0.05
0.40
-0.10
0.20 -0.15
0.00 -0.20
-0.20 -0.25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Weak fiscal position Strong fiscal position
Note: The graphs show the conditional impulse responses (percentage points) for the strong (blue) and the weak (red) fiscal positions. These are based on estimates
from the IPVAR model, where model coefficients are conditioned only on fiscal position. Government debt as a percentage of GDP is the measure of fiscal position.
The strong fiscal position corresponds to the 10th percentile of debt-to-GDP ratio from the sample, while the weak fiscal position corresponds to the 90 th percentile.
Solid lines represent the median, and dotted bands a`re the 16-84 percent confidence bands.
19
Table 1: Comparison of Fiscal and Business Cycle States
Full Sample Advanced Developing
a
Relative frequency
Strong fiscal and recessionary state 2.2 2.4 1.8
Weak fiscal and recessionary state 2.1 2.4 3.0
Test of differences
b [52.3, 54.0] [57.3, 57.9] [43.4, 44.6]
In means
0.25 0.76 0.55
Note: The table shows the association (or lack thereof) between different fiscal positions and the recessionary state.
a
The top panel shows the relative frequency (percent of observations) of the strong fiscal position and the recessionary
state, and that of weak fiscal position and the recessionary state. The frequencies are reported for the full sample and
also for specific country groups: advanced and developing economies. The strong (weak) fiscal position corresponds
to the 10th (90th) percentile of debt-to-GDP ratio in each sample. The bottom panel reports results that show the
statistical significance of the difference of those relative frequencies. The recessionary state is determined by the
Harding-Pagan (2002) business cycle dating algorithm.
b
The top entry shows the average debt-to-GDP ratio (in percent) during expansions (left) and recessions (right). The
bottom entry shows the p-values of two-group t-test of difference in means with unequal variances.
20
Figure 5: Fiscal Position-Dependent Multipliers during Recessions
A. On Impact B. 1 Year
0.7 1.6
0.6 1.4
0.5 1.2
1.0
0.4
0.8
0.3
0.6
0.2 0.4
0.1 0.2
0.0 0.0
12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4 12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4
C. 2 Years D. Long Run
2.5 2.5
2.0
2.0
1.5
1.5 1.0
0.5
1.0
0.0
0.5 -0.5
-1.0
0.0
-1.5
-0.5 -2.0
12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4 12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4
Note: The graphs show the conditional fiscal multipliers during recessions for different levels of fiscal position at select horizons. These are based on estimates
from the IPVAR model, where model coefficients are jointly conditioned on fiscal position and the phase of the business cycle. Government debt as a percentage
of GDP is the measure of fiscal position and the values shown on the x-axis correspond to the 5th to 95th percentiles from the sample. Recessions are determined
by the Harding-Pagan (2002) business cycle dating algorithm. Fiscal position is strong (weak) when government debt is low (high). Solid lines represent the
median, and dotted bands are the 16-84 percent confidence bands.
21
Figure 6: Fiscal Multipliers by Business Cycles only
4
Recessions Expansions
3
2
1
0
-1
-2
On Impact 1 year 2 years Long run
Note: The graph shows the conditional fiscal multipliers during recessions at select horizons. These are based on
estimates from the IPVAR model, where model coefficients are conditioned only on the phase of the business cycle.
Recessions are determined by the Harding-Pagan (2002) business cycle dating algorithm. Bars represent the median,
and error bands are the 16-84 percent confidence bands.
22
Figure 7: Transmission Channels
0.15 A. Private Consumption 0.40 B. Private Investment
0.10
0.20
0.05
0.00
0.00
-0.05 -0.20
-0.10
-0.40
-0.15
-0.60
-0.20
-0.25 -0.80
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Weak fiscal position Strong fiscal position
Note: The graphs show the conditional impulse responses (percentage points) of private consumption and private investment due to a positive shock to government
consumption for the strong (blue) and the weak (red) fiscal positions. These are based on estimates from the IPVAR model, where model coefficients are conditioned
only on fiscal position. Government debt as a percentage of GDP is the measure of fiscal position. The strong fiscal position corresponds to the 10th percentile of
debt-to-GDP ratio from the sample, while the weak fiscal position corresponds to the 90 th percentile. Solid lines represent the median, and dotted bands are the 16-
84 percent confidence bands.
