Output Spillovers from Fiscal Policy — Auerbach and Gorodnichenko, NBER Working Paper 18578 (November 2012)
Summary
NBER Working Paper 18578, dated November 2012, is titled Output Spillovers From Fiscal Policy and is by Alan J. Auerbach and Yuriy Gorodnichenko of the Department of Economics at the University of California, Berkeley. The abstract says the authors estimate the cross-country spillover effects of government purchases on output for a large number of OECD countries, letting the multipliers vary with the state of the economy and using real-time forecast data to purge policy innovations of their predictable components. The paper says fiscal spillovers are significant in both statistical and economic terms, are particularly high in recessions and quite modest in expansions, and are increased further when both the recipient and source countries are in recession.
Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used
- Issuer
- Source documents
- Document type
- Output Spillovers From Fiscal Policy
- Case
- Output Spillovers From Fiscal Policy
Cite this document
“Output Spillovers from Fiscal Policy — Auerbach and Gorodnichenko, NBER Working Paper 18578 (November 2012).”
Original at www.nber.org. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/output-spillovers-from-fiscal-policy/
, SHA-256 567e3a07578346884746f1e253ef3f5107e8bf6754924bd78a041986d5d56419.Full text
NBER WORKING PAPER SERIES
OUTPUT SPILLOVERS FROM FISCAL POLICY
Alan J. Auerbach
Yuriy Gorodnichenko
Working Paper 18578
http://www.nber.org/papers/w18578
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
November 2012
Alan Auerbach acknowledges the Robert D. Burch Center for research support, and has no material
outside financial relationships that may relate to this research. Yuriy Gorodnichenko acknowledges
the National Science Foundation for research support, and has no material outside financial relationships
that may relate to this research. This paper will be presented at the American Economic Association
meetings, San Diego, January 2013. We thank Olivier Coibion for comments on an earlier version
of the paper. We are grateful to Kyle Kennelly for research assistance. The views expressed herein
are those of the authors and do not necessarily reflect the views of the National Bureau of Economic
Research.
NBER working papers are circulated for discussion and comment purposes. They have not been peer-
reviewed or been subject to the review by the NBER Board of Directors that accompanies official
NBER publications.
© 2012 by Alan J. Auerbach and Yuriy Gorodnichenko. All rights reserved. Short sections of text,
not to exceed two paragraphs, may be quoted without explicit permission provided that full credit,
including © notice, is given to the source.
Output Spillovers from Fiscal Policy
Alan J. Auerbach and Yuriy Gorodnichenko
NBER Working Paper No. 18578
November 2012
JEL No. E32,E62
ABSTRACT
In this paper, we estimate the cross-country spillover effects of government purchases on output for
a large number of OECD countries. Following the methodology in Auerbach and Gorodnichenko
(2012a, b), we allow these multipliers to vary smoothly according to the state of the economy and
use real-time forecast data to purge policy innovations of their predictable components. We also consider
the responses of other key macroeconomic variables. Our findings suggest that cross-country spillovers
have an important impact, and also confirm those of our earlier papers that fiscal shocks have a larger
impact when the affected country is in recession.
Alan J. Auerbach
Department of Economics
530 Evans Hall, #3880
University of California, Berkeley
Berkeley, CA 94720-3880
and NBER
auerbach@econ.berkeley.edu
Yuriy Gorodnichenko
Department of Economics
508-1 Evans Hall #3880
University of California, Berkeley
Berkeley, CA 94720-3880
and NBER
ygorodni@econ.berkeley.edu
1. Introduction
One of the challenges facing a country attempting to maintain economic stability is the economic
shocks emanating from abroad, which through trade and other linkages may have important
effects on domestic conditions. While such shocks may have many sources, one of particular
interest is fiscal policy. Indeed, a common justification for fiscal agreements like the Stability
and Growth Pact and successive measures adopted by Eurozone countries is that, having
relinquished independent monetary and exchange rate policies, individual countries need some
protection from the shocks of uncoordinated fiscal policies. Economic observers long
appreciated the importance of fiscal spillovers but it is the current economic environment of ever
increasing globalization and of conflicting calls for fiscal austerity and fiscal stimuli that
demands clear and robust evidence to navigate policymakers through the Great Recession and its
aftermath. Specifically, there are at least three key questions: (1) What is the effect of fiscal
austerity/stimulus in one country on economic conditions in another country? (2) Can countries
short of fiscal ammunition (e.g., Greece during the Great Recession) be supported by positive
fiscal stimulus in other countries? (3) Does the strength of fiscal spillovers vary over the business
cycle? If so, what should be the scope of coordinated fiscal policies in recession? In this paper,
we try to shed new light on these questions, with results that have a number of immediate policy
implications.
Our paper is certainly not the first empirical investigation of fiscal spillovers. For
example, Bénassy-Quéré and Cimadomo (2006) estimate the impact of fiscal spillovers from
Germany to the remaining G-7 countries using an augmented SVAR model. Beetsma et al.
(2005) estimate the spillover impact of shocks via the trade channel by combining panel VAR
estimates of the effects of domestic shocks on output and estimates of the effects of output on
imports from abroad. In a recent paper, Hebous and Zimmermann (2012) adopt a panel VAR
approach to estimate the effects of domestic and foreign fiscal shocks among Eurozone
countries, representing foreign shocks as a common, GDP weighted aggregate of the individual
country shocks. 1
We extend the existing literature in a number of ways. First, we consider fiscal spillovers
among OECD countries, a larger and more heterogeneous group than the G-7 or the Eurozone.
This allows us to consider, for example, the differences in effects among countries sharing fixed
exchange rates versus those that do not. Second, although we allow shocks to depend on trade
linkages, we directly estimate the effects of shocks in one country on another country’s output.
This makes interpretation of estimated coefficients in our econometric specification particularly
straightforward and transparent. Third, we allow multipliers to vary across states of the business
cycle, thus relaxing standard assumptions that our previous analysis of domestic shocks in the
United States and the OECD (Auerbach and Gorodnichenko, 2012a, b, respectively) suggested
was important. Fourth, we enhance identification of fiscal shocks by removing predictable
innovations in government spending by controlling for information contained not only in the lags
of macroeconomic variables but also in professional forecasts. This aspect is important because
anticipated and unanticipated fiscal shocks may have different effects and mixing these two
types of shocks, which is likely to happen in VAR models (see e.g. Ramey 2011 and Auerbach
and Gorodnichenko 2012a, 2012b), can lead to understating estimated effects of fiscal shocks.
Finally, we consider the effects of fiscal spillovers not only on output but also on a number of
other macroeconomic variables; hence we can paint a more detailed picture of how fiscal shocks
propagate across countries.
1
There have also been a number of papers estimating the impact of fiscal spillovers using DSGE models. See, for
example, Bénassy-Quéré (2006) and Corsetti, Meier, and Müller, (2010).
