1

Does fiscal decentralization mitigate the adverse effects of corruption on public

deficit?

Daniel Oto-Peralías*, Diego Romero-Ávila and Carlos Usabiaga

Pablo de Olavide University, Spain

Abstract

The current economic crisis has led several rich countries to experience severe fiscal

deficits. Among other factors responsible for the situation, corruption is considered harmful

to public finances and appears closely related to fiscal deficits. This paper opens a new

avenue in addressing the effects of corruption on public deficits through fiscal

decentralization. Focusing on a sample of 31 OECD countries over the period 1986-2010,

we find that fiscal decentralization contributes to mitigating the perverse effects of

corruption in public deficits. In addition, our findings indicate heterogeneity in the effect of

fiscal decentralization, since it appears related to lower deficits in countries with higher

levels of corruption, but not in less corrupt countries. Therefore, the results suggest that

bringing the government closer to the people in relatively corrupt countries may lead to a

more responsible fiscal management.

Keywords Corruption, Public Deficit, Fiscal Decentralization, OECD

JEL Classification D73, H62, H71, H72

-----------------------------------------------------

Acknowledgements: The authors acknowledge financial support from the Spanish Ministry of Science and Technology through grant ECO2009-13357, from the Spanish Ministry of Economics and Competitiveness through grant ECO2012-35430 and from the Andalusian Council of Innovation and Science under Excellence Project SEJ-4546.

* Corresponding Author: Daniel Oto-Peralías, Universidad Pablo de Olavide, Departamento de Economía, Métodos Cuantitativos e Historia Económica, Carretera de Utrera, Km. 1, Sevilla, Spain. E-mail: dotoper@upo.es.

2

1. Introduction

Nowadays, achieving sustainable public finances is a first order concern in rich countries. The current

economic crisis has led several countries to experience severe fiscal deficits, going so far as to make it

impossible for some of them to obtain funding without international bailouts. Among other factors

responsible for this situation, corruption is considered an important institutional feature which is

harmful to public finances and appears closely related to fiscal deficits. In the view of the World Bank

(2012a), corruption “reduces the effectiveness of public administration and distorts public expenditure

decisions, [...] erodes the rule of law and harms the reputation of and trust in the state”. More

specifically, corruption leads to adverse budgetary consequences by decreasing state revenue and

promoting wasteful spending. Corruption practices are associated with tax evasion, unofficial

economy, illegal customs administration, irregular procurements, thefts and bribes or “white elephant”

investment projects (World Bank, 2012b).

As pointed out by Kaufmann (2010), it is interesting to note that, contrary to common belief, there

exist large differences regarding the presence of corruption among industrialized countries. To get an

overview of these differences, focusing on our sample of OECD countries, the Global Corruption

Barometer 2006 of Transparency International reports bribe payments values ranging between 1

percent (Finland, Sweden and Switzerland) and 17 percent (Czech Republic and Greece).1 With

respect to the consequences on public finances, Kaufmann (2010) observes a strong relationship

between corruption and fiscal deficits in rich countries. An illustration of this relationship is presented

in Figure 1. 2

[Insert Figure 1 about here]

Bearing this in mind, this paper opens a new avenue in accounting for the fact that the adverse effects

of corruption on public deficits may be mitigated through fiscal decentralization. The intuition behind

this proposition is that when the government does not work well, by misusing public office for private

gain, a close citizen control of politicians is particularly necessary. The monitoring of political activity

–in general– and fiscal management –in particular– can be facilitated by bringing the government

closer to the people, which leads to higher information about local affairs and a higher degree of

1 Literately, the question included in the Barometer is as follows: “In the past 12 months, have you or anyone living in your household paid a bribe in any form?” (Transparency International, several years). 2 See below for the explanation of the variables employed.

3

democracy and accountability. The same logic suggests that if corruption is absent, to decentralize

may be less necessary in this regard. In addition, it may be argued that in a context of rampant

corruption, it is preferable to have several levels of government rather than an omnipotent and opaque

central government that appears to be very far from citizen control.

When one observes decentralization across OECD countries, the differences are striking. Considering

the amount of subnational tax revenue as a percentage of total general government tax revenue, we

find great variability within our sample over the period 1986-2010. The range runs from countries like

Greece, with a value lower than one percent, to countries like Canada, with a value near 50 percent.

Casual observation supports our argument. Greece, a country relatively corrupt, presents both the

lowest level of fiscal decentralization and the highest deficit in the analyzed period. This stands in

stark contrast to the experience of countries like Poland, which reports higher decentralization and

lower deficits, despite having at least similar levels of corruption. The comparison of the two countries

suggests a connection between more fiscal decentralization and fewer deficits. In contrast, among less

corrupt countries, decentralization appears unrelated to public deficit. Thus, the Netherlands and

Canada present similar levels of corruption and similar records of deficits, the former being much less

decentralized than the latter. Table 1 summarizes the data of the four countries mentioned.3

[Insert Table 1 about here]

These observations lead us to systematically study the relationship between corruption, fiscal

decentralization and public deficit. Using a sample of 31 OECD countries, we show that the

relationship between corruption and public deficit is nonlinear and varies with the degree of fiscal

decentralization. Thus, we find that the effect of corruption on increasing fiscal deficit goes down as

the level of decentralization rises. In addition, we observe that fiscal decentralization is related to

lower deficit only in countries with higher levels of corruption, but not in less corrupt countries.

Therefore, the central message this paper conveys is that fiscal decentralization mitigates the adverse

effects of corruption on public deficit. We adopt several econometric approaches to provide empirical

3 In general, the literature on fiscal decentralization predicts ambiguous effects on public deficits. On the one hand, to decentralize means bringing the government closer to the people and encourages competition among subnational jurisdictions. This leads to better information about local affairs, higher accountability, more efficient provision of public services, responsible fiscal management and moderate size of government (Oates, 1999; Brennan and Buchanan, 1980). But decentralization is considered as a “double-edged sword”. Several problems have been identified, such as the risk of duplicating functions and wasting resources, coordination failures, a “race to the bottom” due to interjurisdictional competition –with the subsequent erosion of government revenue sources–, “soft budget constraints” and the lower expertise of local governments (Prud’homme, 1995; Weingast, 2009).

4

evidence supporting our main argument. First, we estimate cross-section and panel regressions for the

period 1986-2010 by using an interaction model where public deficit is explained through corruption,

fiscal decentralization and an interaction term of both variables. Second, the sample is divided into

three groups according to the level of corruption so that we can examine whether decentralization is

related to lower deficit in each of them. For the first purpose, we use ordinary least squares (OLS) and

the system GMM estimator and, for the second, fixed effects models. The results hold for both revenue

and expenditure indicators of decentralization and they are fairly robust to different specifications.

The remainder of the paper is organized as follows. Section 2 reviews the relevant literature on

corruption, fiscal federalism and public deficit, as well as outlines possible hypotheses about the

impact of the interaction between corruption and decentralization on public deficits. Section 3 presents

the methodological approach and takes a first look at the data. Section 4 presents the results from the

main regression analysis. Section 5 provides several robustness checks as well as a complementary

analysis that divides the sample into three groups according to the average corruption level and

focuses on within-country variation. Section 6 puts forward some policy implications and concludes.

2. Theoretical and empirical background

Our research is placed between two well-differentiated branches of the institutional literature, namely,

the consequences of corruption –on one side– and decentralization –on the other– in the economy and

governance. We first review both branches separately and we then adopt an integrative view in order

to interpret the interaction between corruption and decentralization.

For more than a decade the consequences of corruption have been studied profusely by economists

and political scientists. A seminal contribution on the effects of corruption on the economy is Mauro

(1995, 1997), who finds that corruption reduces investment and hence economic growth. Since then,

many authors have found evidence on the negative impact of corruption on different areas of

economic activity, such as foreign direct investment, countries’ infrastructures, investment, public

spending effectiveness or sustainable development (see, among others, Wei, 1997a and 1997b; Tanzi

and Davoodi, 1998; Aidt et al., 2008; Rajkumar and Swaroop, 2008; Aidt, 2009). Regarding public

finances, the World Bank (2012b) stresses that corruption is harmful to state coffers by decreasing

state revenue and promoting wasteful spending. Corruption encourages tax evasion, tax regressivity

and the unofficial economy. It is also associated with corrupt procurement with cost overrun, cases of

theft, “white elephant” investment projects and purchasing of political support. High level of

5

corruption reduces tax revenue particularly from social security tax and –to a lower extent– sales tax.

Extrabudgetary accounts (with lower political and administrative controls) and goods and services

provided below market prices are also fertile grounds for corruption to flourish (Tanzi, 1999). In a

study of US states, Depken and Lafountain (2006) find that corruption is related to lower bond rating,

which implies the payment of a premium for state bonds.

Focusing on industrialized countries, Kaufmann (2010) lists a number of channels whereby corruption

can affect public finance, which can be divided into direct and indirect channels. The former are a) the

erosion of tax revenue through evasion and ineffective tax collection institutions and b) the increase of

public expenditures associated with inflated bureaucracies, expensive public investments and

inefficient composition of public spending. The indirect channels are a) the rising cost of debt service

and the lack of transparency of the financial sector, b) corrupt data of financial and national statistics,

which leads to uncertainty and mistrust in the financial markets, c) the shadow economy, which leads

to an increase in official tax rates to compensate for the overall fall in tax revenues, and d) lower

productivity, competitiveness and growth (by raising –for instance– the cost of doing business), which

are the ultimate sources of government revenue. Kaufmann (2010) finds that corruption is strongly

related to fiscal deficits in rich countries. One standard deviation improvement in the corruption

indicator leads to a 3.5 percentage point reduction in fiscal deficits. Besides Kaufmann’s work, very

little has been done to investigate the relationship between corruption and public finance in rich

countries. In this respect, we provide further evidence showing that corruption increases fiscal deficits

in rich countries.

Another extensively studied topic over the last two decades is the impact of decentralization on

economic activity and governance. Nowadays, we are witnessing a global trend of devolution process

(see, for instance, Rodríguez-Pose and Gill, 2003). Many states have conducted and are still involved

in decentralization processes in order to achieve economic, social and democratic goals. Although

initially decentralization initiatives were seen with optimism, scholars soon began to doubt about the

merit of such endeavours. The attitude change was due, at least partially, to the dismal performance of

subnational levels in some countries (Oates, 2008). Argentina and Brazil suffered severe fiscal crisis

largely motivated by irresponsible behavior of subnational governments (Dillinger and Webb, 1999).4

4 We focus particularly on fiscal decentralization since our interest lies in the distribution of fiscal competences among different levels of government. Nevertheless, it is important to keep in mind that the different facets of decentralization are related among them. For example, vertical fiscal imbalances, –i.e., dependence on financial resources from higher levels of government–, endanger the political autonomy of subnational governments (Weingast, 2009).