23
Table 2: Robustness Checks: Fiscal Multipliers
Fiscal position only Recessions and fiscal position
Impact 1 year 2 years Long run Impact 1 year 2 years Long run
Alternative fiscal position
Fiscal balances Strong [0.16, 0.31] [0.29, 0.63] [0.43, 1.10] [0.39, 1.38] [0.37, 0.57] [1.34, 2.00] [1.76, 2.66] [1.09, 2.35]
Weak [-0.08, 0.08] [0.16, 0.54] [0.19, 0.76] [-0.05, 0.97] [-0.02, 0.24] [1.04, 1.78] [1.25, 2.16] [0.65, 1.74]
Alternative business cycle dates
Auberch and Gorodnichenko (2012b) Strong … … … … [0.54, 0.85] [1.23, 1.75] [1.13, 1.78] [0.61, 1.48]
Weak … … … … [0.48, 0.80] [1.36, 1.96] [0.85, 1.52] [-0.56, 0.54]
Flexible exchange rate and fiscal position Fixed exchange rate and fiscal position
Controlling for exchange rate regime
Strong [0.02, 0.23] [-0.06, 0.44] [-0.13, 0.62] [-0.37, 0.81] [0.72, 1.01] [2.11, 2.84] [2.26, 3.29] [1.44, 3.18]
Weak [-0.14, 0.07] [-0.23, 0.15] [-0.63, -0.08] [-2.28, -0.90] [0.53, 0.90] [1.57, 2.32] [1.10, 2.20] [-7.81, -0.72]
Note: The table presents estimates of fiscal multipliers from alternative specifications of the IPVAR model for the strong and the weak fiscal positions. The top
panel presents the multipliers using an alternative measure of fiscal position. The middle panel considers an alternative business cycle dating scheme. The bottom
panel presents estimates of fiscal position-dependent multipliers for flexible and fixed exchange rate regimes. Fiscal position is strong (weak) when government
debt is high (low) or when fiscal balances are low (high). When fiscal position is measured in terms of government debt, the strong position corresponds to the 10th
percentile and the weak position corresponds to the 90th percentile. When fiscal balances are taken as the measure of fiscal position, the strong position corresponds
to the 90th percentile and the weak position corresponds to the 10th percentile. Numbers reported in square brackets are the 16-84 percent confidence range.
24
List of Figures and Tables in the Supplementary Appendix
Figure A1: Distribution of Fiscal Position
Figure A2: Unconditional Multipliers
Figure A3: Fiscal Position-Dependent Multipliers by Country Groups
Figure A4: Conditional Impulse Responses – Alternative Cut-offs
Figure A5: Fiscal Position-Dependent Multipliers during Expansions
Figure A6: Transmission Channels – Alternative Cut-offs
Figure A7: Fiscal Position-Dependent Multipliers with Private Consumption
Figure A8: Fiscal Position-Dependent Multipliers with Private Investment
Figure A9: Fiscal Position-Dependent Multipliers – All Horizons
Table A1: Country Coverage
Table A2: Data Sources
Table A3: Distribution of Fiscal Position
25
Supplementary Appendix
Do Fiscal Multipliers Depend on Fiscal Positions?
Raju Huidrom, M. Ayhan Kose, Jamus J. Lim, and Franziska L. Ohnsorge
This appendix provides additional results to the main paper.
26
Table A1: Country Coverage
Advanced Developing
Country Period Country Period
Australia 1980:1--2014:1 Argentina 1993:1--2014:1
Belgium 1991:1--2014:1 Bulgaria 1999:1--2014:1
Canada 1980:1--2014:1 Brazil 1995:1--2014:1
Germany 1991:1--2014:1 Chile 1989:1--2014:1
Denmark 1999:1--2014:1 Colombia 2000:1--2014:1
Spain 1995:1--2014:1 Czech Republic 1999:1--2014:1
Finland 1998:1--2014:1 Croatia 2000:1--2014:1
France 1980:1--2014:1 Hungary 1995:1--2014:1
United Kingdom 1980:1--2014:1 Israel 1999:1--2014:1
Iceland 1997:1--2014:1 Mexico 1991:1--2014:1
Italy 1999:1--2014:1 Poland 1999:1--2014:1
Lithuania 1995:1--2014:1 Romania 1998:1--2014:1
Netherlands 1988:1--2014:1 Slovak Republic 1999:1--2014:1
Norway 1996:1--2014:1 South Africa 1993:1--2014:1
Puerto Rico 1980:1--2014:1 Turkey 1998:1--2014:1
Slovenia 1995:1--2014:1
Sweden 1993:1--2014:1
United States 1980:1--2014:1
Note: The table shows the list of countries in the sample. Coverage corresponds to maximum temporal coverage for
each country in the baseline specification of the IPVAR model. The coverage differs for specifications used in the
robustness exercises.
