What Makes Discretionary Counter-Cyclical Fiscal Policy so Difficult? An Analysis of 32 OECD Countries

Abstract

We examine the stance of fiscal policy for 32 OECD countries from 1986 to 2023 by comparing for each country-year observation the signs of the output gap and the change in the cyclically adjusted budget balance. We find that fiscal policies are often pro-cyclical. We test possible explanations using comparative statistics for country-year observations with pro- and counter-cyclical policies. Our evidence suggests that fiscal rules and institutional quality are not related to the prevalence of pro-cyclicality. Likewise, several political-economy arguments suggested for explaining pro-cyclical fiscal policies do not receive strong support. However, poor access to finance makes fiscal policies more pro-cyclical. (JEL codes: E62, H50, and H62)

Full text

What Makes Discretionary Counter-Cyclical
Fiscal Policy so Difficult? An Analysis of
32 OECD Countries
Jakob de Haan*
,†
and Bram Gootjes*
*Faculty of Economics and Business, University of Groningen, The Netherlands. E-mail:
jakob.de.haan@rug.nl (corresponding author) and
†
CESifo, Munich, Germany
Abstract
We examine the stance of fiscal policy for 32 OECD countries from 1986 to 2023 by
comparing for each country-year observation the signs of the output gap and the
change in the cyclically adjusted budget balance. We find that fiscal policies are
often pro-cyclical. We test possible explanations using comparative statistics for
country-year observations with pro- and counter-cyclical policies. Our evidence sug-
gests that fiscal rules and institutional quality are not related to the prevalence of
pro-cyclicality. Likewise, several political-economy arguments suggested
for explaining pro-cyclical fiscal policies do not receive strong support. However,
poor access to finance makes fiscal policies more pro-cyclical. (JEL codes: E62, H50,
and H62)
Key words: fiscal policy; economic stabilization; pro-cyclical
1. Introduction
Fiscal policy can be used for macroeconomic stabilization purposes. From a normative per-
spective, a reasonable counter-cyclical goal of fiscal policy would be the same as that of
monetary policy: keep real GDP close to potential GDP when inflation is on target (Taylor
2000). If the economy is in a recession, say, government may increase spending and/or
lower taxes and finance its revenues shortfall by issuing government debt. This expansion-
ary policy will stimulate the economy, just like a cut in monetary policy rates would.
Likewise, during an economic boom, the government may decide to increase taxes and/or
lower spending to reduce government debt (i.e. contractionary fiscal policy). Fiscal policy
that stabilizes the business cycle is counter-cyclical, while it is neutral if it has no systematic
impact on the cyclicality of economic activity and pro-cyclical if it tends to amplify fluctua-
tions. As pointed out by the IMF (2005), symmetry in the fiscal response between good and
bad times is important for three main reasons: (1) rebuilding buffers ahead of the next cyc-
lical downturn; (2) reducing the risk of overheating; and (3) avoiding a ratcheting up of
public debt over successive cycles.
V
CThe Author(s) 2023. Published by Oxford University Press on behalf of Ifo Institute, Munich.
This is an Open Access article distributed und er the terms of the Creative Commons Attribution License (https://creativecommons.org/
licenses/by/4.0/),which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is
properly cited.
CESifo Economic Studies, 2023, 1–20
https://doi.org/10.1093/cesifo/ifad001
Original article
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

A distinction is often made between discretionary fiscal policy and automatic stabilizers.
Discretionary changes in taxes and spending require decisions by the policymaker, whereas
automatic stabilizers are mechanisms that automatically loosen the fiscal stance when the
economy slows down and that tighten the fiscal stance when the economy is booming (e.g.
unemployment benefits). The strength of automatic stabilizers depends on factors such as
the progressivity of the tax and transfer system. Both types of fiscal policy impact aggregate
demand, but the automatic stabilizers are more predictable and work more quickly than
discretionary fiscal policy actions (Taylor 2000).
1
This is due to several lags. The govern-
ment needs parliamentary approval to change spending or taxes and this process takes time
(the legislative lag). Moreover, once a fiscal policy bill is accepted by parliament, it takes
time to implement it (the implementation lag). Finally, it takes time for fiscal policy to have
an effect (the impact lag). These long lags make discretionary fiscal policy a less effective
stabilization tool than monetary policy. Furthermore, it was widely believed that monetary
policy was strong enough to do the job. Fiscal policy was simply not necessary.
Furthermore, some economists had doubts about its effectiveness (Barro 1974).
At the beginning of this century, most economists therefore dismissed discretionary fis-
cal policy and argued that automatic stabilizers should be relied upon (see, for instance,
Taylor 2000;Beetsma 2013;Bradford de Long and Tyson 2013 for discussions of the litera-
ture). However, more recently views have changed (Fata´s and Mihov 2012;Beetsma 2013).
For instance, Bradford de Long and Tyson (2013) argue that ‘with the reduction of short-
term nominal policy interest rates to zero toward the end of 2008, there came recognition
that conventional open-market operations were not powerful enough to accomplish the
desired stabilization of aggregate demand. This recognition led to another: that discretion-
ary fiscal policy might have an appropriate stabilization role...’. Furthermore, recent re-
search suggests that fiscal policy has gained importance as fiscal multipliers were found to
be higher than in the past. The fiscal multiplier is the impact on GDP of a one-unit increase
in the government deficit. Research suggests that multipliers depend on several factors (see
Bradford de Long and Tyson 2013;Bonam et al. 2022 for discussions of this literature).
First, several studies report a reduced effect of fiscal policy when sovereign debt levels are
high. Expectations of sustained deficits leading to higher public debt may undermine invest-
ors’ confidence in the government’s ability to face its obligations in full and in all circum-
stances. Second, multipliers tend to be higher when the interest rate is at its effective lower
bound (Bonam et al. 2022). Finally, there is evidence that the size of the multiplier is signifi-
cantly larger during downturns than during expansions, while countries operating with
fixed exchange rate systems have larger multipliers than countries with flexible exchange
rate systems.
The previous discussion would suggest that discretionary fiscal policy should be used
for stabilization purposes. In practice, however, fiscal policy is frequently pro-cyclical
(Bartsch et al. 2020).
2
This is problematic because pro-cyclical fiscal policies during an
1 Based on an analysis of the cyclical behavior of fiscal policy among 23 OECD countries, Fata´s
and Mihov (2012) conclude that for most countries, automatic changes in the budget balance
play a stronger role in stabilizing output than discretionary fiscal policy. When compared across
countries, changes in fiscal policy stance are predominantly linked to differences in government
size.
2 Previous studies report that fiscal policy is often pro-cyclical in emerging markets and develop-
ing economies (see, e.g., Gavin and Perotti, 1997), but Ilzezki and Ve´gh (2008) find evidence of
2CESifo Economic Studies, 2023, Vol. 00, No. 0
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