2
We document that fiscal spillovers are significant in both statistical and economic terms.
The effect, however, varies tremendously over the business cycle with the spillovers being
particularly high in recessions and quite modest in expansions, with the output multiplier in
recessions being even larger than those found in our previous work for domestic shocks, based
on what would expect given the strength of trade linkages. We also find that fiscal spillovers are
increased further when both recipient and source countries are in recession.
2. Modeling Fiscal Spillovers
To model the effects of fiscal spillovers, we extend the approach taken in Auerbach and
Gorodnichenko (2012b) and use data for a panel of OECD countries to estimate fiscal spillover
multipliers using direct projections. Specifically, for our baseline model we run a series of
regressions for different horizons, h = 0, 1, …, H of the form:
𝑌𝑖,𝑡+ℎ −𝑌𝑖,𝑡−1 𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 Δ𝑌 Δ𝐺
(1) = 𝛼ℎ + ∑𝑚
𝑠=1 𝛽ℎ𝑠 𝑌
𝑖,𝑡−𝑠
+ ∑𝑚
𝑠=1 𝛿ℎ𝑠 𝑌
𝑖,𝑡−𝑠
+ 𝜙ℎ𝑖 + 𝜇ℎ𝑡 + 𝑒𝑟𝑟𝑜𝑟𝑖ℎ𝑡
𝑌𝑖,𝑡−1 𝑌𝑖,𝑡−1 𝑖,𝑡−𝑠−1 𝑖,𝑡−𝑠−1
where 𝑡 and 𝑖 index time and countries, Y is real GDP, and G is real government purchases (both
Y and G are measured in terms of local currency in fixed prices of the base year), 𝜙 and 𝜇 are
horizon-specific country and time fixed effects, and 𝐺𝑆ℎ𝑜𝑐𝑘 is the government spending
spillover shock emanating from other countries, which we will specify further. The impulse
response for 𝐻 periods is constructed from a sequence of estimated {𝛼ℎ }𝐻
ℎ=0 .
Before we continue, a few observations about equation (1) are useful. First, variables in
equation (1) are in differences, scaled by lagged GDP, which follows the approach in Hall (2009)
and Barro and Redlick (2011). Estimated {𝛼ℎ }𝐻
ℎ=0 in this specification directly correspond to
multipliers while in our previous work we estimated elasticities of output with respect to a
3
government spending shock and then converted them into multipliers. Also, scaling by lagged
output puts the observations from different dates and countries in similar units. 2 Second, we
omit lagged values of taxes from our specification. Experiments with the inclusion of taxes and
other potential own-country controls, such as rates of interest, inflation and unemployment, did
not have an appreciable impact on the results, and so we omit them in the interest of conserving
degrees of freedom. Third, we do not include own-country fiscal shocks in equation (1). While
the analysis of the effects of such shocks was the main objective of our previous papers,
including them here would require a significant reduction in sample size, for there are many
country-years for which we do not have OECD forecasts of government spending. Analysis of
models including own-country shocks using this reduced sample suggests that their inclusion has
little impact on the estimated effects of fiscal spillovers; given the larger standard errors caused
by this reduction in sample size, we omit own-country shocks from the results we report.
To construct the fiscal spillover shock 𝐺𝑆ℎ𝑜𝑐𝑘, we regress real-time one-period-ahead
forecast errors for government spending from the OECD’s “Outlook and Projections Database”
in each country on that country’s lagged macroeconomic variables (output, government
spending, exchange rate, inflation, investment, and imports) as well as a set of country and
period fixed effects. Since the residual from this regression captures innovations in government
spending orthogonal to professional forecasts and lags of macroeconomic variables, we take this
residual as a measure of unanticipated government spending shocks. Denote this policy shock in
source country q with eq,t. Then we aggregate eq,t across countries using bilateral trade as a
2
In principle, if G and Y are cointegrated, their ratio is stable in the long run and a regression based on first
differences should include an error correction term reflecting this relationship in levels. However, our estimates are
for a relatively short time period (1985-2012), during which there are trends and swings in the G/Y ratio. Moreover,
specifications with included levels of G and Y (or their ratio) barely affected the coefficients of interest and therefore
are not reported separately here.
4
measure of inter-country linkages. In particular, our base specification for the fiscal shock
affecting country i in year t is:
𝑴𝒊𝒒,𝑩
∑𝒒≠𝒊 ×𝒆𝒒,𝒕 ×𝑮𝒒,𝒕−𝟏 ×𝑬𝒒,𝑩
𝑮𝒒,𝑩
(2) 𝑮𝑺𝒉𝒐𝒄𝒌𝒊,𝒕 = 𝑬𝒊,𝑩
where 𝑀𝑖𝑞,𝑡 is country q’s imports from country i in year t, Ej,t is country j’s US dollar exchange
rate in year t, and B is a base year. The term in curly brackets in expression (2), 𝑒𝑞,𝑡 × 𝐺𝑞,𝑡−1 ×
𝐸𝑞,𝐵 , equals the dollar value of country q’s fiscal shock, calculated using a base-year exchange
rate. The first term in the numerator, 𝑀𝑖𝑞,𝐵 /𝐺𝑞,𝐵 , scales this shock by the ratio of imports from
country i to government purchases, and division by the base-year dollar exchange rate of country
i converts the shock into units of the recipient-country’s currency. One may interpret 𝑀𝑖𝑞,𝐵 /𝐺𝑞,𝐵
as a weight which corrects for heterogeneity of countries in the strength of the trade linkage
between source country q and recipient country i and in the size of government in source country
q. This ratio also captures the idea that a certain factor of government purchases translates
(directly or indirectly) into imports from other countries, which stimulate demand in those
countries.
It is possible that a dollar increase in government spending in country q is going to be
converted into less than 𝑀𝑖𝑞,𝐵 /𝐺𝑞,𝐵 dollars of imports from country i. Indeed, discretionary
government spending shocks are often designed to support the domestic economy. For example,
the 2008-2009 fiscal stimulus in the United States had many restrictions, such as that firms
receiving federal aid had to hire U.S. citizens or purchase inputs from U.S. suppliers. While
examples that go in the other direction are less obvious (military spending abroad would be one
instance), indirect effects matter, too; the propensity to import out of induced changes in private
5
spending in country q must also be taken into account. Our specification in (2) effectively
𝑀𝑖𝑞,𝐵
assumes that spillover shocks occur through imports, and that 𝜃 is imported from country i
𝐺𝑞,𝐵
for each dollar increase in government spending in country q, with 𝜃 being constant across
countries. Obviously, 𝜃 = 1 is an important special case. However, even when 𝜃 ≠ 1, the
estimated 𝛼ℎ in specification (1) will absorb 𝜃 and our interpretation of the estimated 𝛼ℎ is not
affected. The higher the value of θ, the higher we would expect the estimate of 𝛼ℎ to be.