27
Table A2: Data Sources
Variable Definition Frequency Source
Output Real gross domestic product (GDP) Quarterly Ilzetzki, Mendoza, and Vegh (2013), OECD, Haver Analytics
Private consumption Real personal consumption expenditure Quarterly Ilzetzki, Mendoza, and Vegh (2013), OECD, Haver Analytics
Private investment Real private gross fixed capital formation Quarterly
Government consumption Real government consumption expenditure a Quarterly Ilzetzki, Mendoza, and Vegh (2013), OECD, Haver Analytics
a
Government investment Real government gross fixed capital formation Quarterly OECD, Haver Analytics, Eurostat
b
Real effective exchange rate Real effective exchange rate Quarterly Ilzetzki, Mendoza, and Vegh (2013), BIS
Current account Current account as percent of GDP Quarterly Ilzetzki, Mendoza, and Vegh (2013), WEO
Government debt General government debt as percent of GDP Annual WEO
Fiscal balance Overall fiscal balance as percent of GDP Annual WEO
Government consumption-to-GDP ratio Government consumption as percent of GDP Annual WDI
Government investment-to-GDP ratio Government investment as percent of GDP Annual WDI
Interest rate Short term nominal interest rate Quarterly Ilzetzki, Mendoza, and Vegh (2013)
Note: The main source for the quarterly series is Ilzetzki, Mendoza, and Vegh (2013). This database which ends around 2008 is extended by splicing from different
sources as mentioned in the table.
a
This refers to general government for most countries while for a few countries central government is taken. See Ilzetzki, Mendoza, and Vegh (2013).
b
The narrow index wherever available is taken while the remainder uses the broad index. Details are available upon request.
28
Figure A1: Distribution of Fiscal Position
25
20
Frequency (%)
15
10
5
0
0 10 20 30 40 50 60 70 80 90 100 110 120 130
Debt-GDP Ratio
Note: The graph shows the distribution of fiscal position, taken to be the annual government debt-to-GDP ratio, from the sample of advanced and developing
economies during the period 1980-2014.
Table A3: Distribution of Fiscal Position
Percentile 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95
Debt-GDP Ratio 12.4 17.3 24.8 28.7 32.2 35.9 38.5 40.7 42.9 45.1 47.9 51.4 56.1 60.1 65.5 71.3 80.5 92.4 107.4
Note: The table shows the percentile values of fiscal position, taken to be annual government debt-to-GDP ratio, from the sample of advanced and developing
economies during the period 1980-2014.
29
Figure A2: Unconditional Multipliers
A. Advanced Economies B. Developing Economies
0.6 0.6
0.4 0.4
0.2 0.2
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
-0.8 -0.8
-1.0 -1.0
Impact 1 year 2 years Long run Impact 1 year 2 years Long run
Note: The graph shows the unconditional fiscal multipliers for select horizons. Panel A uses a sample of advanced economies only while Panel B uses only
developing economies. These are based on estimates from the SVAR model of Ilzetzki, Mendoza, and Vegh (2013) that features with no interaction terms. Bars
represent the median, and error bands are the 16-84 percent confidence bands.
30
Figure A3: Fiscal Position-Dependent Multipliers by Country Groups
Advanced Economies
A. On Impact B. Long Run
2.0 3.0
2.0
1.0 1.0
0.0
-1.0
0.0
-2.0
-3.0
-1.0 -4.0
-5.0
-2.0 -6.0
12 26 35 41 47 53 60 71 90 116 12 26 35 41 47 53 60 71 90 116
Developing Economies
C. On Impact D. Long Run
0.4 2.0
1.5
0.3 1.0
0.5
0.2 0.0
-0.5
0.1 -1.0
-1.5
0.0 -2.0
-2.5
-0.1 -3.0
12 15 20 27 29 32 35 37 38 40 42 43 45 47 53 60 66 72 81 12 15 20 27 29 32 35 37 38 40 42 43 45 47 53 60 66 72 81
Note: The graphs show the conditional fiscal multipliers for different levels of fiscal position at select horizons. These are based on estimates from the IPVAR
model, where model coefficients are conditioned only on fiscal position. The top (bottom) panel is based a sample of only advanced (developing) economies.
Government debt as a percentage of GDP is the measure of fiscal position and the values shown on the x-axis correspond to the 5th to 95th percentiles from each
sample. Fiscal position is strong (weak) when government debt is low (high). Solid lines represent the median, and dotted bands are the 16-84 percent confidence
bands.
31
Figure A4: Conditional Impulse Responses – Alternative Cut-offs
1.20 A. Government Consumption 0.10 B. GDP
1.00
0.05
0.80
0.60 0.00
0.40 -0.05
0.20
-0.10
0.00
-0.20 -0.15
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Weak fiscal position Strong fiscal position
Note: The graphs show the conditional impulse responses (percentage points) for the strong (blue) and the weak (red) fiscal positions. These are based on estimates
from the IPVAR model, where model coefficients are conditioned only on fiscal position. Government debt as a percentage of GDP is the measure of fiscal position.
The strong fiscal position corresponds to the 25th percentile of debt-to-GDP ratio from the sample, while the weak fiscal position corresponds to the 75 th percentile.
Solid lines represent the median, and dotted bands are the 16-84 percent confidence bands.