economic downswing only make bad times worse. This argument has received considerable
attention in the literature (see, for instance, Alesina et al. 2015), but recently the other side
of the coin has also become relevant: pro-cyclical fiscal policy can further overheat the
economy that is in an upswing, thereby making the job of the central bank to keep inflation
on target more difficult. Using data for a panel of ten euro-area (EA) countries and a panel
smooth transition local projections model, Kloosterman et al. (2022) report that a contrac-
tionary monetary shock only reduces inflation if fiscal policy is also contractionary.
Furthermore, business cycle stabilization is important because macroeconomic volatility
can hamper medium-term growth (Furceri and Jalles 2018).
3
Pro-cyclical fiscal policy may
also undermine the sustainability of public finances, particularly so in European Economic
and Monetary Union (EMU) countries. As their economies are heavily integrated, fiscal
policy in one EMU country may have large spillovers to other economies; these spillovers
are generally not internalized (Eyraud et al. 2017).
This paper takes stock of research on pro-cyclical discretionary fiscal policy. First,
Section 2 examines the stance of fiscal policy for OECD countries. We update the analysis
of Bartsch et al. (2020) for 32 OECD countries for the period 1986–2023, so that the epi-
sode of the COVID-19 pandemic is included.
4
We focus on OECD countries, as previous
research suggests that the drivers of pro-cyclical policy differ among advanced and develop-
ing countries (Halland and Bleaney 2011). To determine the stance of fiscal policy, we com-
pare—for each country-year observation—the signs of the output gap and the change in the
cyclically adjusted budget balance. Previous research on the cyclicality of fiscal policy uses
panel-regression models to link fiscal variables to the business cycle; the sign of the (static
or time-varying) coefficient associated with output (or the output gap) is interpreted as the
cyclicality of the fiscal variable.
5
Following such a method is helpful for measuring the ex-
tent to which the average fiscal policy stance is pro-cyclical, but it does not reveal the prob-
ability that fiscal policy is pro-cyclical in any given year. Our approach has the advantage
that the cyclicality of fiscal policy can be determined for every country-year observation.
The other sections zoom in on several explanations that have been put forward in the lit-
erature to explain why fiscal policy is most often pro-cyclical. We test these explanations
by using simple comparative statistics for country-year observations with pro- and counter-
cyclical policies. Section 3 discusses the view that fiscal rules and poor institutional quality
make fiscal policy pro-cyclical, while Section 4 reviews the argument that lack of access to
pro-cyclical fiscal policies in high-income countries as well. Some other studies report that fiscal
policy is most-often counter-cyclical (see, e.g., Wyplosz 2006;Be´ne´trix and Lane 2013 for EMU
member countries, Fata´s and Mihov 2012 for OECD countries, and Frankel et al. 2013 for develop-
ing economies), but recent studies do not confirm this finding. For instance, Bova et al. (2018)
and Eyraud et al. (2017) do not find that the pro-cyclical bias has declined over time in resource-
rich and euro area countries, respectively. As explained below, all these studies use a different
methodology than Bartsch et al. (2020) and the present paper to identify pro-cyclical fiscal
policy.
3Aghion et al. (2007) report that (i) counter-cyclical fiscal policy is growth enhancing and (ii) the
effect of counter-cyclical fiscal policy is larger at lower levels of financial development. Similar
results are reported by Woo (2009) and McManus and Ozkan (2015).
4Bartsch et al. (2020) focus on 22 OECD countries for the 1986-2019 period.
5 This is the approach followed by studies such as Lane (2003),Alesina et al. (2008),Fata´s and
Mihov (2012),Combes et al. (2017),Furceri and Jalles (2018),Jalles (2021), and Afonso and
Carvalho (2022).
CESifo Economic Studies, 2023, Vol. 00, No. 0 3
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

finance may make fiscal policy pro-cyclical. This section also evaluates political-economy
reasons suggested to explain pro-cyclical fiscal policies. Section 5 concludes.
2. The Facts
Our metric for fiscal policy is the government budget balance excluding (i) the estimated ef-
fect of the business cycle on the budget; (ii) interest expenditures (which reflect past deficits
and are largely outside the control perimeter of the government); and (iii) one-off revenues
or expenses. Although automatic stabilizers are clearly part of fiscal stabilization policies, it
is common practice to focus on policy changes resulting from discretionary decisions by
policymakers (Bartsch et al. 2020). Hence, international institutions usually measure the
fiscal stance by the (change in the) cyclically adjusted primary balance (CAPB). We follow
Golinelli and Momigliano (2009) and examine the change in the CAPB (as a percentage of
GDP). As the actual budget balance is affected by the business cycle, its change is not an ad-
equate indicator of discretionary fiscal policy. The calculation of the CAPB involves an esti-
mation of what revenues and expenditure (and thus the budget balance) would be if the
economy were at its trend output, rather than its actual output. If the CAPB decreases
(increases), fiscal policy is expansionary (contractionary).
6
The data come from the
OECD.
7
Our sample period is 1986–2023 and we focus on 32 OECD countries.
8
When in a certain year the change of the CAPB and the output gap (also from the OECD)
have opposite signs, and if the absolute change in the budget balance exceeds 0.2% of GDP,
fiscal policy is considered pro-cyclical. If output is below trend (i.e. the output gap is nega-
tive), counter-cyclicality requires that fiscal policy should be expansionary (i.e. the change in
the CAPB is negative). To ensure that years with very small changes in the budget balance do
not distort the analysis, we impose that the absolute change in the CAPB should be more than
0.2% of GDP. Our analysis is descriptive and based on ‘ex-post’ data (as opposed to ‘real-
time’ measures capturing information at the time when the decision is made).
Bartsch et al. (2020) propose an alternative to define fiscal policy cyclicality. These
authors relate the change in the CAPB to the change in the output gap. They ‘define a fiscal
policy as counter-cyclical if the year-on-year change in the underlying primary balance has
the same sign as the change in the output gap (for example, it rises when actual GDP
growth exceeds potential growth)’ (p. 35). To examine whether our results are sensitive to
our definition of cyclicality, we have also used the approach of Bartsch (2020) to define
pro- and counter-cyclical fiscal policy. The results are reported in the Appendix.
6 So, we focus on the direction of the fiscal policy stance for each country-year observation. The ap-
proach used by most previous studies on the cyclicality of fiscal policy as described in the previous
section does not yield this outcome for each and every country-year observation but focuses on the
average fiscal policy stance over the estimation period. However, an advantage of this approach is
that it allows inferences about the extent to which fiscal policy is pro- or counter-cyclical.
7 We have also calculated the output by applying the HP filter to output. The correlation of both
output gaps is 0.8.
8 We use the November 2022 edition of the OECD Economic Outlook to retrieve data on fiscal and
economic variables. The dataset includes projections of the output gap and the CAPB through
2024, but we omit the last year from the analysis because revisions in the forecast could be large
and fiscal policy for 2024 is still in the planning phase. Note that the data for 2022 and 2023 is
sensitive to revisions.
4CESifo Economic Studies, 2023, Vol. 00, No. 0
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