6

The literature on fiscal federalism provides arguments both in favor and against decentralization. First

of all, to decentralize means bringing the government closer to the people. This has many implications,

namely, a higher level of knowledge and information about local affairs, leading to a more efficient

and better provision of public services (Oates, 1999), and higher degree of democracy and

accountability, allowing the control of corrupt behavior. Secondly, it has been argued that the

existence of a number of sub-national jurisdictions encourages competition among them in order to

satisfy people’s preferences. The widely known “Tiebout model” supports the fact that mobile

households select those jurisdictions that better match with their preferences about public goods and

taxes by voting “with their feet” (Tiebout, 1956). Officials and politicians will be compelled to a

responsible management of fiscal resources and to provide valuable public services (Oates, 1999).

Also, in a related fashion, a plurality of subnational governments is considered to favor the functioning

of economic markets (“market-preserving federalism” –Weingast, 1995) and limit the size of the

“Leviathan” (Brennan and Buchanan, 1980). For all these reasons, second generation fiscal federalism

models stress the important role played by own revenue generation –tax autonomy– by subnational

governments (Careaga and Weingast, 2003; Rodden, 2003).5

Regarding the arguments against fiscal decentralization, interjurisdictional competition can generate a

“race to the bottom” because it creates incentives for subnational governments to reduce tax rates and

regulation standards in order to attract business to the jurisdiction. This erodes their revenue sources,

thus causing serious fiscal imbalances. On the other hand, “soft budget constraints” and an excessive

dependence on intergovernmental transfers are related to irresponsible fiscal management (Weingast,

2009).6 Additionally, as mentioned by Treisman (2002), it has been argued that multi-tier governments

are more likely to duplicate functions and waste resources. Likewise, when economies of scale exist,

subnational governments are less efficient in providing public goods and services if they lack the

adequate size (Rodríguez-Pose and Ezcurra, 2011). Other negative aspects of decentralization include

coordination problems due to the presence of a high number of veto players and the lower capacity

and skills of local governments (Prud’homme, 1995).

5 In favor of federalism, it has been also said that it allows for a higher degree of institutional and policy experimentation. A federal system offers the possibility of public policy innovation by a decentralized process of “learning-by-doing” in different circumscriptions (Oates, 1999). In the US there are several examples of federal policies that were designed and used formerly in subnational levels –for instance, the unemployment insurance, the taxation of gasoline and the regulation of emissions standards for motor vehicles (Oates, 2008). 6 It is emphasized the importance of revenue generation by state and local governments. In a review of the second generation models of fiscal federalism, Weingast (2009) asserts that “subnational governments that raise a substantial portion of their own revenue tend to be more accountable to citizens, to provide market-enhancing public goods, and to be less corrupt” (p. 283).

7

Several empirical works have addressed the question of the relationship between decentralization and

public deficits. For a sample of 30 developed and developing countries during the period 1970-1995,

De Mello (2000) finds subnational tax autonomy to be associated with larger deficits, which he relates

to coordination failures in intergovernmental fiscal relations. Focusing on a sample of 17 federations,

Rodden and Wibbels (2002) provide evidence that fiscal decentralization is associated with smaller

overall deficits, particularly when subnational governments have wide autonomy over taxation. In

contrast, public deficits increase as subnational governments rely more on intergovernmental transfers.

Shah (2006) comes to the conclusion that decentralized fiscal systems permit (or facilitate) to a higher

extent improved macroeconomic governance compared to centralized systems. In general, he finds

fiscal decentralization to be related to better fiscal performance.7 In addition, Baskaran (2010) explores

the relationship between fiscal decentralization and public debt within a panel of 17 OECD countries

from 1975 to 2001. His findings indicate that expenditure decentralization reduces public debt,

whereas tax decentralization and vertical fiscal imbalances are not significantly related to it. Baskaran

(2012) analyzes the influence of sub-national tax autonomy in public deficits with a sample of 23

OECD countries over the last quarter a century. Sub-national tax autonomy shows a U-shaped effect,

so the ‘average OECD country’ could reduce the fiscal deficit by decreasing sub-national tax

autonomy.

We also find a paper by Neyapti (2010) that is related to our analysis. Her study shows that both

expenditure and revenue decentralization reduce budget deficits for a sample of developing and

developed countries over the period 1980-98. In addition, she uses an interaction model providing

some preliminary evidence that the effect of fiscal decentralization on reducing budget deficits

increases with population size and that the effect of expenditure decentralization is reduced when

governance improves. Our analysis differs from Neyapti’s in several important respects. First, unlike

her sample of 16 countries that only includes nine OECD countries (while the rest are developing

countries as diverse as Bolivia, Brazil, Colombia, Iran, Malaysia, Mexico and South Africa), our

dataset is based on a more homogeneous sample comprising 31 OECD countries. The advantage of

doing so is that most OECD countries are characterized by having democratic regimes and relatively

educated populations, and the risk of omitted variables bias and measurement errors is attenuated.8

7 In Shah’s view, that result is expected since decentralized fiscal systems imply clarity in the role of decision-making units, transparency in the rules and a careful institutional design thought to achieve fair play and restrict rent seeking behavior. 8 So, taking as reference the year 2000, all of the 31 OECD countries of our sample are “Free” according to the Freedom House’s classification of political status. Regarding the level of education, which is important in order to make an effective control of government activity feasible, the secondary school enrolment is above 85% in all

8

This allows us to better identify the effects of fiscal decentralization and its interaction with corruption

on fiscal deficits, thereby enabling us to provide some policy implications and advice better suited for

industrialized countries. Second, by taking the period covering the years between 1986 and 2010 we

are able to extend Neyapti’s time span by 12 years, which entails the most recent period in which tax

and expenditure decentralization has intensified (Stegarescu, 2005).

Third, while this author treats fiscal decentralization (as well as the other regressors) as exogenous, we

allow all explanatory variables to be endogenous with respect to fiscal deficits.9 We deal with this

issue by using the system GMM estimator of Arellano and Bover (2005) with both annual and 5-year

average data (which should also help to correct for real business cycle effects) versus the pooled OLS

estimator of Neyapti that fails to accommodate the presence of endogeneity in the regressors. Fourth,

whereas she interacts –in only one model– fiscal decentralization with a broad measure of governance

which includes a wide variety of factors such as control of corruption, rule of law, political instability,

government efficiency, voice accountability and regulatory quality, our paper focuses the whole

empirical analysis on corruption and its interaction with fiscal decentralization. We do so because

corruption is the aspect of governance that is generally emphasized in both the theoretical and

empirical literature on fiscal decentralization and fiscal outcomes.10 Fifth, we conduct extensive

robustness checks where we introduce many additional control variables, test the presence of outliers

and even use alternative empirical approaches. Not surprisingly, our results differ substantially from

Neyapti for the revenue decentralization measure. Whereas we find robust evidence that tax

decentralization leads to lower deficits only in countries with high corruption, her evidence points to

the fact that “governance [does] not seem to influence the effectiveness of revenue decentralization”

(Neyapti, 2010, p. 161).

After reviewing the literature on the consequences of corruption and decentralization in public finance,

we raise the question of what the effect of their interaction is. Is the effect of corruption independent

of the level of decentralization? Does fiscal decentralization mitigate or aggravate the adverse effects

of corruption on public deficit? What result is to be expected from that interaction? An a priori answer

is necessarily ambiguous. On the one hand, decentralization may exacerbate public finance

countries (World Bank, 2011). In addition, common forces such as economic integration measured by trade and financial openness are found to increase fiscal decentralization in the OECD (Stegarescu, 2009). 9 It is not difficult to show how, for instance, GDP growth affects negatively fiscal deficits and how the latter in turn reduces economic growth by inducing spending cuts and tax increases. 10 In addition, while the governance indicator is only available since 1996, our corruption measure from the International Country Risk Guide is available for the whole period under scrutiny (1986-2010).

9

mismanagement in a corrupt context. Local government may be captured by local elites and corrupt

politicians without proper citizen control. Political patronage is another pathology of corrupt

governments that may be aggravated at local levels, with negative consequences for spending and

revenue decisions. More actors involved in fiscal management implies that the pie must be divided

into more pieces, which may lead to more private appropriation of public funds. In addition,

coordination problems and fiscal imbalances may be exacerbated when corruption exists, since budget

monitoring and public finance transparency tend to be lower. Hence, a plausible hypothesis is that

when the government is corrupt, many government levels will worsen rather than improve the

situation.11

However, a more optimistic view is possible. A corrupt government is characterized by opaqueness

and lack of accountability, whose decisions are taken without citizen control. In this case, central

government is a Leviathan that extracts rents from population and uses them for private gain of

politicians and officials. Decentralizing fiscal powers gives citizens the opportunity to increase the

control of public money management. Arguably, if the central government is corrupt, fiscal

decentralization is more necessary as an institutional mechanism that helps to prevent the irresponsible

management of public funds. A government closer to the people is a government subject to more

oversight and more effective accountability. This is because citizens are more interested and informed

about local affairs, thereby being better able to punish irresponsible political behavior, for instance,

through subnational and local elections. By contrast, when government is transparent and open, and

politicians and public employees honest and upright, then the central government may function well

without need for decentralization. Although it may be argued that decentralization –as well as

democracy– is always desirable, in a context of low corruption it is less necessary because the

government performs properly and does not misuse public funds. In this case, citizen control is not so

11 A related question is whether decentralization increases corruption. Fisman and Gatti (2002) obtain a strong and significant effect of government expenditure decentralization on reducing corruption. Along similar lines, Ivanyna and Shah (2010), focusing on the local government level for a large sample of countries, find that local fiscal decentralization has a significant negative effect on the incidence of corruption. This would accord with the prediction of theoretical models such as the ones by Shleifer and Vishny (1993) and Kotsogiannis and Schwager (2006) for which greater decentralization limits corruption by raising competition among subnational governments. Arikan (2004) provides less clear-cut evidence of a reduction in corruption due to higher decentralization, since only about half of the decentralization coefficients appear statistically significant. In contrast, using measures of corruption experiences and a large sample of countries around the world, Fan et al. (2009) point to the danger of uncoordinated rent-seeking when government structures increase in complexity. In this sense, a larger number of tiers and a larger local staff –“with pockets to fill” – increase the chances of “overgrazing” in the rents of public office. In conclusion, these works show that there is no consensus about the effect of decentralization on corruption.

10

essential, and one can expect the positive effect of fiscal decentralization on reducing budget deficits

to be lower in a transparent country.12

Since there is theoretical ambiguity about the effect of the interaction between corruption and

decentralization on fiscal deficits, an empirical inquiry is warranted. This is what we do in the rest of

the paper.

3. Methodology and data

We select a sample of 31 OECD countries, which includes all current members except Chile, Mexico

and Turkey, due to missing data. The use of a sample of rich countries has several advantages, such as

greater data availability and lower heterogeneity across countries. This last aspect implies that we are

comparing relatively homogeneous countries, with well-developed economies and democratic

regimes. When analyzing the mechanism by which decentralization affects governance, it is important

to keep in mind the features characterizing the countries included in our sample (high average levels of

education and income, political freedom, levels of corruption relatively lower than non-OECD

countries, and so on). Also, this greater sample homogeneity leads to more meaningful results and

reduces the risk of omitted variable bias. The period under scrutiny spans from 1986 to 2010, since

available data on corruption begin in the mid-eighties. The source of institutional data is the

International Country Risk Guide (ICRG) produced by the Political Risk Services Group (PRS Group,

2012).