One can re-write the expression in (2) as follows
𝑀 𝐸𝑞,𝐵 𝑒𝑞,𝑡 ×𝐺𝑞,𝑡−1 𝑒𝑞,𝑡 ×𝐺𝑞,𝑡−1
(2’) 𝐺𝑆ℎ𝑜𝑐𝑘𝑖,𝑡 = 𝑀𝑖𝑇,𝐵 ∑𝑞≠𝑖 𝑀𝑖𝑞,𝐵 𝐸 × = 𝑀𝑖𝑇,𝐵 ∑𝑞≠𝑖 𝜔𝑖𝑞,𝐵 ×
𝑖𝑇,𝐵 𝑖,𝐵 𝐺𝑞,𝐵 𝐺𝑞,𝐵
where 𝑀𝑖𝑇,𝐵 is the total imports from country i to other countries measured in the currency of
country i (i.e., country i’s total exports to those countries from which fiscal shocks emanate) and
𝜔𝑖𝑞,𝐵 is the share of these imports from country i going to country q. The last term (curly
brackets) can be interpreted as an unanticipated shock to government spending measured in
percent terms. Thus, 𝐺𝑆ℎ𝑜𝑐𝑘𝑖,𝑡 is the weighted average percent shock in government spending
in other countries scaled by the level of imports from country i.
Equation (2) uses fixed, base-year exchange rates and import intensity for scaling fiscal
shocks. An alternative is to allow these factors to change over time reflecting changing trade
patterns and exchange rates. With time-varying weights, we specify country i’s fiscal shock in
year t as:
𝑴𝒊𝒒,𝒕−𝟏 𝑷𝒒,𝒕−𝟏
∑𝒒≠𝒊 ×𝒆𝒒,𝒕 ×𝑮𝒒,𝒕−𝟏 ×𝑬𝒒,𝒕−𝟏 ×
𝑮𝒒,𝒕−𝟏 𝑷𝑼𝑺,𝒕−𝟏
(3) 𝑮𝑺𝒉𝒐𝒄𝒌∗𝒊,𝒕 = 𝑷𝒊,𝒕−𝟏
𝑬𝒊,𝒕−𝟏 ×
𝑷𝑼𝑺,𝒕−𝟏
6
𝑃
where 𝑃𝑗,𝑡−1 is country j’s price index in year 𝑡 − 1 and hence 𝐸𝑗,𝑡−1 × 𝑃 𝑗,𝑡−1 is country j’s
𝑈𝑆,𝑡−1
real US dollar exchange rate in year 𝑡 − 1. An advantage of using time varying weights is that
they will pick up trends in exchange rates and trade linkages and hence more accurately
represent current economic conditions. A disadvantage is that imports, government purchases
and exchange rates are potentially correlated with the shocks we measure (to the extent that we
have not fully purged the shocks of predictable components) and also can exhibit volatility over
short periods. To deal with the latter issue, we smooth the time-varying weights in equation (3)
by averaging values over two lagged periods.
In addition to the standard real GDP series, we will examine responses of other key
macroeconomic variables to government spending spillover shocks, following the same general
approach as in Auerbach and Gorodnichenko (2012b). As discussed there, a benefit of
estimation using direct projections is that consideration of these other variables involves only a
small modification to equation (1). For example, equation (1) for a variable of interest 𝑋 is
specified as follows:
𝑋𝑖,𝑡+ℎ −𝑋𝑖,𝑡−1 𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 Δ𝑌 Δ𝑋
(1’) = 𝛼ℎ + ∑𝑚
𝑠=1 𝛽ℎ𝑠 𝑌
𝑖,𝑡−𝑠
+ ∑𝑚
𝑠=1 𝛿ℎ𝑠 𝑋
𝑖,𝑡−𝑠
+ 𝜙ℎ𝑖 + 𝜇ℎ𝑡 + 𝑒𝑟𝑟𝑜𝑟𝑖ℎ𝑡 ,
𝑋𝑖,𝑡−1 𝑋𝑖,𝑡−1 𝑖,𝑡−𝑠−1 𝑖,𝑡−𝑠−1
𝑋𝑖,𝑡+ℎ −𝑋𝑖,𝑡−1 𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 Δ𝑌 Δ𝑋
(1’’) = 𝛼ℎ + ∑𝑚
𝑠=1 𝛽ℎ𝑠 𝑌
𝑖,𝑡−𝑠
+ ∑𝑚
𝑠=1 𝛿ℎ𝑠 𝑋
𝑖,𝑡−𝑠
+ 𝜙ℎ𝑖 + 𝜇ℎ𝑡 + 𝑒𝑟𝑟𝑜𝑟𝑖ℎ𝑡 .
𝑋𝑖,𝑡−1 𝑌𝑖,𝑡−1 𝑖,𝑡−𝑠−1 𝑖,𝑡−𝑠−1
7
𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡
In some instances, we prefer using equation (1’’) with rather than because 𝑋
𝑌𝑖,𝑡−1 𝑋𝑖,𝑡−1
could be measured in units or scale different from 𝐺𝑆ℎ𝑜𝑐𝑘 (e.g., real exchange rate, hours per
worker). 3
Whichever version of the fiscal shock we use, we also wish to allow the impact of shocks
to vary over the business cycle, given the findings in our earlier work that multipliers vary over
the cycle and are larger in recessions. Following our earlier approach of allowing smooth
transitions between the states we label “recession” and “expansion,” we modify specification (1)
as follows:
𝑌𝑖,𝑡+ℎ −𝑌𝑖,𝑡−1 𝐺𝑆ℎ𝑜𝑐𝑘𝑖,𝑡 𝐺𝑆ℎ𝑜𝑐𝑘𝑖,𝑡
(4) = 𝛼𝑅,ℎ 𝐹𝑧𝑖,𝑡−1 + 𝛼𝐸,ℎ 1 − 𝐹𝑧𝑖,𝑡−1
𝑌𝑖,𝑡−1 𝑌𝑖,𝑡−1 𝑌𝑖,𝑡−1
𝑚 Δ𝑌𝑖,𝑡−𝑠 𝑚 Δ𝑌
+ ∑𝑠=1 𝛽𝑅,ℎ𝑠 𝐹𝑧𝑖,𝑡−1 𝑌 + ∑𝑠=1 𝛽𝐸,ℎ𝑠 1 − 𝐹𝑧𝑖,𝑡−1 𝑌 𝑖,𝑡−𝑠
𝑖,𝑡−𝑠−1 𝑖,𝑡−𝑠−1
Δ𝐺𝑖,𝑡−𝑠 Δ𝐺𝑖,𝑡−𝑠
+ ∑𝑚
𝑠=1 𝛿𝑅,ℎ𝑠 𝐹𝑧𝑖,𝑡−1 𝑌 + ∑𝑚
𝑠=1 𝛿𝐸,ℎ𝑠 1 − 𝐹𝑧𝑖,𝑡−1 𝑌
𝑖,𝑡−𝑠−1 𝑖,𝑡−𝑠−1
+𝜙ℎ𝑖 + 𝜇ℎ𝑡 + 𝑒𝑟𝑟𝑜𝑟𝑖ℎ𝑡
where 𝐹𝑧𝑖,𝑡 can be interpreted as a measure of probability of being in a recession in country i
at time t based on a measure of the state of the business cycle, 𝑧𝑖,𝑡 . The impulse response for the
𝐻 𝐻
multiplier in recession is given by 𝛼𝑅,ℎ ℎ=0, while for expansion it is 𝛼𝐸,ℎ ℎ=0 . We construct
exp−𝛾𝑧
𝐹𝑧𝑖,𝑡 = 1+exp−𝛾𝑧𝑖,𝑡 and, as before, normalize 𝑧𝑖,𝑡 to have zero mean and unit variance and
𝑖,𝑡
fix 𝛾 = 1.5 so that an economy spends about 20 percent of the time in recession. We calculate
𝑧𝑖,𝑡 after removing very low frequency movements in the data using the Hodrick-Prescott filter
(smoothing parameter 𝜆 = 10,000), so one can think of 𝑧 as (normalized) deviations from trend.