Figure 1 shows the share of countries with pro-cyclical fiscal policies over time.
9
The
percentage of countries with pro-cyclical policies peaks at 78% in 1996 after which it grad-
ually drops to 31% in 2003. After that year, the share of countries with pro-cyclical policies
increased again until 2008; during the financial crisis, fiscal policy was often counter-
cyclical. Likewise, during the height of the COVID-19 pandemic in 2020, only 6% of the
countries had pro-cyclical fiscal policies, but more recently the share of countries with pro-
cyclical policies increased again to almost 60% in 2023.
Looking at the percentage of pro-cyclical years over the same period for each country
(Figure 2), it seems that Denmark, Switzerland, Korea, and Japan seem less prone to pro-
cyclical policies than the other countries. Pro-cyclical fiscal policies are most frequent in
Ireland, Greece, Poland, and Lithuania.
Bartsch et al. (2020) argue that large countries may have more leeway to conduct
counter-cyclical policies, due to higher fiscal multipliers and more de facto monetary inde-
pendence (regardless of the exchange rate regime). Hence, they should be less prone to fiscal
pro-cyclicality. The first bars in Figure 3 provide only weak evidence for this. On average,
large OECD countries (the USA, the UK, Japan, Germany, France, Italy, and Spain) have
pro-cyclical fiscal policies only slightly less often than the other OECD countries in our
sample (in 45% versus 49% of all country-year observations in both subsamples). A two-
sample t-test indicates that the difference is not statistically significant (P-value is 0.259; see
Table A2).
10
However, these averages mask variation over time. Figure A2 shows that dur-
ing the 1990s, fiscal policy is more often pro-cyclical in the large economies of our sample,
while the smaller economies tended to exhibit more pro-cyclicality in the past two decades
(except during the COVID-19 pandemic).
Likewise, as shown in the second part of Figure 3, countries that have an above-median
level of real GDP per capita on average have only slightly less often pro-cyclical fiscal poli-
cies than countries with a below-median level of economic development (47% versus 49%;
P-value of the t-test is 0.530). In this case, there is not a clear variation over time (results
available on request). Our finding is in contrast with the results of some other studies. For
instance, Jalles (2021) concludes that more developed economies tend to have a smaller de-
gree of pro-cyclicality of government spending. Lane (2003) finds similar results for govern-
ment spending, but the coefficient of his proxy for economic development is not significant
in the model for the cyclicality of the budget balance. As pointed out in Section 1, these
studies employ a different methodology to assess fiscal pro-cyclicality.
3. Are Fiscal Rules and Poor Institutional Quality to Blame?
Fiscal rules are often blamed for causing pro-cyclical policies. For instance, the European
Stability and Growth Pact has been criticized a lot. One of the main concerns was that these
9 As shown in Figure A1, the approach of Bartsch et al. (2020) yields fairly similar results, al-
though there are some differences, notably during the global financial crisis and at the end of
the sample period when economies were recovering from the COVID-19 pandemic. In both peri-
ods, our measure indicates that more countries pursued pro-cyclical policies than the measure
of Bartsch et al. (2020), which reflects that in both periods the level and the change of the out-
put gap differed.
10 Table A3 replicates Table A2 if we define pro- and counter-cyclical fiscal policy following
Bartsch et al. (2020).
CESifo Economic Studies, 2023, Vol. 00, No. 0 5
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

fiscal rules might restrain governments from implementing expansionary policies during
downturns, which would make fiscal policy pro-cyclical (cf. Buiter 2004). Gal
ı and Perotti
(2003) were among the first to examine whether the EU fiscal rules forced countries into
following pro-cyclical policies. The authors compare fiscal policy for the euro countries
during the period preceding the Maastricht Treaty (1980–1991) to the period after the
Treaty was signed (1992–2002). They find that there is a trend toward less pro-cyclicality.
This contrasts with the view that the EU fiscal rules have hampered stabilization.
Specifically, the estimates suggest that before Maastricht, fiscal policies in eurozone coun-
tries were pro-cyclical, while after Maastricht they were acyclical. Bartsch et al. (2020) re-
port that when they split their sample into EA and non-EA countries, the pro-cyclicality of
fiscal policy in EA countries stands out in the 1990s (in the run-up of monetary unification)
and again in the 2010s (during the sovereign debt crisis). Aghion et al. (2007) find that
EMU countries have more often pro-cyclical policies than the other countries in their
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Figure 1. Cyclicality of fiscal policies over time, 32 OECD countries, 1986–2023 [share of the number of
countries (%)]. Note: Fiscal policy is pro-cyclical if (i) DCAPB 60.2 (% of potential GDP) and (ii)
DCAPB and the output gap have opposite signs. Source: own calculations based on OECD data.
0%
10%
20%
30%
40%
50%
60%
70%
Figure 2. Cyclicality of fiscal policies across countries, 32 OECD countries, 1986–2023 [share of the
number of years (%)]. Note: Fiscal policy is pro-cyclical if (i) DCAPB 60.2 (% of potential GDP) and (ii)
DCAPB and the output gap have opposite signs. Source: own calculations based on OECD data.
6CESifo Economic Studies, 2023, Vol. 00, No. 0
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

sample of OECD countries. Afonso and Carvalho (2022) assess the cyclicality of fiscal pol-
icy in the 19 euro-area countries for the 1995–2020 period using a time-varying measure of
fiscal cyclicality. Their results suggest that for six of the 19 countries (Estonia, France,
Greece, Lithuania, Portugal, and Slovakia), fiscal policy was pro-cyclical on average. These
authors also find that during periods with negative output growth discretionary fiscal pol-
icy becomes more pro-cyclical.
The third set of bars in Figure 3 shows the extent to which fiscal policy in the first 11
Member States of the EA was pro-cyclical before and after 1999 (the start of EMU). We
find some evidence of a difference in cyclicality. The graph indicates that between 1986
and 1998, fiscal policy is pro-cyclical in 53% of the years, while in the years after the start
of the EMU fiscal policy is pro-cyclical in 45% of the years. The difference in averages
thus seems to indicate that the fiscal policies of the eleven countries that first entered the
eurozone became less pro-cyclical after the start of the EMU, but a simple t-test rejects this
(P-value is 0.132). However, this result seems to mask variation over time. Figure A3
shows that around the period of the sovereign debt crisis (2011–2013), fiscal policy in the
EA was pro-cyclical in 85% of the years. In other years, fiscal policy appears less pro-
cyclical (40% on average). Moreover, while the fourth part of Figure 3 points out that
since the start of the EMU, there is no substantial difference between EA and non-EA
countries (47% versus 51%; P-value of the t-test is 0.416), non-EA countries had a consid-
erably lower percentage of pro-cyclical fiscal policies during the sovereign debt crisis (fig-
ure showing the variation over time between EA and non-EA countries is available on
request).
Of course, EU countries are not the only ones having fiscal rules. Quite a few studies
have examined the impact of fiscal rules using cross-country samples. In contrast to popular
views, some studies suggest that well-designed fiscal rules may reduce the pro-cyclicality
of fiscal policy (Nerlich and Reuter 2015;Combes et al. 2017;Guerguil et al. 2017;
40%
42%
44%
46%
48%
50%
52%
54%
Country size Economic
development
EA EMU Fiscal rules Institutional
quality
No / Below median Yes / Above median
Figure 3. Pro-cyclical policies: economic and institutional factors (percentage of country-year observa-
tions). Notes: When the variable of interest is a 0–1 dummy variable [country size; euro area (EA);
Economic and Monetary Union (EMU)], the percentage of pro-cyclical fiscal policies is split based on
no ¼0, yes ¼1. When the variable of interest is continuous, the sample is split based on whether the
value of the variable of interest is below or above the median. The median is calculated on a year-by-
year basis, so that it may increase or decrease over time. Source: own calculations based on OECD
data.
CESifo Economic Studies, 2023, Vol. 00, No. 0 7
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