Firstly, we estimate an interaction model where public deficit is explained by corruption, fiscal

decentralization, the interaction between these two and several control variables. We calculate cross-

section regressions with OLS by averaging the variables over the whole period. This can be thought of

as the analysis of the long-term relationship between the variables under study. We also estimate panel

regressions in order to exploit the time variation of the data, using both annual and five-year averaged

data. In this case, we use pooled OLS and the system GMM estimator. Secondly, we divide the sample

into three groups based on the average level of corruption. This exercise can be proven meaningful

since corruption –as many other institutional variables– is remarkably persistent over time.13 With

12 Likewise, one can argue that other disciplining mechanisms working in decentralized governments such as jurisdictional competition play a more important role when corruption is higher. We will insist on this point in Section 5.3. 13 In order to check the persistence of corruption over time, we can divide the sample into two periods (1984-1997 and 1998-2010) and calculate the average value of corruption for each country in both periods (using ICRG data). The correlation between the two periods is 0.87.

11

countries classified into three groups, we ask whether there are significant cross-country differences

regarding the effect of decentralization on public deficits. For that purpose, we use a fixed effects

model, which controls for all time-invariant characteristics and exploits within country variations. In

this way, corruption is held constant and instead we focus on the variation in fiscal decentralization.

Regarding the measure of corruption, we concentrate on public corruption and adopt a widely

extended definition of corruption as the misuse of public office for private gain (see, among others,

Tanzi, 1999). Our main measure is a perception-based indicator from the ICRG (hereafter, ICRG

corruption). This indicator assesses, on a scale of 0 to 6, the presence of corruption within the political

system.14 It measures different facets of corruption, such as financial corruption (requirements of

special payments and bribes), political patronage, “wire-pulling”, secret party funding, and ties

between politics and business (PRS Group, 2012). We prefer this variable over other alternatives due

to the greater time span availability. Unlike our corruption measure, the indicators Control of

Corruption from the World Bank Governance Indicators (Kaufmann et al., 2010) and Corruption

Perceptions Index from Transparency International are only available since 1996 and 1995,

respectively. Experience-based indicators from comparable international surveys are much more

limited both in the time dimension and in the sample of countries covered. Correlations among

perception-based indicators are high, indicating that they are measuring more or less the same thing;

while correlations among these indicators and experience-based indicators are lower.15

With respect to fiscal decentralization, our preferred indicator is subnational (state/regional and local)

tax revenue as a percentage of total general government tax revenue (hereafter, subnational tax

revenue), which is a measure of revenue decentralization and comes from the OECD Fiscal

Decentralisation Database (OECD, 2012a). This indicator is interesting because it focuses on tax

revenue generated at subnational levels, rather than other sources of revenue such as

intergovernmental transfers, which depend on the central government. It has been argued that

decentralization based on own revenue leads to more accountability, less corruption, more market-

enhancing public goods and more political autonomy (Weingast, 2009). Nevertheless, our indicator is

imperfect, since it does not reflect the degree of autonomy of subnational governments in deciding the

14 Country-specific annual scores are computed as the average over the 12 months for each year (PRS Group, 2012). This makes the variable take values over the continuum between 0 and 6. We rescale the variable so that higher scores imply more corruption. 15 The correlation of ICRG corruption with Control of Corruption is 0.82 (N = 337) and with Corruption Perception Index 0.76 (N = 478). The correlation with the variable bribe payments (percentage of population paying bribes in the last year), an experience-based indicator from the Global Corruption Barometer, is 0.53 (N = 134).

12

tax characteristics (tax base, rates, reliefs, etc.). Stegarescu (2005) –following the classification

proposed by the OECD (1999)– provides a measure of subnational tax autonomy (that spreads until

2001) by selecting only autonomous own tax revenue of subnational governments, that is, those for

which sub-central government determines the tax rate and/or the tax base. For robustness purposes, we

update Stegarescu’s best measure of tax decentralization (i.e. TD1) by using the information about tax

autonomy of state and local governments provided by the Fiscal Decentralization Database of the

OECD for the years 2002, 2005 and 2008 (OECD, 2012a). Remarkably, the subnational tax revenue

and TD1 indicators are strongly correlated.16 In addition, in order to test whether the effect of

expenditure decentralization is different from that of tax decentralization, we also estimate regressions

using “subnational government expenditure as a percentage of total general government expenditure”

–hereafter, subnational govern. expend. (OECD, 2012a).

Our dependent variable is general government deficit over GDP obtained from the Economic Outlook

of the OECD (2012b). We do not use subnational or central deficits because what matters is the

behavior of the whole political system. Moreover, irresponsible fiscal policies of subnational

governments may not be associated with higher subnational deficits if the central government issues

debt and increases transfers to subnational units. As Treisman (2002) notes, it makes no sense to

advocate decentralization on the basis of better local government performance when it is offset by a

deterioration of central government accounts.

In all the estimations we always include a basic set of control variables. Total government revenue

over GDP is included to control for countries’ capacity to collect taxes and for government size. The

expected sign is ambiguous since greater revenue capacity allows covering expenditures, but large

governments are related to economic inefficiencies and distortions. Real per capita GDP growth and

unemployment rate control for business cycle effects. We expect economic growth to reduce and

unemployment rate to increase fiscal deficits. The logarithm of per capita GDP is also included to

control for the level of economic development. The effect of this variable is uncertain since wealthier

countries are associated with a higher demand for goods and services provided by the government (the

Wagner’s law) but also these countries can afford more effective tax administration and are less likely

to be affected by tax evasion. Finally, population is introduced due to the important variability in the

16 The correlation between subnational tax revenue and Stegarescu’s indicator of tax decentralization is 0.874 (N = 357). After updating the indicator, the correlation is almost identical (0.875, N = 403).

13

size of the countries in the sample and because it probably affects public finances (scale effects, for

instance) and promotes decentralization.17

Before starting the regression analysis, Table 2 provides a first look at the relationship between

corruption, fiscal decentralization and public deficit. Countries are classified into three groups

according to their level of corruption.18 Within each group, countries are separated on the basis of their

degree of tax decentralization. The entries represent the average values over the period 1986-2010.

The aggregate data for each group (rows with bold letters) reveal a positive association between

corruption and public deficit, which fits with the relationship depicted in Figure 1 (see Introduction).

We see that more corrupt countries exhibit higher deficits, being the difference relative to less corrupt

countries of nearly 4 percentage points (3.98 vs. 0.08). In each corruption group, one observes that

corruption affects deficits differently depending on tax decentralization. Considering the ‘high

corruption group’, less decentralized countries suffer higher deficits than more decentralized ones,

with a difference of 3.5 percentage points.19 In the ‘intermediate corruption group’, less decentralized

countries also suffer higher deficits, but the difference with respect to more decentralized ones is lower

(0.68 percentage points). Finally, focusing on the ‘low corruption group’, we observe a very different

pattern. In this case, less decentralized countries experience lower deficits than those more

decentralized, the difference being 2.3 percentage points. Therefore, this preliminary evidence

indicates heterogeneity in the relationship between decentralization and public deficit, depending on

the level of corruption. Fiscal decentralization is related to lower public deficits in a context of

relatively high corruption, but not in a context of low corruption.

[Insert Table 2 about here]

4. Regression analysis

4.1 Cross-section regressions analysis

We estimate in this section a basic interaction model of the following form:

17 The definitions and sources of the variables are presented in Appendix I. The descriptive statistics are provided in an unpublished appendix available from the authors upon request. 18 We calculate the average value of our main indicator of corruption, ICRG corruption, by using all available data (the period 1984-2010). Then, countries are divided by the 33th and 66th percentiles of this variable. More details are provided in subsection 5.2. 19 Note that this difference is not simply due to a higher level of corruption among less decentralized countries. In fact, the level of corruption is higher among more decentralized countries (2.67 vs. 2.20).

14

iiiiii Xdecentrcorruptiondecentrcorruptiondeficit εδβββα +⋅+×⋅+⋅+⋅+= 321 (1)

where deficiti is general government deficit, α is a constant term, corruptioni is our main corruption

indicator (ICRG corruption), decentri represents fiscal decentralization, corruption×decentri stands for

the interaction term between the last two variables, Xi is a vector of control variables and εi is the error

term.20 All the variables are included as the average over the period 1986-2010.

Table 3 reports results from cross-section regressions estimated with OLS. Column 1 shows the basic

model without fiscal decentralization variables. Corruption is significantly related to higher fiscal

deficits. An increase by one standard deviation in corruption (0.89) is associated with an increase of

1.15 percentage points in public deficit. The next column adds subnational tax revenue, which enters

with a negative and significant coefficient. Now, corruption appears statistically insignificant. But

since we suspect the existence of heterogeneity in the relationship between corruption and public

deficit depending on the degree of fiscal decentralization, column 3 includes the interaction term

between corruption and tax decentralization. In this model the interpretation of the coefficients is very

different. The coefficient on corruption shows the effect on public deficit when subnational tax

revenue is equal to zero, while the negative coefficient on the interaction term indicates that the effect

of corruption on fiscal deficit decreases as tax decentralization increases. Therefore, we must conduct

a conditional interpretation of the marginal effect of corruption for each value of tax decentralization

(see Brambor et al., 2006). The marginal effect is given by:

decentrcorruption

deficit⋅+=

∂

∂31 ββ

(2)

where β1 is the coefficient on corruption and β3 the coefficient on the interaction term. It is also

necessary to properly calculate the standard error of the marginal effect, which is given by (see Aiken

and West, 1991):21

)ˆˆcov(2)ˆvar()ˆvar(ˆ 3132

1 ββββσ ⋅⋅⋅+⋅+=∂

∂ decentrdecentrcorruption

deficit

(3)

At the bottom of Table 3, we calculate the marginal effects and standard errors of corruption for

different values of tax decentralization (percentiles 10th to 90th). The marginal effect of corruption on 20 Note that the model includes all constitutive terms of the interaction. In order for coefficients to be meaningful, Brambor et al. (2006) show that interaction models must comprise all constitutive terms of the interaction. 21 In an interaction model, for interpretation purposes, the relevant standard errors are those calculated in this way rather than the standard errors of the constitutive or interaction terms (Brambor et al., 2006).

15

public deficit is positive and significant only when tax decentralization is low. For values equal or

greater than the 50th percentile, that is, subnational tax revenue equal to 10.5%, the marginal effect is

no longer significant and even the sign changes (see column 3). Regarding the control variables,

economic growth and unemployment show the expected negative and positive signs, both variables

being statistically significant. Total government revenue as a percentage of GDP presents a negative

but insignificant coefficient, as it is the case also for per capita GDP. Population appears with a

positive coefficient, sometimes statistically significant.