For our base case, as before, we measure 𝑧𝑖,𝑡 using the growth rate of real GDP, but we also
3
We find similar results when in equations (1’) and (1’’) we control for lags of the government spending growth
rate rather than lags of the output growth rate. We prefer the latter specification as it has a better fit and produces
smaller standard errors.
8
consider the logarithm of real GDP and the unemployment rate. When we use the latter two
variables, one can use the slump/boom dichotomy rather than the recession/expansion
dichotomy. For each variation, we lag 𝑧𝑖,𝑡 by one period to minimize contemporaneous
correlations between fiscal shocks and macroeconomic variables.
3. Data
The macroeconomic series we use in our analyses come from the OECD’s Statistics and
Projections database. Aside from the benefits these data provide in terms of standardized
variable definitions and measurement, we also have available semiannual forecasts prepared by
the OECD for key macroeconomic variables such as GDP and government spending in June and
December of each year. 4 For further information on these forecasts and their quality, see
Auerbach and Gorodnichenko (2012b) or the OECD’s website. 5
The OECD’s forecasts are consistently available since 1984 for “old” members of the
OECD (e.g., the United States) and since the mid-1990s for newer members (e.g., Poland). A
drawback of using the OECD projections is that, for most of the available sample, they are
available only semiannually rather than quarterly. Thus, we estimate our equations at this
frequency. To avoid a further reduction in sample size, as discussed above, we omit own-
country fiscal shocks from our specification, in order to include observations for countries in
years for which their own forecasts of government purchases are unavailable. Also, we treat
countries for which forecasts are unavailable as having zero weight in the fiscal shock
calculations for other countries. This undoubtedly introduces some error to our key variable, but
4
Consistent with the OECD definitions and the previous literature on fiscal multipliers, our government spending
series is the sum of real public consumption expenditure and real government gross capital formation. That is, it
excludes imputed rent on the government capital stock, unlike under the current U.S. NIPA convention.
5
http://www.oecd.org/faq/0,3433,en_2649_33733_1798284_1_1_1_1,00.html
9
we have complete data for the large leading economies that account for a considerable share of
world imports. Finally, we exclude from our analysis all observations from a few small
economies—Greece, Estonia, Luxembourg, and Turkey—for which there are large and volatile
changes in reported government spending, in at least some cases due to data revisions. In the
end, we have 30 countries in our sample.
For all model specifications presented below, we estimate impulse responses for six
semiannual periods, starting in the first half of 1985 (because our projections of government
spending are available beginning in 1984). That is, we set the maximum horizon H = 5 and
estimate equations for 0 to 5 periods ahead. Also, we set m, the number of lags of changes in
real GDP and government purchases included as control variables, equal to 4. All estimates are
reported along with Driscoll and Kraay (1998) standard errors that allow arbitrary correlations of
the errors across countries and time.
Figure 1 shows time series of fiscal spillover shocks for selected countries and Table 1
reports basic descriptive statistics. The average (across countries) standard deviation of scaled
𝐺𝑆ℎ𝑜𝑐𝑘𝑖,𝑡
fiscal spillover shocks × 100 is about 0.09 but there is considerable variation across
𝑌𝑖,𝑡−1
countries from 0.2 for Canada to 0.025 for the USA. The magnitude of these shocks is fairly
small for big economies and, given plausible multipliers, they are thus unlikely to explain a large
fraction of variation in output and other macroeconomic variables. The correlation of shock
series across countries varies between -0.4 to 0.99 with the mean correlation of approximately
0.4. In general, countries that have very different trading partners tend to have low or negative
correlation (e.g., Canada and the Czech Republic), while countries sharing the same key trading
partners (e.g., Hungary and Slovakia or Canada and Mexico) have highly correlated series. The
10
average 𝑀𝑖𝑞,𝐵 /𝐺𝑞,𝐵 across countries and trading pairs is 1.6 but again there is dramatic variation
across countries from 3.6 in Slovakia to 0.6 in Japan and 0.66 in the United States.
4. Results
A. Basic Results
Table 2 presents results for a variety of specifications based on the approach outlined above.
1
Each entry in the table provides the average (6 ∑5ℎ=0 𝛼ℎ ) real GDP multiplier of fiscal spillovers
over the six-period (i.e., three-year) horizon window, with standard errors in parentheses.
To help in the interpretation of these coefficients, recall that the fiscal shocks are measured in
units of real government spending, scaled by the ratio of bilateral imports to government
spending in the source country. Thus, if the percent shock to government spending, 𝑒𝑞𝐺 , were
uniform across all source countries q, this shock would be scaled by the sum of imports from
country i by all source countries (e.g. 𝑀𝑖𝑇,𝐵 in (2’)) before being included in the regression as the
variable 𝐺𝑆ℎ𝑜𝑐𝑘. In contrast, when using a framework similar to specification (1), the literature
estimating domestic government spending multipliers for country i scales percent shock 𝑒𝑖𝐺 by
the level of government spending 𝐺𝑖,𝑡−1 . Thus, if the impact of spillover shocks is in proportion
to the ratio of imports from country i to government spending—if 𝜃 = 1 in our previous
terminology—we should expect the estimated multipliers to be of the same magnitude as those
estimated for domestic shocks. Otherwise, as discussed, the estimated spillover multipliers will
be larger (if 𝜃 > 1) or smaller (if 𝜃 < 1).