Gootjes and de Haan 2022a;Reuter et al. 2022),
11
but Bergman and Hutchison, (2015)
suggest that this result only holds when these rules are supported by a sufficiently high level
of government efficiency/quality.
Most studies, however, conclude that fiscal rules did not reduce pro-cyclicality (see, for
example, Furceri and Jalles 2018). Moreover, Jalles (2018) uses time-varying country-spe-
cific measures of fiscal policy counter-cyclicality for a large sample of 60 countries between
1980 and 2014 and finds that fiscal rules reduce the degree of fiscal counter-cyclicality. The
result is especially strong for debt-based rules in advanced economies. Bova et al. (2014)
show that the adoption of fiscal rules did not help emerging markets and developing econo-
mies escape the pro-cyclicality trap. Similarly, Bova et al. (2018) find that the adoption of
fiscal rules does not reduce the pro-cyclical stance of fiscal policy in a significant way.
Instead, these authors present results showing that the efficiency/quality of government
institutions helps to limit pro-cyclicality. Caldero´ n et al. (2016) also conclude that the qual-
ity of the institutional framework plays an important role in countries’ ability and willing-
ness to implement counter-cyclical fiscal policies. Similar results are reported by Frankel
et al. (2013), who argue that the political institutional framework is pivotal for developing
countries’ ability to graduate from pro-cyclical to counter-cyclical policies. For developing
countries, Ardanaz and Izquierdo (2022) find that upswings are associated with increases
in current primary expenditures, while public investment falls and current spending remains
acyclical during downturns. Using a similar institutional quality index as Frankel et al.
(2013), the authors present evidence suggesting that this asymmetrical response is more
pronounced in countries where incumbent politicians face shorter time horizons (proxied
by the number of years left in office) and weak institutions.
The two sets of bars on the right-hand side in Figure 3 show the extent to which fiscal
policy is pro-cyclical if we split our sample based on whether a country-year observation is
above or below the median of the IMF fiscal rules index and the World Bank governance
index, respectively (see Table A1 for details). As the strength of fiscal rules increases over
time (Gootjes and de Haan 2022b), we classify country-year observations using the median
of the fiscal rules index for the countries in our sample for each year. We do the same for
the institutional quality index. The graphs do not provide evidence that fiscal rules and in-
stitutional quality are strongly related to the prevalence of pro-cyclical fiscal policy. The
share of country-year observations with pro-cyclical policies is the same when countries are
split by the median score of the fiscal rules index.
12
Likewise, the percentages for below-
and above-median institutional quality observations do not differ a lot either (50% versus
11 Using data of the EU28 Member States for the period 1996–2015, Reuter et al. (2022) report that
rules reduce fiscal volatility. They document that this result only holds for rules which are
designed to be unaffected by the current state of the business cycle. Furthermore, the strin-
gency of fiscal rules makes their stabilising effect more pronounced, while the effects of the
rules can be observed even if they are not complied with.
12 We have also examined whether different types of fiscal rules affect our findings. We find no
significant effect for different types of fiscal rules (results are available on request). Likewise,
most fiscal rule characteristics do not make a difference in the extent to which fiscal is pro-
cyclical, except for an independent body outside the government being responsible for monitor-
ing the implementation of fiscal rules. Discretionary fiscal policy is pro-cyclical 50% of the time
in countries without fiscal councils versus 43% in countries with these institutions (the P-value
of the t-test statistic is 0.081).
8CESifo Economic Studies, 2023, Vol. 00, No. 0
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

47%; P-value of the t-test is 0.479). However, if we follow the definition of Bartsch et al.
(2020) to define the cyclicality of fiscal policy, the P-value of the t-test for institutional
quality is significant as shown in Table A3.
All in all, our evidence does thus not provide strong support that fiscal rules and institu-
tional quality are related to the prevalence of pro-cyclical fiscal policies.
4. Access to Finance and Political Distortions
According to Frankel et al. (2013), two main explanations, which are not necessarily incon-
sistent, have been given in the literature to explain pro-cyclical fiscal policy: (i) imperfect
access to international credit markets and lack of financial depth and (ii) political distor-
tions. Lack of access to (international or national) credit markets in bad times forces gov-
ernments to cut spending and raise taxes, whereas political pressures for additional
spending in good times are hard to resist.
Although the evidence is mixed, most studies find that access to credit is related to the
cyclicality of fiscal policy. For instance, Jalles (2021) reports that pro-cyclical government
spending is negatively associated with financial development, as it is easier for governments
to raise money during bad times. Similar results are reported by Aghion et al. (2007),Jalles
(2018),Furceri and Jalles (2018), and Afonso and Carvalho (2022). However, the latter
three studies do not find evidence that capital account openness (proxied by the Chinn–Ito
index) makes fiscal policy less pro-cyclical. In contrast, Frankel et al. (2013) report that
both more financial integration and depth are associated with more counter-cyclicality/less
pro-cyclicality of government spending.
The first sets of bars in Figure 4 show the relationship between access to domestic and
international credit and the prevalence of pro-cyclical fiscal policy. The first two sets of
bars focus on two widely used indicators of financial development, namely the ratio of pri-
vate credit to GDP and the IMF financial development index (see Table A1 for details),
while the third set of bars presents the relationship between fiscal pro-cyclicality and capital
account openness (proxied by the Chin-Ito index; cf. Frankel et al. 2013;Afonso and
Carvalho 2022). The graphs suggest that governments that have a more difficult time
financing budget deficits often have pro-cyclical fiscal policies. For instance, 52% of the
country-year observations turn out to be characterized by pro-cyclical fiscal policy in coun-
tries with a below-median level of financial development, while in countries with an above-
median level of financial development, this percentage amounts to 45% (the two-sample
t-test indicates that the difference in means is significant with a P-value of 0.023). Similarly,
53% of the country-year observations turn out to be characterized by pro-cyclical fiscal
policy in countries with a below-median level of capital account openness, while in coun-
tries with an above-median level of capital account openness, this percentage is 46%
(P-value of the t-test is 0.056).
13
Private credit does not seem to make a difference (47%
versus 45%; P-value of the t-test is 0.481). However, consistent with the results for the
financial development and financial openness variables, the fourth set of bars in Figure 4
suggests that countries with an above-median level of long-term interest rates more often
13 When we use the definition of Bartsch et al. for pro- and counter-cyclical fiscal policy, the
t-test for capital account openness is not significant (see Table A3).
CESifo Economic Studies, 2023, Vol. 00, No. 0 9
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