[Insert Table 3 about here]

It is interesting to note that we are controlling for government revenue, economic growth, and so on,

which are some of the potential indirect channels through which corruption can affect public deficits

(Kaufmann, 2010). For comparison purposes, column 4 excludes the set of control variables. As

expected, the impact of corruption is now larger and tax decentralization contributes to a lower extent

to reduce this effect. Thus, the threshold of tax decentralization beyond which corruption does not

increase public deficit is now higher. Columns 5 to 7 are equivalent to columns 2 to 4 but with

subnational govern. expend. as an alternative indicator of fiscal decentralization. The conclusions

derived from this alternative indicator are similar22, even though Australia, Greece and Japan are

omitted due to unavailable data on expenditure decentralization. Figure 2 depicts the marginal effect

of corruption on public deficit depending on the degree of fiscal decentralization. The figure

summarizes the main finding of the paper: corruption raises public deficit when fiscal decentralization

is low, whereas no significant effect is found when the degree of decentralization increases.

[Insert Figure 2 about here]

Finally, it is possible to read the results from a different angle, namely, focusing on the marginal effect

of fiscal decentralization depending on the level of corruption. This is what we do in Figure 3, which

reports heterogeneity in the marginal effect of tax and expenditure decentralization, depending on the

corruption level. Interestingly, when corruption is low the impact of fiscal decentralization is

insignificant, but as corruption becomes more prevalent, decentralizing decreases fiscal deficits.

[Insert Figure 3 about here]

22 To interpret the results, one must keep in mind the range of variation of fiscal decentralization indicators. While values for subnational tax revenue vary from 0.9 to 46.4, subnational govern. expend. ranges from 10.3 to 61.

16

4.2 Panel data analysis

We now turn to panel estimation methods to further test the empirical regularity found. This is

appropriate for several reasons. Panel estimations exploit the temporal variation in the data, thus

improving efficiency. Also, some variables change significantly over time, so taking averages for

extended periods is not highly recommended. Moreover, better estimators are available to control for

endogeneity by using lags of the variables as instruments. We use both annual and averaged data over

5-year intervals for our computations, since it is unclear the exact time frame one should expect before

corruption and decentralization changes influence public deficits.

Table 4 presents the results from the panel analysis using subnational tax revenue as the fiscal

decentralization indicator. The first two columns show the coefficients obtained with pooled OLS,

where we also include time-specific effects (year dummies) to account for common shocks affecting

all countries in a given period.23 Column 1 indicates that corruption is strongly associated with public

deficits, while the next column shows that the effect of corruption depends on the degree of tax

decentralization. Compared to cross-section analysis (Table 3), in this case a higher level of

decentralization is necessary for the effect of corruption to cease to be significant (subnational tax

revenue around 20%). The results remain unaltered in columns 5 and 6 that employ data averaged over

5-year intervals.

[Insert Table 4 about here]

However, pooled OLS only serves as a first approximation. Other methods are necessary to control for

countries’ permanent characteristics as well as for endogeneity problems. Fixed effects models with

country-specific effects are not recommended in this case due to the fact that our main explanatory

variable –corruption– is highly persistent over time and many of the regressors are thought to be

endogenous. The difference GMM estimator (Arellano and Bond, 1991) addresses endogeneity

problems by using previous realizations of the regressors to instrument for their current values in the

first-differenced specification. However, Arellano and Bover (1995) and Blundell and Bond (1998)

show that in the case of highly persistent regressors, lagged levels of the variables are weak

instruments for the first-differenced regressors. This leads to a fall in precision as well as to biased

coefficients. In order to overcome these shortcomings, these authors recommend the use of the system

GMM estimator that utilizes instruments in levels and first-differences to improve in efficiency.

23 Strictly speaking, this estimator is labelled as one-way fixed effect model with time dummies.

17

Consequently, we will estimate the model using this estimator, which employs previous realizations of

the regressors to instrument for their current values in the first-differenced specification and the lagged

differences for the regression in levels. In order to avoid using an excessive number of instruments in a

context with a relatively short cross-country dimension, we follow the suggestion of Roodman (2009)

and limit the set of instruments to the minimum, i.e. to the first available: 2−itx for the specification in

first-differences and 1−∆ itx for the specification in levels.24 We use the one-step estimator since

standard errors for the two-step estimator are biased downwards. We apply orthogonal deviations,

which help to maximize the sample size in panels with gaps, like ours (Roodman, 2006). All our

regressors are treated as endogenous variables (except time-period dummies).

The consistency of the system estimator depends on the validity of the instruments and the absence of

serial correlation of second-order in the first-differenced error term. Therefore, we test these

assumptions using the Hansen test for over-identifying restrictions and the test for second-order

autocorrelation proposed by Arellano and Bond (1991). Failure to reject the null hypotheses of overall

validity of the instruments and absence of second-order serial correlation in the first-differenced error

for the respective tests would give support to the model.

Columns 3 and 7 confirm that corruption increases public deficit. The interaction models specified in

columns 4 and 8 show that this effect is conditioned to the level of tax decentralization. They indicate

that a one point increase in the corruption index leads to about 1.4 and 1.5 percentage points rises in

public deficits when subnational tax revenue is lower than the 10th percentile. However, as the degree

of decentralization rises, the effect of corruption becomes smaller, being insignificant at sufficiently

high levels of decentralization (60th percentile). Moreover, it is important to note that the Hansen test

for over-identifying restrictions and the test for second-order serial correlation are not rejected, thereby

supporting the validity of the model.

5. Robustness checks

5.1 Sensitivity analyses of baseline results

24 An additional way to further reduce the instrument set would be to only instrument for corruption, decentralization and their interaction. But this has the shortcoming that several other regressors are clearly endogenous to fiscal balances due to the operation of automatic stabilizers over the cycle such as GDP growth and the unemployment rate. If left uncontrolled, endogeneity would bias the results. Therefore, we opt for instrumenting all the regressors with the minimum possible instrument set: 2−itx and 1−∆ itx .

18

In this section we conduct sensitivity analysis to check the robustness of our baseline results. We

begin with Table 5 that introduces subnational govern. expend. as an alternative fiscal decentralization

indicator. Remarkably, we find the same pattern observed in Table 4. The main difference is that in the

panel with data averaged over 5-year intervals, when using the system GMM estimator, the effect of

corruption becomes insignificant at a lower threshold, namely, when expenditure decentralization

reaches the 20th percentile (see column 4).

[Insert Table 5 about here]

Table 6 analyzes the robustness of our reference estimation, which is the panel based on 5-year

averages using the system GMM estimator. Column 2 presents the results for an alternative tax

decentralization measure which is an updated version of Stegarescu’s TD1 indicator using OECD

information for the data points 2002, 2005 and 2008.25 It is remarkable that our main finding regarding

the heterogeneity in the effect of fiscal decentralization remains robust to the use of this alternative tax

decentralization indicator. As in the baseline analysis, we find that for values of tax decentralization

greater than the 50th percentile, the positive marginal effect of corruption becomes insignificant.

Hence, the effect of corruption on fiscal deficits is significantly positive when tax decentralization is

relatively low, whereas corruption does not significantly affect fiscal balances if a country is highly

decentralized.

The remaining columns introduce additional control variables which are thought to affect fiscal

deficits in industrialized economies or at least have some effect on it through either fiscal

decentralization or corruption. Debt service proxies for interest payments that in the short-run are

mostly affected by the existing stock of debt, which have an immediate positive influence on current

deficits. The government bond yield at a 10-year maturity tries to capture the fact that debt interest

payments above some threshold can generate a permanent increase in the public deficit of sufficient

magnitude as to make the stock of public debt unsustainable. The current financial crisis has also

shown how important high interest rates on sovereign bonds are as a disciplining mechanism of

international investors on governments (acting as a “hard budget constraint”), thus leading to lower

future fiscal deficits. The fiscal cost of a banking crisis controls for the effect that financial distress

25 In order to update Stegarescu’s TD1 indicator, we calculate the share of sub-central tax revenues on which sub-central government has discretion over tax rates and/or tax bases –a) to c) components of the OECD classification of taxing power (OECD, 2012a)– as a percentage of general government total tax revenue. Then, we link Stegarescu’s data with the new series by applying the growth rate of these ratios to the former to avoid structural change in the series.

19

and bank bailouts can bring on public sector deficits. As an alternative measure, we also employ a

dummy variable for the years where a banking crisis was in place. The dummy federal represents the

presence of a federalist constitution implying a legislated division of powers between central and

subnational authorities. This variable tries to control for the possibility of overgrazing in federal

constitutions due to multiple and sometimes overlapping layers of government.26 Government and

parliament fragmentation are variables that depend on the efficiency of decentralization according to

Persson and Tabellini (2003). The other controls (inflation rate, trade openness, freedom of press,

democracy indicator (polity2) and the square of tax decentralization) are factors that can affect fiscal

deficits either directly or through their effect on corruption.27

The results are reported in columns 3 to 14 of Table 6. In all cases, our baseline results remain fairly

robust, while debt service, freedom of press, the banking crisis dummy, democracy and the square

term appear with significant coefficients (with a positive sign for the five variables).28 In addition, as a

further sensitivity analysis we check for the possible influence of outliers by excluding countries one

at a time. We also delete observations whose standardized residuals are higher than two in absolute

value. The evidence obtained indicates that the results are neither driven by any particular country nor

by observations with high residuals.29

[Insert Table 6 about here]

5.2 Within-country effects: An alternative empirical approach

The above results can be read from the angle of the effect of decentralization on public deficits

depending on countries’ corruption levels: decentralization only leads to lower public deficits when

the level of corruption is high. This section makes a complementary analysis to further explore this

non-linear effect by using an alternative approach. We divide the sample into three groups according

to the average level of corruption and then we analyze whether there are significant differences among

them regarding the effect of decentralization on public deficits. In this way, corruption is held constant

and considered a country feature that only changes very gradually over time, and instead we focus on

26 Indeed, Treisman (2000) provides robust evidence that federal states are more corrupt. 27 Focusing on a broad cross-section of 64 countries over the period 1980-2000, Lessmann and Markwardt (2010) provide evidence that a free press is necessary for fiscal decentralization to reduce the extent of corruption. 28 Note that the freedom of press indicator ranges from 0 to 100, where a higher value implies lower press freedom. 29 We do not report these regressions for space considerations. They are available in the unpublished appendix. Likewise, in the unpublished appendix we also provide the results using annual data. Remarkably, our main findings hold irrespective of the frequency of the data considered.

20

the variation in fiscal decentralization.30 We expect to find heterogeneity in the relationship between

this variable and public deficit across the three groups of countries. Decentralization is expected to

have more negative effects on public deficits in the group with higher corruption.