32
Figure A5: Fiscal Position-Dependent Multipliers during Expansions
A. On Impact B. 1 Year
0.3 0.6
0.2 0.4
0.1 0.2
0.0 0.0
-0.1 -0.2
-0.2 -0.4
-0.3 -0.6
12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4 12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4
C. 2 Years D. Long Run
1.2 3.0
0.8 2.0
1.0
0.4
0.0
0.0
-1.0
-0.4
-2.0
-0.8 -3.0
-1.2 -4.0
12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4 12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4
Note: The graphs show the conditional fiscal multipliers during expansions for different levels of fiscal position at select horizons. These are based on estimates
from the IPVAR model, where model coefficients are jointly conditioned on fiscal position and the phase of the business cycle. Government debt as a percentage
of GDP is the measure of fiscal position and the values shown on the x-axis correspond to the 5th to 95th percentiles from the sample. Expansions are determined
by the Harding-Pagan (2002) business cycle dating algorithm. Fiscal position is strong (weak) when government debt is low (high). Solid lines represent the
median, and dotted bands are the 16-84 percent confidence bands
33
Figure A6: Transmission Channels – Alternative Cut-offs
0.10 A. Private Consumption 0.30 B. Private Investment
0.20
0.05
0.10
0.00 0.00
-0.05 -0.10
-0.20
-0.10
-0.30
-0.15 -0.40
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Weak fiscal position Strong fiscal position
Note: The graphs show the conditional impulse responses (percentage points) of private consumption and private investment due to a positive shock to government
consumption for the strong (blue) and the weak (red) fiscal positions. These are based on estimates from the IPVAR model, where model coefficients are conditioned
only on fiscal position. Government debt as a percentage of GDP is the measure of fiscal position. The strong fiscal position corresponds to the 25th percentile of
debt-to-GDP ratio from the sample, while the weak fiscal position corresponds to the 75th percentile. Solid lines represent the median, and dotted bands are the 16-
84 percent confidence bands.
34
Figure A7: Fiscal Position-Dependent Multipliers with Private Consumption
A. On Impact B. 1 Year
0.3 0.6
0.2
0.5
0.2
0.4
0.1
0.1 0.3
0.0
0.2
-0.1
0.1
-0.1
-0.2 0.0
12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4 12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4
C. 2 Years D. Long Run
1.0 1.5
0.8 1.0
0.6 0.5
0.4 0.0
0.2 -0.5
0.0 -1.0
-0.2 -1.5
-0.4 -2.0
12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4 12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4
Note: The graphs show the conditional fiscal multipliers for different levels of fiscal position at select horizons. These are based on estimates from the IPVAR
model that includes private consumption. The model coefficients are conditioned only on fiscal position. Government debt as a percentage of GDP is the measure
of fiscal position and the values shown on the x-axis correspond to the 5th to 95th percentiles from the sample. Fiscal position is strong (weak) when government
debt is low (high). Solid lines represent the median, and dotted bands are the 16-84 percent confidence bands.
35
Figure A8: Fiscal Position-Dependent Multipliers with Private Investment
A. On Impact B. 1 Year
0.3 0.8
0.2 0.6
0.4
0.1
0.2
0.0
0.0
-0.1 -0.2
-0.2 -0.4
12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4 12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4
C. 2 Years D. Long Run
1.0 2.0
1.0
0.5
0.0
0.0 -1.0
-0.5 -2.0
-3.0
-1.0
-4.0
-1.5 -5.0
12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4 12.4 24.8 32.2 38.5 42.9 47.9 56.1 65.5 80.5 107.4
Note: The graphs show the conditional fiscal multipliers for different levels of fiscal position at select horizons. These are based on estimates from the IPVAR
model that includes private investment. The model coefficients are conditioned only on fiscal position. Government debt as a percentage of GDP is the measure of
fiscal position and the values shown on the x-axis correspond to the 5th to 95th percentiles from the sample. Fiscal position is strong (weak) when government debt
is low (high). Solid lines represent the median, and dotted bands are the 16-84 percent confidence bands.
36
Figure A9: Fiscal Position-Dependent Multipliers – All Horizons
A. Fiscal Position Only B. Fiscal Position and Recessions
Note: The surf plots show the conditional fiscal multipliers for different levels of fiscal position and across all horizons. These are based on estimates from the
IPVAR model. The left panel is when model coefficients are only conditioned on the fiscal position, and in the right panel they are jointly conditioned on the fiscal
position and the phase of the business cycle. Government debt as a percentage of GDP is the measure of fiscal position and the values shown on the x-axis
correspond to the 5th to 95th percentiles from the sample. Recessions are determined by the Harding-Pagan (2002) business cycle dating algorithm. Fiscal position
is strong (weak) when government debt is low (high). Numbers shown are the median estimates of the multiplier
37