have pro-cyclical fiscal policies than countries with lower interest rates (51% versus 43%;
the difference in percentages is significant with a P-value of 0.011).
The final sets of bars in Figure 4 show the relationship between fiscal pro-cyclicality and
trade openness and the size of the public sector. In contrast to previous studies, we do not
find strong evidence that trade openness (proxied by the sum of exports and imports as a
share of GDP) and the size of the public sector (proxied by general government spending as
a share of GDP) is related to pro-cyclical fiscal policy in our sample of OECD countries.
14
The share of country-year observations with pro-cyclical fiscal policy is the same for obser-
vations with a below- or above-median level of trade openness. Country-year observations
with an above-median level of public sector size have a slightly higher (49%) degree of pro-
cyclical fiscal policy than those with a below-median level of government spending (46%),
but the difference is not significant (P-value of the t-test is 0.441).
According to some political-economy arguments, politicians generally tend to prefer ex-
pansionary fiscal policy over contractionary policy. Notably during economic good times,
windfall revenues create more room for their preferred spending and tax programs. As a
consequence, fiscal policies are often pro-cyclical, especially in good times (Alesina et al.
2008). The first set of bars in Figure 5 does not provide support for this view. The share of
country-year observations with pro-cyclical fiscal policies in good and bad times does hard-
ly differ.
Alesina et al. (2008) argue that the pro-cyclical bias of fiscal policy arises from voters’
demands. As voters cannot directly observe government capital spending, they demand
higher utilities during a flourishing economy, fearing that the government will otherwise
appropriate part of the tax revenues for political rents. The empirical evidence of Alesina
et al. (2008) shows that pro-cyclical fiscal policy is more prevalent in more corrupt coun-
tries. The authors also find that corruption affects pro-cyclicality only in democracies.
Halland and Bleaney (2011) confirm the importance of corruption and democracy for their
40%
42%
44%
46%
48%
50%
52%
54%
Private credit Financial
development
Capital account
openness
Long-term interest
rates
Trade openness Government
spending
No / Below median Yes / Above median
Figure 4. Pro-cyclical policies: financial-economic conditions (percentage of country-year observa-
tions). Note: See notes to Figure 3.Source: own calculations based on OECD data.
14 Lane (2003),Aghion et al. (2007),Jalles (2018), and Furceri and Jalles (2018) report that
trade openness and government size are positively correlated with counter-cyclical fiscal pol-
icy. When we use the definition of Bartsch et al. for pro- and counter-cyclical fiscal policy, the
t-test for public sector size is significant (see Table A3).
10 CESifo Economic Studies, 2023, Vol. 00, No. 0
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

sample of developing countries, but they do not find convincing evidence that it is the com-
bination of the two that matters, as stressed by Alesina et al. (2008). For a sample of 37
African countries, Thornton (2007) finds that, in contrast to Alesina et al. (2008), less cor-
ruption leads to more pro-cyclicality. The author suggests that this may be due to corrup-
tion leading to lower levels of tax collection, generating lower government expenditure.
Better control of corruption may thus increase the tax revenues available for fiscal expend-
iture and cause fiscal policies to become pro-cyclical.
The second set of bars in Figure 5 does not offer very strong evidence for the role of cor-
ruption in fostering pro-cyclical fiscal policies in our sample of OECD countries. The share
of country-year observations with pro-cyclical fiscal policy is 50% for observations above
the median level of corruption, while it is only slightly lower (47%) for observations below
the median level of corruption (P-value of the t-test is 0.240).
15
One of the reasons that
democracy may matter, is that fiscal policy may be used by the incumbent during elections
to enhance chances for re-election (see de Haan and Klomp 2013 and Dubois 2016 for
reviews of research on political budget cycles). However, the third set of bars in Figure 5
does not offer strong support for this view. The share of pro-cyclical fiscal policy in election
and non-election years hardly deviate from one another.
Fata´s and Mihov (2006) find that US states with more Democrats in Congress have less
volatile fiscal policies. This is consistent with the partisan theory according to which left-
wing and right-wing parties follow policies that are in the interest of their constituencies
(see Potrafke 2017 for a survey of recent evidence). We therefore examine whether pro-
cyclicality is different under left-wing and other types of government.
16
As shown in
Figure 5, our results do not suggest that government ideology is significantly related to the
prevalence of pro-cyclical fiscal policy (the P-value of the t-test statistic is 0.58). We also
40%
42%
44%
46%
48%
50%
52%
Economic
upswing
Control of
corruption
Elections Government
ideology
Margin of
majority
Government
fragmentation
Checks and
Balances
Output
volatiltiy
Income
inequality
No / Below median Yes / Above median
Figure 5. Pro-cyclical policies: political-economy distortions (percentage of country-year observa-
tions). Note: See notes to Figure 3.Source: own calculations based on OECD data.
15 When we use the definition of Bartsch et al. for pro- and counter-cyclical fiscal policy, the t-test for
corruption is significant (see Table A 3 ).
16 Results are similar when we use data from the CPDS dataset.
CESifo Economic Studies, 2023, Vol. 00, No. 0 11
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

examine whether a larger majority of legislative seats held by the government matters.
Again, we do not find evidence that this variable is related to the prevalence of pro-cyclical
fiscal policies (the P-value of the t-test statistic is 0.46).
Lane and Tornell (1999) pose that a windfall in government revenues in good times is
considered to be a common pool by various interest-groups competing for it. As these
groups have no incentive to reduce their claims (knowing that the economic windfall will
otherwise accrue to other interest-groups), the windfall revenues will be spend making fis-
cal policy pro-cyclical. Lane (2003, p. 2665) summarizes the framework as follows: ‘mul-
tiple power blocs compete for a share in fiscal revenues. The notion of multiple powerful
fiscal groups is open to a number of interpretations. It can refer to different branches of
government (i.e. the executive versus congress); individual parties within a coalition; or
even individual ministries within the government. .... An important result in this modeling
approach is that the intensity of fiscal competition increases during upturns: the impact of
this ‘voracity effect’ is that spending can even grow more than proportionally relative to the
increase in income. The intuition is that the incentive to act prudently is low: each group
knows that if it refrains from increasing its appropriation rate during expansions, the result
is not that the government runs a budget surplus but that the other groups can increase their
appropriate rate by an even greater amount. Symmetrically, recessions have a chilling effect
on fiscal competition. Accordingly, a basic prediction of this approach is that political sys-
tems in which power is diffused among a number of agents will witness a higher degree of
fiscal pro-cyclicality relative to a unitary system’. Lane (2003) reports a significantly posi-
tive coefficient on the power dispersion index of Henisz (2002), indicating that power dis-
persion leads to more pro-cyclical surpluses. Furceri and Jalles (2018) find similar results.
The sixth and seventh sets of bars in Figure 5 show the relationship between the preva-
lence of pro-cyclical fiscal policy and proxies for government fragmentation and checks
and balances in the political system (the number of veto players). The graphs do not suggest
that these variables are clearly related to the pro-cyclicality of fiscal policies.
In the model of Talvi and Ve´ gh (2005), political pressures for additional spending intensify
during revenue booms and weaken during economic slumps, such that the government finds it
optimal to respond in a pro-cyclical manner to fluctuations in the tax base. As pointed out by
Lane (2003), an important feature of the Talvi–Vegh model is that, in general, these spending
pressures are an increasing, convex function of the incipient primary surplus (the larger the
boom, the more severe the political distortion). Accordingly, high output volatility is conducive
to generating pro-cyclical fiscal policy. Lane (2003) reports support for this. Figure 5 shows
that the share of observations with pro-cyclical fiscal policy amounts to 50% for country-year
observations with an above-median level of output volatility, while it is 46% for observations
that have a below-median level of volatility. The difference in means is not significant (P-value
of the t-test is 0.148), but note that when we use the definition of Bartsch et al. for pro- and
counter-cyclical fiscal policy the t-test for output volatility is significant (see Table A3).
Woo (2009) develops a model that links the polarization of preferences over fiscal
spending to the pro-cyclicality bias of fiscal policy. The author presents evidence that social
polarization as measured by income inequality and educational inequality is consistently
and positively associated with fiscal pro-cyclicality. However, Halland and Bleaney (2011)
do not confirm that income inequality plays a significant role in explaining pro-cyclical fis-
cal policies in developing countries. Likewise, the last set of bars in Figure 5 only indicates
a small difference in the percentage of pro-cyclical fiscal policies in OECD countries with
inequality below and above the median (48% versus 52%; P-value of the t-test is 0.414).
12 CESifo Economic Studies, 2023, Vol. 00, No. 0
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