A fixed effects model is used in order to focus on “within-country variation”. Thus, by controlling for

all time-invariant characteristics, we ask for each group whether countries are more likely to reduce

the public deficit as they become more decentralized. More specifically, the equation to estimate is:

titititi

tititi

Xcorrhighdecentr

corrdecentrdecentrdeficit

,,,3

,2,1,

_

_int

εθαδβ

ββ

+++⋅+×⋅

+×⋅+⋅=

(4)

where decentr×int_corri,t and decentr×high_corri,t are the interaction terms between fiscal

decentralization and the intermediate and high corruption groups, respectively; αi and θt are sets of

country and time specific effects and the remaining variables are the same as in Section 4. β1 estimates

the effect of decentralization on public deficits for the “low corruption group” and β2 and β3 measure

the differences with respect to that group.31

Table 7 presents the results obtained from the fixed effects model for our main indicator of fiscal

decentralization (subnational tax revenue). Column 1 shows high heterogeneity in the coefficient on

tax decentralization across corruption groups for the panel based on annual data. The coefficient is

positive and significant for the countries included in the “low corruption group”. The “intermediate

corruption group” shows a coefficient on tax decentralization significantly different from the previous

one, being its value close to zero (-0.63 + 0.62 = -0.01). The coefficient associated with the “high

corrupt group” is also significantly different from the reference group and negative (-0.82 + 0.62 = -

0.20). Therefore, there is clear-cut heterogeneity in the effect of tax decentralization across corruption

groups. Remarkably, this variable is related to lower deficits only in relatively corrupt countries.

30 We average the indicator ICRG corruption over the period 1984-2010 (all available years). Taking long-term averages makes sense because corruption is a highly persistent institutional characteristic. In this way, we reduce measurement error and avoid doubts about the accuracy of corruption and other government indicators to capture the temporal variation of the data (Treisman, 2007; Arndt and Oman, 2006). The “high corruption group” comprises those countries with average values of corruption higher than the 66th percentile, the “intermediate corruption group” those with average values between the 33th and 66th percentiles, and the “low corruption group” those with values below the 33th percentile. Table 2 shows the countries included in each group. We create a dummy variable for each corruption group. 31 Hence, the effect of fiscal decentralization on public deficits for the intermediate and high corruption groups are (β1 + β2) and (β1 + β3), respectively. The significance level of β2 and β3 indicates whether differences with respect to β1 are statistically significant; therefore, these two coefficients allow us to test the presence of heterogeneity in the effect of fiscal decentralization.

21

[Insert Table 7 about here]

We test whether the previous results are due to the particular classification of countries made with

ICRG corruption. Columns 2 to 4 use alternative indicators to create the groups. Column 2 utilizes

Corruption Perception Index from Transparency International, while column 3 uses Control of

Corruption from the World Bank Governance Indicators. The resulting classifications are very similar

to the previous one, despite the use of different sources and period coverage.32 Consequently, the

results shown in columns 2 and 3 are very consistent with those of column 1. Rather than a perception-

based indicator, column 4 uses bribe payments, a measure of actual corruption experiences obtained

from the Global Corruption Barometer of Transparency International.33 Although in this case there are

three countries with missing data (Estonia, Slovak Republic and Belgium) and matches with the

reference classification are lower (68%), the results appear highly consistent. The effect of tax

decentralization on public deficit is heterogeneous across corruption groups and is only significantly

negative for the “high corruption group”.

Columns 5 to 8 repeat the above analysis but using data averaged over 5-year intervals. The results

remain remarkably robust: the coefficient on tax decentralization is positive for the “low corruption

group”, while for the other two groups the coefficient is significantly lower, being very close to zero

for the intermediate group and negative for the “high corruption group”.

To sum up, this subsection has shown from a different approach the presence of heterogeneity in the

relationship between fiscal decentralization and public deficit depending on the level of corruption.

Focusing on “within-country variation”, we observe that decentralizing public finances is associated

with a reduction in fiscal deficits only in relatively more corrupt countries.

5.3 Discussion of results

In this subsection we try to discuss the theoretical mechanisms that could help us to explain the fact

that decentralization only leads to lower public deficits when the level of corruption is high, whereas it

does not reduce fiscal deficits in a context with low corruption. The line of arguments provided is

essentially associated with the fact that in a corrupt environment, fiscal decentralization brings about

32 Due to data availability, we take the average over the years 1996-2009. Matches between classifications made with ICRG indicator and Corruption Perception Index are 87.1%, and between classifications made with ICRG indicator and Control of Corruption are 96.8%. 33 We take the average over the years 2006-2010.

22

some disciplining mechanisms such as local accountability or greater interjurisdictional competition

that can mitigate the negative effect of corruption on fiscal balances. In contrast, when the government

is diligent and honest, fiscal decentralization has less scope to discipline fiscal management, and as a

result, the effect on reducing fiscal deficits is less clear.

Arikan (2004) develops a tax-competition model with rent-seeking that predicts that a horizontal rise

in the number of jurisdictions that compete for residents and business investment reduces corruption

through a fall in the revenue expropriated by rent-seeking bureaucrats. Along similar lines, in

Kotsogiannis and Schwager (2006) interjurisdictional competition lowers the possibilities and

incentives for rent-seeking on the part of political officials. Establishing hard-budget constraints make

local officials responsible for financing the cost of public goods provision, thus acting as a disciplining

mechanism of local accountability. However, in their absence, i.e. when soft-budget constraints34 and

implicit bailout from the central government prevail, local officials will be more prone to pursuing

rent-seeking and over-invoicing in the provision of public goods.35 In this particular case, the existence

of municipal or state elections can serve as an alternative disciplining device to raise local

accountability, thus helping to prevent corrupt behavior among local officials.

The rationale for decentralizing tax and expenditure decision-making to the lower levels of

government lays in the fact that peer monitoring can take place to directly control agents’ behavior

(Stiglitz, 1999). Fiscal decentralization aligns spending and funding responsibilities, which improves

the efficiency in the provision of public goods. Along these lines, Hindricks and Lockwood (2009) and

Weingast (2009) emphasize that greater fiscal autonomy enhances incumbent politicians’ performance

via electoral accountability. Persson and Tabellini (2000) develop a model on decentralization where

political officials try to minimize effort and maximize the probability of re-election. Unlike in a

centralized system where bureaucrats are responsible for multiple tasks affecting many jurisdictions,

decentralization makes bureaucrats deal with only one task specific to a single locality. By better

linking effort and rewards, decentralization improves the performance of political officials, which can

be held accountable for their actions by their local constituents.

34 These can result from the lack of tax compliance and the existence of unconditional grants from the central level of government to subnational levels. 35 Indeed, Weingast (2009) notes that for greater competition to be an effective mechanism for policing corruption, the following must be satisfied: common market conditions, sufficient subnational authority in decision-making and a hard budget constraint.

23

Empirically this is supported by Khemani (2001) who provides evidence for 14 major Indian states

over the period 1960-92 that voters are more vigilant in monitoring the performance of the state

government tier in terms of output growth and income inequality compared to the national legislature,

where they act more myopically. This thus supports the presence of greater government accountability

in elections at the lower levels of government. Finally, focusing on a sample of more than 110

countries over the period 1980-95, Adserá et al. (2003) find that electoral accountability and the

exercise of democracy explain more than half of the variability in the levels of government

performance and corruption. The reason for this is that informed voters are in a better position to

discipline politicians, thereby preventing the latter from engaging in political rent creation and

seeking.

6. Conclusions

This paper has aimed at providing a practical policy advice regarding a first-order concern in

industrialized countries, the public deficit. Throughout the analysis, we have argued that one of the

major factors causing public deficits, corruption, may be addressed by decentralizing public finances.

Focusing on OECD countries, our article provides extensive empirical support for that proposition.

The effect of corruption is non-linear: corruption increases public deficit when fiscal decentralization

is low, but as it rises, the effect becomes smaller, being insignificant at sufficiently high levels of

decentralization. The result implies heterogeneity in the effect of fiscal decentralization on public

deficit, in the sense that decentralizing public finances is more likely to reduce fiscal deficits in

relatively more corrupt countries.

The case of Greece clearly illustrates the results. This country has a relatively high level of corruption,

being its average level of public deficit over GDP during the period of analysis of almost 7%, the

greatest of the sample. Interestingly, Greece is also the most centralized country, with a degree of tax

decentralization slightly lower than 1%. From our findings, it follows that Greece could reduce public

deficit by increasing fiscal decentralization. According to the results from the 5-year period panel

(Table 4, column 8), for this country the effect of corruption on public deficits is 3.1. By increasing

subnational tax revenue to the level of Poland or France (about 10.5%), the effect would be 2.1,

leading to a reduction of one percentage point in the public deficit. Compared to combating

corruption, which requires time to dismantle well-established networks of patronage and rent creation

and seeking, decentralizing public finances is more feasible to implement, particularly in the short-

24

term. Hence, a good policy set would be to combine a program of fight against corruption while

increasing fiscal decentralization.

Before making generalizations, a few caveats must be considered. First, we analyze a sample of

OECD countries. Conclusions derived from our results cannot be in principle extrapolated to

developing countries for a number of reasons. All our countries are economically developed,

democratic and with an educated population. Furthermore, despite existing variation in corruption

levels, OECD countries are relatively less corrupt when compared to many developing countries.

Therefore, further research is necessary to test our findings in a sample of developing and emerging

countries. Secondly, the results indicate a stylized fact and suggest but do not prove causality, so

interpretations must be made carefully and complemented with the extant literature on fiscal

decentralization. In particular, a proper design of the decentralization program is essential, with

incentives leading to a responsible fiscal policy (Weingast, 2009). A proper political-institutional

framework is also necessary to guarantee accountability and citizen control of subnational

governments. Moreover, a cautious interpretation of our findings implies that using fiscal

decentralization to mitigate the adverse effects of corruption on public deficits is more advisable when

the country starts from low levels of decentralization, as the case of Greece. When the degree of

decentralization is already medium or high –as the case of Spain–, the advice of increasing

decentralization is much less clear.

Overall, this paper has made a contribution to the literature on decentralization. Our empirical analysis

has uncovered substantial heterogeneity in the effect of fiscal decentralization depending on the

institutional context. We find that decentralization behaves as a disciplining mechanism that reduces

fiscal deficits in a corrupt public environment, but not when corruption is absent. This leads us to an

interesting theoretical debate. The result suggests that bringing the government closer to the people is

more useful or necessary when public administration is not working properly. In this case, the agent

(politicians and public employees) misuses public resources for private gain, and the principal

(citizens) must monitor closely the former. This is more feasible when the government is close to the

people, since citizens are better informed about local affairs and can revoke corrupt rulers in elections.

The argument can be easily extended to related concepts such as political and citizen participation,

which is supposed to be more necessary when the government does not work well. Likewise, other

dimensions of the performance of countries may be considered from this perspective, such as social

policies or environmental quality.

25

References

Adserá, A., Boix, C., and Payne, M. (2003) ‘Are you being served? Political accountability and quality of government,’ The Journal of Law, Economics and Organization 19 (2), 445-490.

Aidt, T.S. (2009) ‘Corruption, institutions, and economic development,’ Oxford Review of Economic Policy 25(2), 271–291.

Aidt, T.S., Dutta, J., and Sena, V. (2008) ‘Governance regimes, corruption and growth: theory and evidence,’ Journal of Comparative Economics 36(2), 195–220.