The first column of Table 2 provides estimates for the linear model, given in (1). The
upper panel corresponds to the fiscal shocks in expression (3); those in the lower panel are based
on the fiscal shocks as specified in expression (2). For each of these specifications, we present
results for three samples: i) our full sample (the base case), ii) our full sample but with fiscal
11
spillover shocks constructed using only eight relatively large economies for which we have
complete data since 1984: the US, UK, France, Italy, Japan, Germany, Canada, and Australia –
the G-7 plus Australia – and our full sample of countries, but truncated at the end of 2007, to
eliminate the possibly unique effects occurring during the Great Recession. The multipliers for
all six variants are similar, given their standard errors, and are statistically significant when the
sample is constrained to exclude post-2007 observations.
The remaining columns of Table 2 show estimates of state-dependent multipliers, based
on economic conditions in the recipient country, i, from expression (4), using our three different
measures to represent the state of the business cycle, based on output growth, log output level,
and the unemployment rate. Results are again relatively similar across specifications within any
column, but are strikingly different across business cycle regimes, with multipliers being much
larger in recession, and much smaller in expansion, than for the linear model. Indeed, the
multipliers in expansion are generally negative, although not significantly so, while those in
recession are considerably larger than those estimated using the linear model. We obtain values
in recession of a higher order of magnitude than those found in our earlier work for the United
States (Auerbach and Gorodnichenko 2012a), which suggests that fiscal spillovers have a greater
impact than would be implied simply by the ratio of imports to government spending. Note that
because shocks are fairly correlated across countries by construction, and because we control for
many lags of macroeconomic variables as well as country and time fixed effects, the standard
errors are relatively large. However, even with this considerable sampling uncertainty,
multipliers in recessions are routinely statistically significant at the conventional levels.
Figure 2 presents impulse responses over the six-period horizon, with 90 percent
confidence bands, reflecting the estimates in the lower panel and first three columns of Table 2,
12
which correspond to fixed-weight fiscal shocks for the linear model and states based on lagged
output growth rates. Aside from overall differences in magnitude, the impulse responses also
differ between recession and expansion in their shape, with the effects in recession growing over
time, unlike those in expansion.
Given the similarity of results for different specifications, for the remaining tables we
will report results only for this particular specification of fiscal shocks, GShock (using fixed
weights over time, as in equation (2)) and the business cycle index z based on lagged values of
real GDP growth.
Table 3 provides multiplier estimates for a variety of other important macroeconomic
aggregates. For example, fiscal spillover shocks have a strong, stimulating effect on exports in
recessions, while in expansions the effect is negative. In contrast, imports have only weak,
statistically insignificant responses to fiscal spillover shocks. These responses are consistent
with the notion that trade can be an important channel of how fiscal shocks are propagated across
countries. In general, the results confirm those in Auerbach and Gorodnichenko (2012b) where
we estimate the effects of own-country fiscal shocks within the OECD, especially when one
constrains the sample to large countries or pre-2008 observations. All components of output rise
more in recession, as does employment.
B. Exchange rate regimes
As discussed earlier, the issue of fiscal shock transmission within currency areas is of particular
interest, given the inability of countries to offset these shocks via monetary and exchange rate
policy. More generally, we would expect the transmission of shocks to differ between fixed and
floating exchange rate regimes, ceteris paribus, because changes in a source country’s demand
for imports would not have the same effects on its exchange rates vis à vis its trading partners.
13
In particular, we would expect smaller effects of shocks under a floating exchange rate regime,
as an increased demand for imports resulting from a positive shock would cause an appreciation
of the recipient country’s bilateral exchange rate.
We modify our previous methodology by allowing the effects of shocks to vary
according to the recipient country’s bilateral exchange rate regime with the source country. That
is, we aggregate fiscal shocks into two components, those for countries with a fixed exchange
rate with the recipient country, and those for countries with a floating exchange rate with the
recipient country. 6 Specifically, for each combination of countries 𝑖 and 𝑞, we create dummy
variable 𝐵𝐼𝐹𝐼𝑋𝑖𝑞 equal to one if countries 𝑖 and 𝑞 have a fixed exchange rate (directly or
indirectly) and zero otherwise. Then measures of fiscal spillover shocks for countries sharing a
fixed exchange rate and countries with floating exchange rates are
𝑀𝑖𝑞,𝐵
∑𝑞≠𝑖 𝐺
× 𝐵𝐼𝐹𝐼𝑋𝑖𝑞 ×𝑒𝑞𝑡 ×𝐺𝑞,𝑡−1 ×𝐸𝑞𝐵
𝐺𝑞,𝐵
𝐹𝑖𝑥𝑒𝑑
𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 = 𝐸𝑖𝐵
𝑀𝑖𝑞,𝐵
∑𝑞≠𝑖 𝐺
×1−𝐵𝐼𝐹𝐼𝑋𝑖𝑞 ×𝑒𝑞𝑡 ×𝐺𝑞,𝑡−1 ×𝐸𝑞𝐵
𝐺𝑞,𝐵
𝐹𝑙𝑜𝑎𝑡
𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 = 𝐸𝑖𝐵
𝐹𝑖𝑥𝑒𝑑 𝐹𝑙𝑜𝑎𝑡
with 𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 + 𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 = 𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 and the linear specification (1) modified as
follows
𝐹𝑖𝑥𝑒𝑑 𝐹𝑙𝑜𝑎𝑡
𝑌𝑖,𝑡+ℎ −𝑌𝑖,𝑡−1 𝐺𝑆ℎ𝑜𝑐𝑘𝑗𝑡 𝐺𝑆ℎ𝑜𝑐𝑘𝑗𝑡
(5) = 𝛼ℎ,𝐹𝑖𝑥𝑒𝑑 + 𝛼ℎ,𝐹𝑙𝑜𝑎𝑡
𝑌𝑖,𝑡−1 𝑌𝑖,𝑡−1 𝑌𝑖,𝑡−1
Δ𝑌 Δ𝐺
+ ∑𝑚
𝑠=1 𝛽ℎ𝑠 𝑌
𝑖,𝑡−𝑠
+ ∑𝑚
𝑠=1 𝛿ℎ𝑠 𝑌
𝑖,𝑡−𝑠
+ 𝜙ℎ𝑖 + 𝜆ℎ𝑡 + 𝑒𝑟𝑟𝑜𝑟𝑖ℎ𝑡 .
𝑖,𝑡−𝑠−1 𝑖,𝑡−𝑠−1
The state-dependent version (4) is modified in a similar spirit.
6
Our categorization of fixed and floating exchange rates is based on Rose (2011). We classify a pair of countries A
and B as sharing a fixed exchange rate if one of the countries has a fixed exchange rate with another directly (e.g. A
pegs to B) or indirectly (i.e., A pegs to C and C pegs to B or B pegs to C).
14
Using this modification, we find (Table 4) that while multipliers are broadly similar, the
magnitudes of multipliers in countries sharing a floating exchange rate tend to be larger than
multipliers in countries sharing a fixed exchange rate, although the differences are not
statistically significant. This pattern emerges in both linear and state-dependent models. While
these results are somewhat surprising, one should keep in mind that the samples of countries with
fixed and floating exchange rate regimes vary in a number of other ways which may affect our
estimates. For example, countries having fixed exchange rates are almost exclusively covered by
the Eurozone.