5. Conclusions
Based on data for 32 OECD countries from 1986 to 2023, we conclude that fiscal policies
are often pro-cyclical, i.e., the signs of the change in the cyclically adjusted budget balance
and the of the output gap are different. In other words, most of the time fiscal policies amp-
lify the business cycle instead of stabilizing it. Previous studies suggest several explanations
for the pro-cyclicality of fiscal policy. One line of research shows that well-designed fiscal
rules and strong government institutions may lower the average degree of fiscal pro-
cyclicality. Our evidence, which is based on a simple year-on-year comparison of the stance
of fiscal policies in countries above and below the median of our proxies for fiscal rules and
institutional quality, suggests that these factors are not strongly related to the prevalence of
pro-cyclical policy. In other words, while fiscal rules or institutional quality may reduce the
extent to which discretionary fiscal policy has a destabilizing effect on the economy, they
do not appear to reduce the likelihood that fiscal policy will actually destabilize.
Our results do also not provide supporting evidence for several political-economy argu-
ments. Pro-cyclical fiscal policies are not more prevalent in corrupt countries, nor does fis-
cal policy more often turn pro-cyclical during election years than during non-election years.
Likewise, government fractionalization and the number of veto players are not clearly
related to the prevalence of pro-cyclical fiscal policies. However, our results suggest that
poor access to domestic or international finance makes fiscal policies more pro-cyclical.
There was already ample evidence showing that liquidity problems induce pro-cyclical fis-
cal policies in developing countries and emerging markets (see, for instance, Gavin and
Perotti 1997). In line with recent research (Aghion et al. 2007;Furceri and Jalles 2018;
Jalles 2018,2021;Afonso and Carvalho 2022), we find that borrowing constraints affect
the pro-cyclicality of fiscal policy in advanced economies just as well.
There is one important caveat to our results. Like most other studies on the pro-
cyclicality of fiscal policy, we employ data on policy outcomes rather than budget plans.
Yet, governments formulate fiscal plans while being uncertain about the economic outlook.
Therefore, the information available to governments at the time of making decisions can be
very different from economic realizations. As a consequence, ex-ante intentions of fiscal
policy can differ substantially from ex-post budgetary results (Croushore 2011;Cimadomo
2016). Indeed, Beetsma and Giuliodori (2010),Eyraud et al. (2017), and Gootjes and de
Haan (2022a) provide empirical evidence that fiscal policy plans often start out neutral, but
in the course of implementation becomes pro-cyclical (see also Demertzis et al. 2023).
Acknowledgment
The authors are grateful to an anonymous reviewer for their helpful feedback on a previous ver-
sion of the article.
References
Afonso, A. and F. T. Carvalho (2022), “Time-Varying Cyclicality of Fiscal Policy: The Case of the
Euro Area”, North American Journal of Economics and Finance 62, 101778.
Aghion, P., I. Marinescu, R. J. Caballero and A. K. Kashyap (2007), “Cyclical Budgetary Policy
and Economic Growth: What Do We Learn from OECD Panel Data?”, NBER Macroeconomics
Annual 22, 251–97.
Alesina, A., O. Barbiero, C. Favero, F. Giavazzi and M. Paradisi (2015), “Austerity in
2009-2013”, Economic Policy 30, 383–437.
CESifo Economic Studies, 2023, Vol. 00, No. 0 13
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

Alesina, A., F. Campante and G. Tabellini (2008), “Why Is Fiscal Policy Often Procyclical?”,
Journal of the European Economic Association 6, 1006–36.
Ardanaz, M. and A. Izquierdo (2022), “Current Expenditure Upswings in Good Times and Public
Investment Downswings in Bad Times? New Evidence from Developing Countries”, Journal of
Comparative Economics 50, 118–34.
Armingeon, K., S. Engler and L. Leemann (2022), Comparative Political Data Set 1960-2020,
Department of Political Science, University of Zurich, Zurich.
Barro, R. J. (1974), “Are Government Bonds Net Wealth?”, Journal of Political Economy 82, 1095–117.
Bartsch, E., A. Be´ nassy-Que´re´, G. Corsetti and X. Debrun (2020), It’s All in the Mix. How
Monetary and Fiscal Policies Can Work or Fail Together. Geneva Reports on the World
Economy 23, ICMBS/London, CEPR, Geneva.
Beetsma, R. and M. Giuliodori (2010), “Fiscal Adjustment to Cyclical Developments in the OECD:
An Empirical Analysis Based on Real-Time Data”, Oxford Economic Papers 62,419–41.
Beetsma, R. (2013), A Survey of the Effects of Discretionary Fiscal Policy. Report for the Swedish
Fiscal Policy Council, Stockholm.
Be´ne´trix, A. S. and P. R. Lane (2013), “Fiscal Cyclicality and EMU”, Journal of International
Money and Finance 34, 164–76.
Bergman, U. M. and M. Hutchison. (2015), “Economic Stabilization in the Post-Crisis World: Are
Fiscal Rules the Answer?”, Journal of International Money and Finance 52, 82–101.
Bonam, D., J. de Haan and B. Soederhuizen (2022), “The Effects of Fiscal Policy at the Effective
Lower Bound”, Macroeconomic Dynamics 26, 149–85.
Bova, E., N. Carcenac and M. Guerguil (2014), “Fiscal Rules and the Procyclicality of Fiscal
Policy in the Developing World”, IMF working paper 14/122. International Monetary Fund,
Washington DC.
Bova, E., P. Medas and T. Poghosyan; European Commission, Belgium (2018), “Macroeconomic
Stability in Resource-Rich Countries: The Role of Fiscal Policy”, Journal of Banking and
Financial Economics 2018, 103–22.
Bradford de Long, J. and L. D. Tyson (2013), “Discretionary Fiscal Policy as a Stabilization Policy
Tool: What Do We Think Now That We Did Not Think in 2007?,” https://delong.typepad.
com/20130531-delong-tyson-1.4.pdf
Buiter, W. H. (2004), “Two Naked Emperors? Concerns about the Stability & Growth Pact and
Second Thoughts about Central Bank Independence”, Fiscal Studies.25, 249–77.
Caldero´ n, C., R. Duncan and K. Schmidt-Hebbel (2016), “Do Good Institutions Promote
Countercyclical Macroeconomic Policies?”, Oxford Bulletin of Economics and Statistics 78,
650–70.
Chinn, M. D. and H. Ito (2006), “What Matters for Financial Development? Capital Controls,
Institutions, and Interactions”, Journal of Development Economics 81, 163–92.
Cimadomo, J. (2016), “Real-Time Data and Fiscal Policy Analysis: A Survey of the Literature”,
Journal of Economic Surveys 30, 302–26.
Combes, J., A. Minea and M. Sow (2017), “Is fiscal Policy Always Counter- (Pro-) Cyclical? The
Role of Public Debt and Fiscal Rules”, Economic Modelling 65, 138–46.
Croushore, D. (2011), “Frontiers of Real-Time Data Analysis”, Journal of Economic Literature
49, 72–100.
de Haan, J. and J. Klomp (2013), “Conditional Political Budget Cycles: A Review of Recent
Evidence”, Public Choice 157, 387–410.
Demertzis, M., M. Dom
ınguez Jime´nez and N. Viegi (2023), “Reviving the Case for Policy
Coordination in EMU”, in D. Adamski, F. Amtenbrink and J. de Haan, eds, The Cambridge
Handbook on European Monetary, Economic and Financial Market Integration. Cambridge
University Press, Cambridge, UK, in press.
Dubois, E. (2016), “Political Business Cycles 40 Years after Nordhaus”, Public Choice 166,
235–59.
14 CESifo Economic Studies, 2023, Vol. 00, No. 0
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