Aiken, L. and West, S. (1991) Multiple regression: testing and interpreting interactions. (London: Sage Publications).

Arellano, M. and Bond, S. (1991) ‘Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations,’ Review of Economic Studies 58(2), 277–297.

Arellano, M. and Bover, O. (1995) ‘Another look at the instrumental variable estimation of error-components models,’ Journal of Econometrics 68(1), 29–51.

Arikan, G.G. (2004) ‘Fiscal decentralization: A remedy for corruption?,’ International Tax and Public Finance 11, 175-195.

Arndt, C. and Oman, C. (2006) Uses and abuses of governance indicators. Paris: OECD Development Centre Studies.

Baskaran, T. (2010) ‘On the link between fiscal decentralization and public debt in OECD countries,’ Public Choice 145(3), 351–378.

Baskaran, T. (2012) ‘Tax decentralization and public deficits in OECD countries,’ Publius: The Journal of Federalism forthcoming.

Blundell, R. and Bond, S. (1998) ‘Initial conditions and moment restrictions in dynamic panel data models,’ Journal of Econometrics 87(1), 115–143.

Brambor, T., Clark, W., and Golder, M. (2006) ‘Understanding interaction models: improving empirical analyses,’ Political Analysis 14(1), 63–82.

Brennan, G. and Buchanan, J. (1980) The power to tax: analytical foundations of a fiscal constitution. (Cambridge: Cambridge University Press).

Careaga, M., and Weingast, B. R. (2003) Fiscal federalism, good governance, and economic growth in Mexico. In Rodrik, D. (Ed.), Search of Prosperity: Analytic Narratives on Economic Growth. (Princeton University Press: Princeton).

De Mello, L.R. (2000) ‘Fiscal decentralization and intergovernmental fiscal relations: a cross country analysis,’ World Development, 28(2), 365–380.

Depken, C.A. and Lafountain, C.L. (2006) ‘Fiscal consequences of public corruption: empirical evidence from state bond ratings,’ Public Choice 126(1), 75–85.

Dillinger, W. and Webb, S. (1999) ‘Fiscal management in federal democracies: Argentina and Brazil,’ World Bank. Policy Research Working Paper, no. 2121.

Fan, C.S., Lin, C., and Treisman, D. (2009) ‘Political decentralization and corruption: evidence from around the world,’ Journal of Public Economics 93(1–2), 14–34.

Fisman, R. and Gatti, R. (2002) ‘Decentralization and corruption: evidence across countries,’ Journal of Public Economics 83(3), 325–345.

Hindriks, J., and Lockwood, B. (2009) ‘Decentralization and electoral accountability: Incentives, separation and voter welfare,’ European Journal of Political Economy 25, 385-397.

26

International Monetary Fund (2012) International Financial Statistics. International Monetary Fund, Washington D.C.

Ivanyna, M. and Shah, A. (2010) ‘Decentralization (localization) and corruption. New cross-country evidence,’ World Bank. Policy Research Working Paper, no. 5299.

Kaufmann, D. (2010) ‘Can corruption adversely affect public finance in industrialized countries?,’ Brookings Institution Opinions, April 19.

Kaufmann, D., Kraay, A., and Mastruzzi, M. (2010) ‘The worldwide governance indicators: methodology and analytical issues,’ World Bank. World Bank Policy Research Working Paper, no. 5430.

Khemani, S. (2001) ‘Decentralization and accountability: Are voters more vigilant in local than in national elections?,’ World Bank Policy Research Working Paper no. 2557, World Bank, Washington D.C.

Kotsogiannis, C., and Schwager, R. (2006) ‘Fiscal equalization and Yardstick competition,’ mimeograph.

Laeven, L. and Valencia, F. (2012). ‘Systemic banking crises database: An update,’ IMF Working Paper. WP/12/163, Washington DC.

Lessmann, C., and Markwardt, G. (2010) ‘One size fits all? Decentralization, corruption, and the monitoring of bureaucrats,’ World Development 38 (4), 631-646.

Mauro, P. (1995) ‘Corruption and growth,’ Quarterly Journal of Economics, 110(3), 681–712.

Mauro, P. (1997) Why Worry About Corruption? Washington, D.C.: International Monetary Fund (IMF).

Neyapti, B. (2010) ‘Fiscal decentralization and deficits: International evidence,’ European Journal of Political Economy 26(2), 155–166.

Oates, W.E. (1999) ‘An essay on fiscal federalism,’ Journal of Economic Literature 37(3), 1120–1149.

Oates, W.E. (2008) ‘On the evolution of fiscal federalism: theory and institutions,’ National Tax Journal 61(2), 313–334.

OECD. (1999) Taxing power of state and local government. OECD Tax Policy Studies, no. 1.

OECD. (2012a) Fiscal decentralization database. URL: http://www.oecd.org/ctp /fiscalfederalismnetwork/oecdfiscaldecentralisationdatabase.htm [Accessed on 11 June 2012].

OECD. (2012b) Economic Outlook, no. 91, Organization for Economic Cooperation and Development, Paris.

Persson, T., and Tabellini, G. (2000) ‘Constitutional determinants of government spending,’ Mimeograph, Bocconi University.

Persson, T., and Tabellini, G. (2003) The Economic Effects of Constitutions. (Cambridge, M.A.: The MIT Press).

PRS Group (2012) ICRG methodology. Political Risk Services Group. On-line publication. URL: http://www.prsgroup.com/ICRG_Methodology.aspx [Accessed on 20 June 2012].

Prud’homme, R. (1995) ‘On the dangers of decentralization,’ World Bank Research Observer 10(2), 201–220.

Rajkumar, A.S. and Swaroop, V. (2008) ‘Public spending and outcomes: does governance matter?,’ Journal of Development Economics 86(1), 96–111.

Rodden, J. (2003) ‘Reviving Leviathan,’ International Organization 57, 695–729.

Rodden, J. and Wibbels, E. (2002) ‘Beyond the fiction of federalism. Macroeconomic management in multitiered systems,’ World Politics 54(4), 494–531.

Rodríguez-Pose, A. and Ezcurra, R. (2011) ‘Is fiscal decentralization harmful for economic growth? Evidence from the OECD countries,’ Journal of Economic Geography 11(4), 619–643.

27

Rodríguez-Pose, A. and Gill, N. (2003) ‘The global trend towards devolution and its implications,’ Environment & Planning C: Government & Policy 21(3), 333–351.

Roodman, D. (2006) ‘How to do xtabond2: an introduction to “difference” and ”system” GMM in Stata,’ Center for Global Development (Washington, D.C.), Working Paper no. 103.

Roodman, D. (2009) ‘A note on the theme of too many instruments,’ Oxford Bulletin of Economics and Statistics 71, 135-158.

Shah, A. (2006) ‘Fiscal decentralization and macroeconomic management,’ International Tax and Public Finance 13(4), 437–462.

Shleifer, A., and Vishny, R. (1993) ‘Corruption,’ Quarterly Journal of Economics 108, 599-617.

Stegarescu, D. (2005) ‘Public sector decentralization: measurement concepts and recent international trends,’ Fiscal Studies 26(3), 301–333.

Stegarescu, D. (2009) ‘The effects of economic and political integration on fiscal decentralization: Evidence from OECD countries,’ Canadian Journal of Economics 42, 694-718.

Stiglitz, J. (1999) ‘Whither reform? Ten years of the transition,’ Paper Prepared for the Annual World Bank Conference on Development Economics, Washington, D.C., April 28-30.

Tanzi, V. (1999) Governance, corruption and public finance: an overview. In: S. Schiavo-Campo, ed. Governance, corruption and public financial management. Manila: Asian Development Bank.

Tanzi, V. and Davoodi, H. (1998) Roads to nowhere: how corruption in public investment hurts growth. Washington, D.C.: International Monetary Fund (IMF).

Teorell, J., Samanni, M., Holmberg, S., and Rothstein, B. (2011) ‘The quality of government dataset,’ [version 6 April 2011]. University of Gothenburg, The Quality of Government Institute. URL: http://www.qog.pol.gu.se.

Tiebout, C. (1956) ‘A pure theory of local expenditures,’ Journal of Political Economy 64(5), 416–424.

Transparency International. Global Corruption Barometer. Several years (2006-2010) URL: http://www.transparency.org/research/gcb/overview.

Treisman, D. (2000) ‘The causes of corruption: A cross-national study,’ Journal of Public Economics 76 (3), 399–457.

Treisman, D. (2002) ‘Decentralization and the quality of government,’ University of California, mimeo. URL: http://www.sscnet.ucla.edu/polisci/faculty/treisman/ Pages /unpublishedpapers.html.

Treisman, D. (2007) ‘What have we learned about the causes of corruption from ten years of cross-national empirical research?,’ Annual Review of Political Science 10, 211–244.

Wei, S.J. (1997a) Why is corruption so much more taxing than tax? Arbitrariness kills. Cambridge, Massachusetts: National Bureau of Economic Research (NBER).

Wei, S.J. (1997b) ‘How taxing is corruption on international investors?,’ NBER Working Paper, no. 6255.

Weingast, B.R. (1995) ‘The economic role of political institutions: market-preserving federalism and economic development,’ Journal of Law, Economics and Organization 11(1), 1–31.

Weingast, B.R. (2009) ‘Second generation fiscal federalism: the implications of fiscal incentives,’ Journal of Urban Economics 65(3), 279–293.

World Bank. (2011) World Development Indicators 2011 Database. Washington, D.C.: World Bank.

World Bank. (2012a) Anticorruption. costs and consequences of corruption. World Bank, Washington D.C. URL: http://go.worldbank.org/BVKN7CIK60 [Accessed on 25 June 2012].

World Bank. (2012b) Anticorruption. Corruption and fiscal stability. World Bank, Washington D.C. URL: http://go.worldbank.org/007SFZXAU0 [Accessed on 25 June 2012].

28

Tables and figures

TABLE 1

CountryICRG

corruptionFiscal decentralization (subnational tax revenue)

Public deficit

Two relatively corrupt countries

Greece 2.1 0.9 6.9vs.

Poland 2.4 10.5 4.6

Two relatively non-corrupt countries

The Netherlands 0.4 3.0 2.6vs.

Canada 0.4 46.4 2.7

Descriptive comparisons among country-pairs

NOTES: The variables represent the average of available data over the period 1986-2010. Thedefinitions of the variables can be found in Appendix I.

TABLE 2

ICRG corruption

Fiscal decentralization (subnational tax revenue)

Public deficit

High corruption 2.43 7.48 3.98

▪ Low tax decentralization

Greece, The Slovak Republic, The Czech Republic, Hungary and Israel.

2.20 3.31 5.73

▪ High tax decentralization

Slovenia, Italy, Poland, Estonia, and Korea (Rep.).

2.67 11.64 2.22

Intermediate corruption 1.57 14.13 3.30

▪ Low tax decentralization

Ireland, Portugal, The United Kingdom, Belgium and France.

1.61 6.13 3.64

▪ High tax decentralization

Austria, Spain, Australia, Japan and The United States.