C. Cyclical properties by source country
Throughout our discussion thus far, we have allowed impulse responses to differ according to the
state of the recipient country’s business cycle. This makes sense, as we might expect the effects
of a given external shock to vary according to the extent of slack in the recipient economy. But
it is also possible that the relationship between the external shock and the source country’s fiscal
shock might depend on that country’s economic conditions. For example, if a source country is
in recession, a positive fiscal shock there might have a bigger local impact on output and on
import demand, and therefore provide a bigger stimulus to recipient country production.
According to this logic, multipliers should be bigger if there is a recession in the source country
as well as if there is one in the recipient country. Whether this prediction is borne out by the data
has immediate policy implications in the current economic environment where economic activity
is depressed in many countries. Specifically, if a fiscal stimulus in one depressed economy has a
more positive effect on another depressed economy, then amplified fiscal spillovers would
increase the argument in favor of coordinated fiscal stimulus so that externalities from fiscal
shocks are internalized.
15
We calculate fiscal shocks from sources countries in recession and in expansion as
follows
𝑀𝑖𝑞,𝐵
∑𝑞≠𝑖 𝐺
× 𝐹𝑧𝑞𝑡 ×𝑒𝑞𝑡 ×𝐺𝑞,𝑡−1 ×𝐸𝑞𝐵
𝐺𝑞,𝐵
𝑅
𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 = 𝐸𝑖𝐵
𝑀𝑖𝑞,𝐵
∑𝑞≠𝑖 𝐺
×1−𝐹𝑧𝑞𝑡 ×𝑒𝑞𝑡 ×𝐺𝑞,𝑡−1 ×𝐸𝑞𝐵
𝐺𝑞,𝐵
𝐸
𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 = 𝐸𝑖𝐵
where, as in equation (4), 𝐹𝑧𝑞𝑡 is a measure of probability of country q being in recession and,
𝑅 𝐸
by construction, 𝐺𝑆ℎ𝑜𝑐𝑘𝑗𝑡 + 𝐺𝑆ℎ𝑜𝑐𝑘𝑗𝑡 = 𝐺𝑆ℎ𝑜𝑐𝑘𝑗𝑡 . which is our original measure of fiscal
shocks. The linear model is then
𝑅 𝐸
𝑌𝑖,𝑡+ℎ −𝑌𝑖,𝑡−1 Δ𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡 Δ𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡
(6) = 𝛼ℎ,𝑅 + 𝛼ℎ,𝐸
𝑌𝑖,𝑡−1 𝑌𝑖,𝑡−1 𝑌𝑖,𝑡−1
Δ𝑌 Δ𝐺
+ ∑𝑚
𝑠=1 𝛽ℎ𝑠 𝑌
𝑖,𝑡−𝑠
+ ∑𝑚
𝑠=1 𝛿ℎ𝑠 𝑌
𝑖,𝑡−𝑠
+ 𝜙ℎ𝑖 + 𝜆ℎ𝑡 + 𝑒𝑟𝑟𝑜𝑟𝑖ℎ𝑡
𝑖,𝑡−𝑠−1 𝑖,𝑡−𝑠−1
and the state-dependent version (4) is modified similarly.
Table 5 presents estimates based on expression (6), with a second dimension of
distinction between recession and expansion based on the state of the business cycle (measured
using the same method) in source countries. Each pair of columns corresponds to one state of
the business cycle in the source country (linear; expansion; and recession). For each pair, our
previous results hold: multipliers are larger when the recipient country is in recession than when
the recipient country is in expansion. Looking across these sets of columns, we see that when
the recipient country is in recession, (columns 4 and 6 in the table), the multipliers are generally
larger when the source country is also in recession, a result consistent with the reasoning laid out
above.
16
5. Conclusions
In an increasingly globalized world, policies adopted in one country are likely to affect economic
outcomes in other countries. To what extent, if at all, fiscal policies spill over into other
countries is a key question in the current environment with depressed economies and high or
rising levels of public debt in many developed countries. We document that fiscal stimulus in
one country is likely to have economically and statistically significant effects on output in other
countries. Furthermore, the strength of the spillover varies with the state of the economy in the
recipient and source countries, with the output multipliers being large in recessions. These
results suggest that fiscal activism may indeed be effective in stimulating demand in economic
downturns and that coordination of fiscal policies may be more valuable than previously thought.
The present paper estimates fiscal spillovers based on the historical experience of OECD
economies since the mid-1980s and thus it may be difficult to generalize these estimates to
different episodes or countries. However, one may reasonably argue that future theoretical and
empirical models should allow for non-linear and potentially strong positive responses of
economies to domestic and foreign fiscal shocks. This approach is likely to provide a solid,
empirically plausible foundation for designing fiscal policies.
17
References
Auerbach, Alan, and Yuriy Gorodnichenko, 2012a. “Measuring the Output Responses to Fiscal
Policy,” American Economic Journal: Economic Policy 4(2), May, 1-27.
Auerbach, Alan, and Yuriy Gorodnichenko, 2012b. “Fiscal Multipliers in Recession and
Expansion,” in A. Alesina and F. Giavazzi, eds. Fiscal Policy after the Financial Crisis,
Chicago: University of Chicago Press, forthcoming.
Barro, Robert, and Charles Redlick, 2011. “Macroeconomic Effects of Government Purchases
and Taxes,” Quarterly Journal of Economics 126(1), February, 51-102.
Beetsma, Roel, Massimo Giuliodori and Franc Klaassen, 2005. “Trade Spillovers of Fiscal
Policy in the European Union: A Panel Analysis,” De Nederlandsche Bank Working
Paper No. 052/2005, August.
Bénassy-Quéré, Agnès, 2006. “Short-Term Fiscal Spillovers in a Monetary Union,” CEPII
Working Paper No. 2006-13, July.
Bénassy-Quéré, Agnès, and Jacopo Cimadomo, 2006. “Changing Patterns of Domestic and
Cross-Border Fiscal Policy Multipliers in Europe and the US,” CEPII Working Paper No.
2006-24, December.
Corsetti, Giancarlo, André Meier, and Gernot J. Müller, 2010. “Cross-Border Spillovers from
Fiscal Stimulus,” International Journal of Central Banking 6(1), March, 5-37.
Driscoll, John C., Aart C. Kraay, 1998. “Consistent Covariance Matrix Estimation With Spatially
Dependent Panel Data,” Review of Economics and Statistics 80(4), 549-560.
Hall, Robert E., 2009. “By How Much Does GDP Rise if the Government Buys More Output?
Brookings Papers on Economic Activity 2009(2), 183-231.