Eyraud, L., V. Gaspar and T. Poghosyan (2017), “Fiscal Politics in the Euro Area”, in V. Gaspar,
ed, Fiscal Politics. International Monetary Fund, Washington, DC, pp. 439–76.
Fata´s, A. and I. Mihov (2006), “The Macroeconomic Effects of Fiscal Rules in the US States”,
Journal of Public Economics 90, 101–17.
Fata´s, A. and I. Mihov (2012), “Fiscal Policy as a Stabilization Tool”, B.E. Journal of
Macroeconomics 12, 1–68.
Frankel, J. A., C. A. Ve´ gh and G. Vuletin (2013), “On Graduation from Fiscal Procyclicality”,
Journal of Development Economics 100, 32–47.
Furceri, D. and J. Jalles (2018), “Determinants and Effects of Fiscal Counter-Cyclicality”,
Ensayos Sobre Pol
ıtica Economica 36, 137–51.
Gali, J. and R. Perotti (2003), “Fiscal Policy and Monetary Integration in Europe”, Economic
Policy.18, 533–72.
Gavin, M. and R. Perotti (1997), “Fiscal Policy in Latin America”, in B. Bernanke and J.
Rotemberg, eds, NBER Macroeconomics Annual 1997, MIT Press, Cambridge, MA, pp. 11–72.
Golinelli, R. and S. Momigliano (2009), “The Cyclical Reaction of Fiscal Policies in the Euro
Area: The Role of Modelling Choices and Data Vintages”, Fiscal Studies 30, 39–72.
Gootjes, B. and J. de Haan (2022a), “Procyclicality of Fiscal Policy in European Union
Countries”, Journal of International Money and Finance 120, 102276.
Gootjes, B. and J. de Haan (2022b), “Do Fiscal Rules Need Budget Transparency to Be Effective?
”, European Journal of Political Economy 75, 102210.
Guerguil, M., P. Mandon and R. Tapsoba (2017), “Flexible Fiscal Rules and Countercyclical
Fiscal Policy”, Journal of Macroeconomics 52, 189–220.
Halland, H. and M. Bleaney (2011), “Explaining the Procyclicality of Fiscal Policy in Developing
Countries”, CREDIT Research Paper 11/09, The University of Nottingham, Centre for
Research in Economic Development and International Trade (CREDIT), Nottingham.
Henisz, W. (2002), “The Institutional Environment for Infrastructure Investment”, Industrial and
Corporate Change.11, 355–89.
Ilzezki, E. and C. A. Ve´gh (2008), “Procyclical Fiscal Policy in Developing Countries: Truth or
Fiction?”, NBER working paper 14191. National Bureau for Economic Research, Cambridge
(MA).
IMF (2005), “Can Fiscal Policy Stabilize Output?”, in IMF Fiscal Monitor 2005, International
Monetary Fund, Washington, DC (Chapter 2).
Jalles, J. T. (2018), “Fiscal Rules and Fiscal Counter-Cyclicality”, Economics Letters 170,
159–62.
Jalles, J. T. (2021), “Dynamics of Government Spending Cyclicality”, Economic Modelling 97,
411–27.
Kloosterman, R., D. Bonam and K. van der Veer (2022), “The Effects of Monetary Policy across
Fiscal Regimes”, De Nederlandsche Bank working paper 755. De Nederlandsche Bank,
Amsterdam.
Lane, P. (2003), “The Cyclical Behaviour of Fiscal Policy: Evidence from the OECD”, Journal of
Public Economics 87, 2661–75.
Lane, P. and A. Tornell (1999), “The Voracity Effect”, American Economic Review 89, 22–46.
McManus, R. and F. G. Ozkan (2015), “On the Consequences of Pro-Cyclical Fiscal Policy”,
Fiscal Studies 36, 29–50.
Nerlich, C. and W. H. Reuter (2015), “Fiscal Rules, Fiscal Space and Procyclical Fiscal Policy”,
ECB working paper 1872. European Central Bank, Frankfurt.
Potrafke, N. (2017), “Partisan Politics: The Empirical Evidence from OECD Panel Studies”,
Journal of Comparative Economics 45, 712–50.
Reuter, W. H., O. Tka
cevs and K. Vilerts (2022), “Fiscal Rules and Volatility: The Role of
Stabilising Properties and Compliance”, Empirica 49, 21–52.
CESifo Economic Studies, 2023, Vol. 00, No. 0 15
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