1.53 22.13 2.96

Low corruption 0.45 23.23 0.08

▪ Low tax decentralization

The Netherlands, Luxembourg, New Zealand, Norway, Finland and Iceland.

0.36 12.61 -0.99

▪ High tax decentralization

Germany, Denmark, Sweden, Switzerland and Canada.

0.55 35.97 1.35

Corruption, fiscal decentralization and public deficit

NOTES: The variables represent the average of available data over the period 1986-2010. The definitions of the variables can be found in Appendix I.

Averages by groups

Corruption groups/ Countries

29

TABLE 3

(1) (2) (3) (4) (5) (6) (7)

1.295** 0.812 2.268** 2.702*** 0.872 2.5* 3.926***

(0.56) (0.71) (0.86) (0.9) (0.74) (1.25) (1.39)

-0.071* 0.035 0.068 -0.04 0.018 0.064

(0.04) (0.06) (0.06) (0.03) (0.04) (0.04)

-0.135** -0.091 -0.061 -0.082*

(0.06) (0.06) (0.04) (0.05)

-0.078 -0.105 -0.137 -0.101 -0.124

(0.1) (0.1) (0.1) (0.1) (0.11)

-1.587*** -1.708*** -1.449*** -1.5*** -1.17**

(0.41) (0.36) (0.34) (0.38) (0.41)

0.315* 0.328** 0.267** 0.339** 0.309**

(0.16) (0.15) (0.12) (0.16) (0.14)

-0.311 -0.122 0.093 -0.195 -0.167

(1.03) (0.98) (0.87) (1.06) (1.06)

0.004 0.009* 0.02** 0.006 0.01*

(0.004) (0.004) (0.01) (0.004) (0.005)

R-squared 0.53 0.58 0.65 0.29 0.55 0.58 0.31Number of obs. 31 31 31 31 28 28 28

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2.1) (Exp.: 12.7) 1.98** 2.51*** 1.73* 2.88***(0.787) (0.835) (0.848) (0.91)

Pct 20 (Tax: 4.8) (Exp.: 17.5) 1.62** 2.26*** 1.44* 2.49***(0.71) (0.772) (0.745) (0.77)

Pct 30 (Tax: 5.8) (Exp.: 25) 1.49** 2.18*** 0.98 1.87***(0.69) (0.757) (0.682) (0.648)

Pct 40 (Tax: 7.7) (Exp.: 27.9) 1.23* 2** 0.81 1.64**(0.66) (0.739) (0.697) (0.646)

Pct 50 (Tax: 10.5) (Exp.: 30.4) 0.85 1.75** 0.66 1.44**(0.651) (0.744) (0.727) (0.666)

Pct 60 (Tax: 17.1) (Exp.: 32.6) -0.04 1.15 0.52 1.25*(0.777) (0.889) (0.765) (0.7)

Pct 70 (Tax: 20.8) (Exp.: 37.1) -0.54 0.81 0.25 0.88(0.913) (1.029) (0.868) (0.803)

Pct 80 (Tax: 28.3) (Exp.: 43.8) -1.54 0.14 -0.15 0.34(1.253) (1.373) (1.067) (1.017)

Pct 90 (Tax: 33.1) (Exp.: 55.1) -2.19 -0.3 -0.84 -0.59(1.502) (1.626) (1.473) (1.462)

Cross-section analysis

Dependent variable is public deficit over GDP

Tax decentralization Expenditure decentralization

NOTES: The proxies for tax and expenditure decentralization are subnational tax revenue and subnational govern. expend ., respectively. The variables represent the average of available data over the period 1986-2010. The estimations include aconstant term, which is omitted for space considerations. All regressions are estimated with OLS. Robust standard errors arein parentheses. *, ** and *** denote significance at the 10, 5 and 1% levels, respectively. Percentiles of tax and expendituredecentralization are calculated for average values over the period 1986-2010. The definitions of the variables can be found inAppendix I. The list of countries included in the sample is shown in Table 2.

Ln GDP pc

Population

ICRG corruption

Fiscal decentralization

Corruption x Fiscal decent.

Total government revenue

Growth GDP pc

Unemployment rate

30

TABLE 4

(1) (2) (3) (4) (5) (6) (7) (8)

0.801*** 1.391*** 0.857* 1.413*** 1.099*** 1.674*** 1.048* 1.516***

(0.16) (0.2) (0.48) (0.47) (0.31) (0.34) (0.62) (0.53)

0.016 0.027 0.019 0.005

(0.01) (0.04) (0.03) (0.04)

-0.058*** -0.055* -0.075*** -0.051*

(0.01) (0.03) (0.02) (0.03)

-0.042* -0.072*** -0.035 -0.06 -0.054 -0.086* 0.062 0.046

(0.02) (0.02) (0.07) (0.06) (0.05) (0.05) (0.16) (0.14)

-0.482*** -0.494*** -0.582*** -0.536*** -0.882*** -0.911*** -0.425 -0.642**

(0.08) (0.08) (0.13) (0.11) (0.18) (0.17) (0.31) (0.28)

0.328*** 0.301*** 0.446*** 0.363*** 0.271*** 0.256*** 0.327* 0.318**

(0.04) (0.04) (0.13) (0.11) (0.08) (0.07) (0.17) (0.14)

-1.399*** -1.168*** -1.163 -1.08 -1.273** -1.055* -1.023 -1.095

(0.33) (0.34) (0.73) (0.75) (0.6) (0.61) (0.83) (0.77)

0.009*** 0.016*** 0.009*** 0.014*** 0.007** 0.017*** 0.012 0.019**

(0.002) (0.002) (0.003) (0.005) (0.004) (0.004) (0.01) (0.01)

Arellano-Bond test for AR(2) 0.07 0.11 0.57 0.73Hansen test of overid. rest. 1.00 1.00 0.87 1.00R-squared 0.45 0.48 0.47 0.52Number of obs. 636 632 636 632 142 142 142 142

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2.1) 1.27*** 1.3*** 1.52*** 1.41***(0.189) (0.442) (0.321) (0.521)

Pct 20 (Tax: 4.8) 1.11*** 1.15*** 1.31*** 1.27**(0.178) (0.415) (0.302) (0.52)

Pct 30 (Tax: 5.8) 1.06*** 1.1*** 1.24*** 1.23**(0.175) (0.41) (0.297) (0.523)

Pct 40 (Tax: 7.7) 0.94*** 0.99** 1.09*** 1.13**(0.172) (0.405) (0.292) (0.534)

Pct 50 (Tax: 10.5) 0.78*** 0.84** 0.88*** 0.98*(0.173) (0.415) (0.294) (0.561)

Pct 60 (Tax: 17.1) 0.39** 0.48 0.38 0.65(0.197) (0.505) (0.338) (0.664)

Pct 70 (Tax: 20.8) 0.18 0.28 0.1 0.46(0.221) (0.584) (0.382) (0.741)

Pct 80 (Tax: 28.3) -0.26 -0.13 -0.46 0.09(0.284) (0.771) (0.493) (0.917)

Pct 90 (Tax: 33.1) -0.54 -0.4 -0.82 -0.16(0.331) (0.908) (0.576) (1.044)

5-year average data

OLS System GMM estimator

Panel data analysis. Tax decentralization

Dependent variable is public deficit over GDP

NOTES: The proxy for tax decentralization is subnational tax revenue . The period analyzed is 1986-2010. The estimations include aconstant term and period dummies, which are omitted for space considerations. Robust standard errors are in parentheses. *, ** and ***denote significance at the 10, 5 and 1% levels, respectively. Percentiles of tax decentralization are calculated for average values over theperiod 1986-2010. The definitions of the variables can be found in Appendix I. The list of countries included in the sample is shown in Table2. Regressions 1, 2, 5 and 6 are estimated with OLS. Regressions 3, 4, 7 and 8 are estimated with the one-step system GMM estimator usingthe STATA program xtabond (Roodman, 2006), with the orthogonal deviations transformation applied. For the first-difference equation, thesecond lag of explanatory variables (all treated as endogenous) are used as instruments. For the level equation, the lagged first-difference ofexplanatory variables are used as instruments. The exogenous variables are the period dummies.

Ln GDP pc

Population

Annual data

OLS System GMM estimator

Corruption x Tax decent.

Total government revenue

Growth GDP pc

Unemployment rate

ICRG corruption

Tax decentralization

31

TABLE 5

OLS System GMM estimator OLS System GMM estimator

(1) (2) (3) (4)

2.562*** 2.566*** 2.979*** 2.034*

(0.29) (0.65) (0.53) (1.13)

0.042*** 0.044* 0.039 0.032

(0.01) (0.03) (0.02) (0.04)

-0.068*** -0.064*** -0.074*** -0.056*

(0.01) (0.02) (0.02) (0.03)

-0.112*** -0.101 -0.089 -0.079

(0.03) (0.08) (0.06) (0.11)

-0.418*** -0.397*** -0.675*** -0.428

(0.08) (0.09) (0.18) (0.27)

0.264*** 0.31*** 0.186** 0.383*

(0.05) (0.11) (0.08) (0.22)

-1.538*** -0.959 -1.319** -0.775

(0.35) (0.82) (0.65) (0.93)

0.012*** 0.011*** 0.014*** 0.014***

(0.002) (0.003) (0.004) (0.01)

Arellano-Bond test for AR(2) 0.07 0.41Hansen test of overid. rest. 1.00 1.00R-squared 0.54 0.54Number of obs. 492 492 118 118

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Exp.: 12.7) 1.7*** 1.75*** 2.03*** 1.32*(0.209) (0.48) (0.388) (0.8)

Pct 20 (Exp.: 17.5) 1.38*** 1.44*** 1.68*** 1.05(0.186) (0.429) (0.348) (0.686)

Pct 30 (Exp.: 25) 0.87*** 0.96** 1.11*** 0.63(0.163) (0.378) (0.313) (0.532)

Pct 40 (Exp.: 27.9) 0.67*** 0.78** 0.9*** 0.47(0.16) (0.37) (0.311) (0.487)

Pct 50 (Exp.: 30.4) 0.51*** 0.62* 0.72** 0.33(0.159) (0.369) (0.315) (0.456)

Pct 60 (Exp.: 32.6) 0.36** 0.48 0.55* 0.2(0.161) (0.373) (0.322) (0.437)

Pct 70 (Exp.: 37.1) 0.05 0.19 0.22 -0.05(0.17) (0.393) (0.348) (0.426)

Pct 80 (Exp.: 43.8) -0.4** -0.24 -0.28 -0.42(0.195) (0.449) (0.406) (0.48)

Pct 90 (Exp.: 55.1) -1.17*** -0.97 -1.13** -1.06(0.259) (0.59) (0.541) (0.696)

Panel data analysis. Expenditure decentralization

Dependent variable is public deficit over GDP

Annual data 5-year average data

Ln GDP pc

Population

NOTES: The proxy for expenditure decentralization is subnational govern. expend. The period analyzed is 1986-2010. The estimations include aconstant term and period dummies, which are omitted for space considerations. Robust standard errors are in parentheses. *, ** and *** denotesignificance at the 10, 5 and 1% levels, respectively. Percentiles of expenditure decentralization are calculated for average values over the period 1986-2010. The definitions of the variables can be found in Appendix I. The list of countries included in the sample is shown in Table 2. Regressions 1 and3 are estimated with OLS. Regressions 2 and 4 are estimated with the one-step system GMM estimator using the STATA program xtabond (Roodman,2006), with the orthogonal deviations transformation applied. For the first-difference equation, the second lag of explanatory variables (all treated asendogenous) are used as instruments. For the level equation, the lagged first-difference of explanatory variables are used as instruments. Theexogenous variables are the period dummies.