Hebous, Shafik, and Tom Zimmermann, 2012. “Estimating the Effects of Coordinated Fiscal
Actions in the Euro Area,” CESIFO Working Paper No. 3912, August.
Ramey, Valerie, 2011. “Identifying Government Spending Shocks: It's All in the Timing,”
Quarterly Journal of Economics 126(1), 1-50.
Rose, Andrew, 2011. “Exchange Rate Regimes in the Modern Era: Fixed, Floating, and Flaky,"
Journal of Economic Literature 49(3), 652–472.
18
Table 1. Descriptive statistics
Trade linkage to 𝐺𝑆ℎ𝑜𝑐𝑘𝑖𝑡
Standard deviation of spillover fiscal shocks × 100
𝑌𝑖,𝑡−1
government
spending Time-varying weights Fixed weights
𝑀𝑖𝑞,𝐵 /𝐺𝑞,𝐵
Country Only old/large Only old/large
Constrain Constrain
OECD OECD
the the
Mean St. Dev. base economies in base economies in
sample to sample to
construction of construction of
pre-2008 pre-2008
spillover shocks spillover shocks
(1) (2) (3) (4) (5) (6) (7) (8)
AUS 0.85 0.32 0.056 0.048 0.056 0.062 0.053 0.061
AUT 1.96 0.54 0.089 0.086 0.091 0.073 0.071 0.076
BEL 2.58 0.66 0.141 0.111 0.146 0.155 0.144 0.160
CAN 1.30 0.41 0.204 0.206 0.209 0.162 0.162 0.162
CHE 2.84 0.65 0.068 0.064 0.068 0.058 0.052 0.059
CHL 2.72 0.59 0.058 0.040 0.054 0.053 0.036 0.051
CZE 2.67 0.75 0.152 0.142 0.170 0.174 0.161 0.195
DEU 1.36 0.51 0.051 0.039 0.045 0.045 0.036 0.042
DNK 1.25 0.36 0.059 0.057 0.060 0.059 0.058 0.059
ESP 1.17 0.43 0.032 0.029 0.031 0.021 0.020 0.021
FIN 1.18 0.41 0.059 0.053 0.060 0.059 0.055 0.061
FRA 0.84 0.24 0.042 0.039 0.043 0.037 0.034 0.038
GBR 0.99 0.34 0.036 0.030 0.036 0.028 0.022 0.027
HUN 2.79 1.07 0.141 0.138 0.150 0.071 0.066 0.076
IRL 2.76 1.30 0.190 0.182 0.198 0.149 0.153 0.155
ISL 1.40 0.26 0.062 0.053 0.067 0.054 0.053 0.056
ISR 1.56 0.18 0.073 0.071 0.075 0.046 0.043 0.046
ITA 1.00 0.29 0.039 0.036 0.040 0.031 0.029 0.032
JPN 0.60 0.10 0.032 0.029 0.032 0.026 0.024 0.025
KOR 1.77 0.77 0.084 0.081 0.087 0.059 0.057 0.061
MEX 1.53 0.95 0.128 0.132 0.127 0.089 0.088 0.089
NLD 1.88 0.53 0.126 0.123 0.129 0.114 0.112 0.119
NOR 1.23 0.18 0.084 0.085 0.084 0.064 0.064 0.066
NZL 1.17 0.29 0.068 0.062 0.069 0.061 0.057 0.062
POL 1.95 0.46 0.056 0.058 0.054 0.032 0.031 0.033
PRT 1.44 0.36 0.060 0.058 0.062 0.035 0.034 0.036
SVK 3.58 1.05 0.171 0.132 0.159 0.204 0.162 0.202
SVN 3.09 0.54 0.098 0.099 0.097 0.114 0.105 0.119
SWE 1.13 0.39 0.050 0.044 0.047 0.058 0.055 0.054
USA 0.66 0.28 0.025 0.014 0.025 0.023 0.013 0.023
Total 1.60 0.93 0.094 0.089 0.095 0.085 0.080 0.087
Notes: columns (1) and (2) show time-series standard deviation and mean of import to
government spending ratio by country. Columns (3) through (8) show the time-series standard
deviation of fiscal spillover shocks.
Table 2. Average output multipliers over 3 years
State-dependent multipliers where state measured as deviation from trend
Shock series Linear Output growth rate Output Unemployment rate
Expansion Recession Expansion Recession Expansion Recession
(1) (2) (3) (4) (5) (6) (7)
Time-varying weights
Base 1.34 -0.46 3.38 -0.49 3.16** -1.14 3.45***
(0.91) (1.32) (2.30) (0.98) (1.63) (1.15) (1.35)
Only old/large OECD economies in 1.25 -1.37 4.41 -0.60 2.97 -1.68 3.48**
construction of spillover shocks (0.97) (1.92) (2.77) (1.29) (1.87) (1.15) (1.55)
Constrain the sample to pre-2008 1.63** -0.92 4.64* -0.58 3.51** -0.94 3.20***
(0.85) (1.44) (2.69) (1.06) (1.59) (1.33) (1.16)
Fixed weights
Base 1.60 -1.10 4.63* 0.33 3.30** -1.50 4.06***
(1.00) (1.59) (2.54) (1.45) (1.63) (1.53) (1.05)
Only old/large OECD economies in 1.96* -2.56 6.72*** 0.74 3.19* -0.76 3.90***
construction of spillover shocks (1.16) (2.13) (2.72) (1.64) (1.84) (1.46) (1.26)
Constrain the sample to pre-2008 2.05** -0.93 5.36** 1.19 3.27* -0.95 3.94***
(1.00) (1.63) (2.73) (1.68) (1.71) (2.05) (1.05)
Notes: the table reports average output multipliers from fiscal spillovers. The estimated specifications are given by equation (1) in
column (1) and equation (4) in columns (2) through (7). Driscoll and Kraay (1998) standard errors are in parentheses. ***, **, *
denote significance at 0.01, 0.05, and 0.10 levels.