Talvi, E. and C. A. Ve´gh. (2005), “Tax Base Variability and Procyclical Fiscal Policy in
Developing Countries”, Journal of Development Economics 78, 156–90.
Taylor, J. B. (2000), “Reassessing Discretionary Fiscal Policy”, Journal of Economic Perspectives.
14, 21–36.
Thornton, J. (2007), “Explaining Procyclical Fiscal Policy in African Countries”, Journal of
African Economies.17, 451–64.
Woo, J. (2009), “Why Do More Polarized Countries Run More Procyclical Fiscal Policy? ”, The
Review of Economics and Statistics 91, 850–70.
Wyplosz, C. (2006), “European Monetary Union: The Dark Sides of a Major Success”, Economic
Policy 21, 208–61.
Appendix
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Output gap ∆Output gap
Figure A1. Cyclicality of fiscal policies over time, 32 OECD countries, 1986–2023 [share of the number
of countries (%)]. Note: The dark line denotes our measure of pro-cyclical fiscal policy, while the dotted
line shows the measure proposed by Bartsch et al. (2020).
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Small Large
Figure A2. Pro-cyclical fiscal policies in small and large countries over time, 1986–2023 (%). Note: See
notes to Figure 3.
16 CESifo Economic Studies, 2023, Vol. 00, No. 0
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Figure A3. Pro-cyclical fiscal policies first eleven euro-area countries, 1986–2023 (%). Note: See notes
to Figure 3.
Table A1. Definition and source of control variables
Variable Definition Source
Country size Large countries include the USA, the UK,
Japan, Germany, France, Italy, and
Spain (0–1 dummy variable)
Bartsch et al. (2020)
Economic
development
The logarithm of GDP per capita at pur-
chasing power parity (PPP)
OECD
EA Variable capturing the run-up effect of
the European Economic and Monetary
Union (0–1 dummy variable; only the
first 11 Member States of the euro
area are considered)
Own calculations
EMU Membership of the European Economic
and Monetary Union (0–1 dummy
variable; only European Union coun-
tries are considered)
Own calculations
Fiscal rules Index on the stringency of fiscal rules
(legal basis, coverage, monitoring,
flexibility, support)
IMF; https://www.imf.org/exter
nal/datamapper/fiscalrules/
map/map.htm
Institutional
quality
Index on the quality of government insti-
tutions (rule of law, regulatory quality,
government effectiveness, voice and
accountability)
World Governance Indicators
of the World Bank; https://
databank.worldbank.org/
source/worldwide-govern
ance-indicators
Private credit Domestic private credit (% of GDP) World Bank
Financial
development
Index on the development of financial
markets and financial institutions
based on depth, efficiency, and access
IMF
Capital account
openness
Chinn–Ito index measuring a country’s
degree of capital account openness
Updates of Chinn and Ito
(2006)
(continued)
CESifo Economic Studies, 2023, Vol. 00, No. 0 17
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

Table A1. (continued)
Variable Definition Source
Trade openness Sum of imports and exports (% of GDP) World Bank
Government bond
yields
Interest rates on government bonds
maturing in 10 years
OECD
Government size General government spending (% of
GDP)
OECD
Economic upswing Positive output gap (0–1 dummy
variable)
OECD
Corruption Index on the control of corruption World Governance Indicators
of the World Bank
Government
ideology
A 0–1 dummy variable that is zero if a
left-wing government is in place and
one when a right-wing/centrist govern-
ment is in power.
DPI 2020 database
Margin of majority Fraction of seats held by the government DPI 2020 database
Government
fragmentation
The probability that two deputies picked
at random from among government
parties will be of different parties
CPDS database (Armingeon
et al., 2022)
Checks and
balances
The number of legislative veto players in
the political system
DPI 2020 database
Output volatility The standard deviation of economic
growth
OECD
Elections Elections years (0–1 dummy variable) Updated version of CPDS data-
base (Armingeon et al.,
2022)
Income inequality Gini index World Bank
18 CESifo Economic Studies, 2023, Vol. 00, No. 0
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

Table A2. Summary statistics and t-tests
Variable l0l1r0r1N0N1PðT
jj
>t
jj
;H0:ðl0l1Þ¼0Þ
Country size 0.490 0.450 0.500 0.498 816 260 0.259
Economic development 0.487 0.468 0.500 0.499 499 556 0.530
EA 0.533 0.451 0.501 0.498 122 275 0.132
EMU 0.509 0.467 0.501 0.500 137 338 0.416
Fiscal rules 0.479 0.478 0.501 0.500 455 557 0.961
Institutional quality 0.500 0.474 0.501 0.500 374 416 0.479
Private credit 0.471 0.446 0.500 0.498 367 377 0.481
Financial development 0.519 0.446 0.500 0.498 455 525 0.023
Capital account openness 0.533 0.462 0.500 0.499 240 710 0.056
Long-term interest rates 0.433 0.519 0.496 0.500 497 505 0.011
Trade openness 0.477 0.477 0.500 0.500 512 543 0.989
Government spending 0.465 0.489 0.499 0.500 527 528 0.441
Good times 0.484 0.475 0.500 0.500 630 446 0.776
Control of corruption 0.511 0.469 0.501 0.500 374 416 0.240
Elections 0.477 0.491 0.500 0.501 789 287 0.669
Government fragmentation 0.469 0.493 0.500 0.500 497 479 0.458
Checks and balances 0.493 0.475 0.501 0.500 290 687 0.596
Output volatility 0.459 0.503 0.499 0.500 545 531 0.148
Income inequality 0.483 0.519 0.501 0.501 271 270 0.414
Note: The table shows the summary statistics of the results shown in Figures 3–5and the P-values of the t-tests
of whether the percentage of fiscal pro-cyclicality is different between groups.
CESifo Economic Studies, 2023, Vol. 00, No. 0 19
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023

Table A3. Summary statistics and t-tests using DGap as measure of the economy
Variable l0l1r0r1N0N1PðT
jj
>t
jj
;H0:ðl0l1Þ¼0Þ
Country size 0.444 0.396 0.497 0.490 816 260 0.179
Economic development 0.449 0.415 0.498 0.493 499 556 0.274
EA 0.492 0.455 0.502 0.499 122 275 0.494
EMU 0.434 0.476 0.497 0.500 137 338 0.359
Fiscal rules 0.422 0.458 0.494 0.499 455 557 0.254
Institutional quality 0.500 0.394 0.501 0.489 374 416 0.003
Private credit 0.469 0.401 0.500 0.491 367 377 0.061
Financial development 0.490 0.400 0.500 0.490 455 525 0.005
Capital account openness 0.483 0.428 0.501 0.495 240 710 0.137
Long-term interest rates 0.396 0.451 0.490 0.498 497 505 0.078
Trade openness 0.418 0.444 0.494 0.497 512 543 0.397
Government spending 0.391 0.472 0.488 0.500 527 528 0.008
Good times 0.425 0.442 0.495 0.497 630 446 0.595
Control of corruption 0.484 0.409 0.500 0.492 374 416 0.034
Elections 0.422 0.460 0.494 0.499 789 287 0.268
Government fragmentation 0.429 0.459 0.495 0.499 497 479 0.335
Margin of majority 0.436 0.450 0.496 0.498 502 476 0.675
Government ideology 0.428 0.455 0.496 0.498 311 547 0.436
Checks and balances 0.441 0.443 0.497 0.497 290 687 0.974
Output volatility 0.378 0.488 0.485 0.500 545 531 0.000
Income inequality 0.488 0.467 0.500 0.500 271 270 0.761
Note: The table shows the summary statistics of the results shown in Figures 3–5and the P-values of the t-tests
of whether the percentage of fiscal pro-cyclicality is different between groups.
20 CESifo Economic Studies, 2023, Vol. 00, No. 0
Downloaded from https://academic.oup.com/cesifo/advance-article/doi/10.1093/cesifo/ifad001/7100837 by guest on 05 April 2023