Table 3. Additional macroeconomic variables: average multipliers over 3 years
Only old/large OECD economies in
base Constrain the sample to pre-2008
construction of spillover shocks
Linear Recession Expansion Linear Recession Expansion Linear Recession Expansion
(1) (2) (3) (4) (5) (6) (7) (8) (9)
Consumption 0.58 1.95 -0.77 0.24 3.13*** -2.44*** 0.47 2.24 -1.13
(0.47) (1.30) (1.12) (0.54) (1.29) (1.05) (0.53) (1.62) (1.07)
Investment 0.39 0.78 0.25 0.41 1.49 -0.57 0.46 1.54 -0.49
(0.64) (1.42) (1.09) (0.63) (1.47) (1.08) (0.61) (1.55) (1.10)
Export 0.26 3.33*** -2.41* 0.50 3.37*** -2.79* 0.51 4.01*** -2.77**
(0.56) (1.39) (1.32) (0.62) (1.35) (1.51) (0.56) (1.43) (1.42)
Import 0.67 1.41 2.04 0.86 1.84 0.10 0.75 2.07 1.34
(0.70) (1.87) (1.73) (0.74) (2.00) (1.75) (0.69) (2.00) (1.99)
Real exchange rate 1.85 12.06 -4.40 2.55 7.52 1.93 2.63 12.30 -1.99
(5.13) (10.11) (11.01) (7.16) (10.78) (11.38) (6.01) (12.83) (12.11)
Hours per worker 0.24 -1.70** 2.25*** 0.48 -1.40 2.53*** 0.17 -2.16** 2.38***
(0.48) (0.82) (0.78) (0.57) (0.94) (0.81) (0.48) (0.98) (0.67)
Employment 1.21* 1.34 0.79 1.40* 1.90 -0.18 1.09 2.73 -0.46
(0.73) (2.19) (1.60) (0.84) (2.55) (1.44) (0.70) (1.74) (1.07)
Real wage rate 1.10 0.11 1.91 1.97 3.03 1.02 2.10 0.60 1.24
(1.15) (2.88) (2.43) (1.55) (3.07) (2.00) (1.49) (3.15) (2.30)
Notes: the table reports average multipliers from fiscal spillovers for variables indicated in the left column. A positive coefficient for
the real exchange rate means that the real exchange rate appreciates. State-dependent multipliers are for the case where the state of the
business cycle is measured as deviation of the output growth rate from trend. The estimated specification is given by equation (1’) for
consumption, investment, exports, and imports, and by equation (1’’) for the real exchange rate, hours per worker, employment and
the real wage rate.
Table 4. Average output multipliers over 3 years by exchange rate regime
State-dependent multipliers where state measured
Linear model
as deviation from trend (output growth rate)
Fixed Float
Fixed Float
Expansion Recession Expansion Recession
(1) (2) (3) (4) (5) (6)
Base 1.46 2.51 0.80 2.06 -1.38 4.29
(1.13) (1.93) (1.92) (2.61) (2.40) (2.75)
Only old/large OECD economies in 2.08* 3.77* -1.15 4.53 -1.27 5.89*
construction of spillover shocks (1.16) (2.12) (1.96) (2.85) (2.56) (3.22)
Constrain the sample to pre-2008 1.73 2.98 -0.69 4.16 -1.23 5.86*
(1.17) (1.91) (1.98) (2.90) (2.16) (3.39)
Notes: the table reports average output multipliers from fiscal spillovers. The estimated specifications are given by equation (5) in
columns (1) and (2) and the corresponding modification of equation (4) in columns (3) through (6). Driscoll and Kraay (1998)
standard errors are in parentheses. ***, **, * denote significance at 0.01, 0.05, and 0.10 levels.
Table 5. Average output multipliers over 3 years by exchange rate regime and by business cycle regime
State-dependent multipliers where state measured as deviation
Linear model
from trend (output growth rate)
Expansion in source country Recession in source country
Expansion Recession
Expansion in Recession in Expansion in Recession in
in source in source
recipient recipient recipient recipient
country country
country country country country
(1) (2) (3) (4) (5) (6)
Base -0.58 2.50 0.54 0.38 -2.21 5.34
(2.28) (2.29) (4.51) (4.83) (6.90) (3.69)
Only old/large OECD economies in 1.26 2.86 -0.65 6.92 -5.21 5.21
construction of spillover shocks (2.74) (2.40) (4.36) (4.87) (5.80) (4.29)
Constrain the sample to pre-2008 -2.08 4.19** -2.49 -0.80 3.46 8.05**
(2.02) (2.10) (3.84) (4.29) (5.53) (3.92)
Notes: the table reports average output multipliers from fiscal spillovers. The estimated specifications are given by equation (6) in
columns (1) and (2) and the corresponding modification of equation (4) in columns (3) through (6). Driscoll and Kraay (1998)
standard errors are in parentheses. ***, **, * denote significance at 0.01, 0.05, and 0.10 levels.
Figure 1. Time series of fiscal spillover shocks for selected countries
%-of-GDP shock %-of-GDP shock
USA
-.1 -.05 0 .05 .1 -.6-.4-.2 0 .2 .4 Canada
1985 1990 1995 2000 2005 2010 1985 1990 1995 2000 2005 2010
%-of-GDP shock %-of-GDP shock
France Italy
-.15-.1-.05 0 .05 .1 -.15-.1-.05 0 .05 .1
1985 1990 1995 2000 2005 2010 1985 1990 1995 2000 2005 2010
%-of-GDP shock %-of-GDP shock
Finland Poland
-.2 -.1 0 .1 .2 -.1-.05 0 .05 .1
1985 1990 1995 2000 2005 2010 1985 1990 1995 2000 2005 2010
%-of-GDP shock %-of-GDP shock
Japan
Australia
-.1 -.05 0 .05 .1 -.1 0 .1 .2 .3
1985 1990 1995 2000 2005 2010 1985 1990 1995 2000 2005 2010
Notes: The figure shows time series of spillover shocks as percent of GDP of the corresponding country in the
𝐺𝑆ℎ𝑜𝑐𝑘
previous period, i.e. 𝑌 𝑖𝑡 ∗ 100.The thick, solid black line is constructed as in equation (3) using government
𝑖,𝑡−1
spending innovations in all countries. The red, thin, dashed line is constructed as in equation (3) using
government spending innovations in only large OECD economies. The solid, thin, blue line is constructed as in
equation (2) with fixed weights using government spending innovations in only large OECD economies.
Figure 2. Impulse response functions (fixed weights)
Base, linear Base, expansion Base, recession
-6 -3 0 3 6 9 12 -6 -3 0 3 6 9 12 -6 -3 0 3 6 9 12
0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5
Horizon Horizon Horizon
Old/Large, linear Old/Large, expansion Old/Large, recession
-6 -3 0 3 6 9 12 -6 -3 0 3 6 9 12 -6 -3 0 3 6 9 12
0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5
Horizon Horizon Horizon
pre-2008, linear pre-2008, expansion pre-2008, recession
-6 -3 0 3 6 9 12 -6 -3 0 3 6 9 12 -6 -3 0 3 6 9 12
0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5
Horizon Horizon Horizon
Notes: the figure reports impulse response of output to a GSshock with fixed weights (equation
(2)). The left column reports impulse responses in the linear specification (1) while the right two
columns show responses in the non-linear specification (4). Each row corresponds to the sample
of countries and construction of shocks described in section 4.A. Solid, thick, red lines are point
estimates. Dashed, blue, thin lines are 95% confidence intervals constructed using Driscoll and
Kraay (1998) standard errors.
File and source
- File
- output-spillovers-from-fiscal-policy.pdf
- Size
- 281,421 bytes
- SHA-256
- 567e3a07578346884746f1e253ef3f5107e8bf6754924bd78a041986d5d56419
- Original
- www.nber.org