Corruption

Expenditure decentralization

Corruption x Expenditure decent.

Total government revenue

Growth GDP pc

Unemployment rate

32

TABLE 6

Reference

(T.4-C.8)

Debt service

i bond

Banking

crisis - fiscal

cost

Banking

crisis

dummy

Federal

Governm

ent

fragment.

Parliam

ent

fragment.

Inflation

Openness

Freedom

of

press

Dem

ocracy

(polity

2)Tax decent.

squared

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

1.51

6***

1.61

8***

1.163***

1.027*

1.38

3***

1.73***

1.703***

1.79

3***

1.56**

*1.11

8**

1.434**

1.471***

1.263*

*1.5***

(0.53)

(0.51)

(0.41)

(0.56)

(0.46)

(0.38)

(0.55)

(0.52)

(0.45)

(0.53)

(0.59)

(0.45)

(0.49)

(0.5)

0.00

50.04

20.021

-0.011

0.01

0.014

-0.043

0.017

0-0.006

0.021

0.03

-0.003

-0.335*

(0.04)

(0.06)

(0.04)

(0.04)

(0.04)

(0.04)

(0.05)

(0.04)

(0.04)

(0.04)

(0.05)

(0.04)

(0.04)

(0.17)

-0.051

*-0.043

-0.053*

-0.05

-0.051

-0.065

**-0.048

-0.06*

-0.05

-0.055**

-0.043

-0.072**

-0.037

-0.03

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

0.04

6-0.155*

-0.046

-0.055

-0.1

-0.108

0.038

0.087

0.028

-0.026

0.068

0.007

-0.102

0.019

(0.14)

(0.09)

(0.08)

(0.08)

(0.08)

(0.1)

(0.14)

(0.11)

(0.09)

(0.12)

(0.17)

(0.12)

(0.09)

(0.12)

-0.642**

-0.886*

-0.576**

-0.953

***

-0.705**

-0.598

**-0.646**

-0.632**

-0.616

**-0.692**

-0.787***

-0.625**

-0.7***

-0.598**

(0.28)

(0.54)

(0.28)

(0.19)

(0.28)

(0.28)

(0.29)

(0.29)

(0.27)

(0.27)

(0.26)

(0.3)

(0.21)

(0.27)

0.318**

0.20

50.244*

0.414*

**0.354**

0.331**

0.33**

0.287**

0.265*

*0.35

6**

0.281**

0.386***

0.331*

*0.337*

**(0.14)

(0.13)

(0.13)

(0.14)

(0.15)

(0.17)

(0.16)

(0.13)

(0.12)

(0.16)

(0.12)

(0.15)

(0.14)

(0.11)

-1.095

-2.358

-1.09*

-1.368

-0.668

-0.294

-1.437

-1.44*

-1.13

-1.574

-1.302

-0.278

-1.903

**-0.687

(0.77)

(2.84)

(0.66)

(0.87)

(0.77)

(0.73)

(0.99)

(0.84)

(0.79)

(0.99)

(1.04)

(0.81)

(0.74)

(0.7)

0.019**

0.01**

0.013*

*0.016*

**0.011**

0.011**

0.013

0.02

4***

0.019*

*0.01

8**

0.023

0.016**

0.013*

*0.01

6**

(0.01)

(0)

(0.01)

(0.01)

(0.01)

(0.01)

(0.01)

(0.01)

(0.01)

(0.01)

(0.01)

(0.01)

(0.01)

(0.01)

0.354*

*-0.147

-0.003

3.238***

3.362

0.566

1.591

-0.762

0.015

0.111*

1.839***

0.008*

(0.15)

(0.21)

(0.38)

(1.2)

(2.47)

(1.94)

(5.03)

(0.99)

(0.02)

(0.06)

(0.62)

(0)

Arellano-Bon

d test for AR(2)

0.72

90.98

10.799

0.726

0.76

60.812

0.72

0.824

0.838

0.51

0.765

0.656

0.455

0.794

Hansen test of overid. rest

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

Num

ber of obs

142

112

142

132

142

142

142

142

142

142

142

121

133

142

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2

.1)

1.41***

1.5***

1.05**

*0.92*

1.27***

1.59***

1.6***

1.67***

1.45**

*1**

1.34**

1.32***

1.19

**1.44

***

(0.521)

(0.475)

(0.397

)(0.521)

(0.437)

(0.359)

(0.538

)(0.503)

(0.43)

(0.508)

(0.58)

(0.442)

(0.461

)(0.483)

Pct 20 (Tax: 4

.8)

1.27**

1.43***

0.91**

0.79

1.13***

1.42***

1.47**

*1.5***

1.32**

*0.85*

1.22**

1.12**

1.09

**1.35

***

(0.52)

(0.455)

(0.389

)(0.483)

(0.429)

(0.342)

(0.541

)(0.497)

(0.412

)(0.489)

(0.584)

(0.438)

(0.435

)(0.475)

Pct 30 (Tax: 5

.8)

1.23**

1.35***

0.86**

0.74

1.09**

1.36***

1.43**

*1.45***

1.27**

*0.8*

1.18**

1.06**

1.05

**1.33

***

(0.523)

(0.437)

(0.39)

(0.474)

(0.431)

(0.34)

(0.545

)(0.499)

(0.41)

(0.485)

(0.588)

(0.439)

(0.429

)(0.477)

Pct 40 (Tax: 7

.7)

1.13**

1.27***

0.75*

0.64

0.99**

1.23***

1.33**

1.33***

1.17**

*0.7

1.1*

0.92**

0.98

**1.27

***

(0.534)

(0.425)

(0.397

)(0.458)

(0.441)

(0.342)

(0.56)

(0.508)

(0.413

)(0.481)

(0.603)

(0.448)

(0.422

)(0.487)

Pct 50 (Tax: 1

0.5)

0.98

*0.84

*0.6

0.5

0.84

*1.05***

1.2**

1.16**

1.03

**0.54

0.98

0.72

0.87

**1.18**

(0.561)

(0.472)

(0.422

)(0.45)

(0.472)

(0.361)

(0.593

)(0.533)

(0.432

)(0.486)

(0.635)

(0.473)

(0.424

)(0.518)

Pct 60 (Tax: 1

7.1)

0.65

0.69

0.25

0.17

0.5

0.62

0.88

0.77

0.7

0.18

0.69

0.24

0.63

0.98

(0.664)

(0.522)

(0.527

)(0.498)

(0.596)

(0.459)

(0.715

)(0.639)

(0.534

)(0.542)

(0.754)

(0.574)

(0.485

)(0.644)

Pct 70 (Tax: 2

0.8)

0.46

0.49

0.05

-0.01

0.31

0.38

0.7

0.55

0.51

-0.02

0.53

-0.02

0.49

0.87

(0.741)

(0.612)

(0.605

)(0.558)

(0.687)

(0.535)

(0.802

)(0.718)

(0.615

)(0.596)

(0.839)

(0.648)

(0.545

)(0.737)

Pct 80 (Tax: 2

8.3)

0.09

0.07

-0.34

-0.39

-0.07

-0.1

0.35

0.1

0.14

-0.42

0.21

-0.55

0.22

0.65

(0.917)

(0.833)

(0.782

)(0.722)

(0.892)

(0.71)

(0.999

)(0.902)

(0.803

)(0.735)

(1.033)

(0.819)

(0.7)

(0.948)

Pct 90 (Tax: 3

3.1)

-0.16

-0.49

-0.6

-0.63

-0.32

-0.41

0.11

-0.19

-0.11

-0.69

0-0.9

0.04

0.5

(1.044)

(1.168)

(0.908

)(0.848)

(1.036)

(0.834)

(1.14)

(1.034)

(0.937

)(0.841)

(1.173)

(0.942)

(0.815

)(1.098)

NOTES:The

proxyfortaxdecentralization

issubnationaltaxrevenue(exceptin

column2).T

heperiod

analyzed

is1986-2010.

The

estimations

includeaconstant

term

andperiod

dummies,which

areom

ittedfor

spaceconsiderations.A

llregressionsareestimated

withtheone-step

system

GMM

estimator

usingtheSTATA

program

xtabond(R

oodm

an,2006),withtheorthogonal

deviations

transform

applied.

For

thefirst-

difference

equation,thesecond

lagof

explanatoryvariables(alltreatedas

endo

genous)areused

asinstruments.Fo

rthelevelequation

,thelagged

first-difference

ofexplanatoryvariablesareused

asinstruments.

The

exogenousvariablesaretheperiod

dummies.

Robuststandard

errors

arein

parentheses.

*,**

and***denote

significance

atthe10,5and1%

levels,respectively.Percentiles

oftaxdecentralization

are

calculated

foraveragevalues

over

theperiod

1986-2010.

Incolumn2percentilesare2.6,

4.3,

6.1,

8,18.1,2

1.5,

26.2,3

5.8and48.8.T

hedefinitionsof

thevariablescanbe

foundin

App

endixI.The

listof

countries

included in

the sample is shown in Table 2.

Total governm

ent revenue

Growth GDP pc

Unemployment rate

Ln GDP pc

Population

Additional controls

Robustness checks: A

dditional control variables. S

ystem GMM estim

ator, 5-year average data.

Dependent variable is public deficit over GDP

Corruption

Tax decentralization

Corruption x Tax decent.

Additional control variables

Using th

e TD1

indicator of tax

decentralization

33

TABLE 7

Corruption indicator : ICRG TI WBGI GCB ICRG TI WBGI GCB(1) (2) (3) (4) (5) (6) (7) (8)

0.62*** 0.549*** 0.642*** 0.665*** 0.593** 0.526** 0.607** 0.721**(0.13) (0.12) (0.13) (0.15) (0.27) (0.25) (0.27) (0.32)

-0.632*** -0.564*** -0.656*** -0.622*** -0.588** -0.525** -0.605** -0.746**(0.13) (0.13) (0.14) (0.15) (0.27) (0.25) (0.27) (0.33)

-0.822*** -0.753*** -0.845*** -0.757*** -0.669** -0.604* -0.685** -0.777**(0.14) (0.13) (0.14) (0.15) (0.32) (0.31) (0.32) (0.33)

-0.162*** -0.161*** -0.162*** -0.132** -0.148 -0.145 -0.146 -0.052(0.05) (0.05) (0.05) (0.05) (0.11) (0.11) (0.11) (0.11)

-0.288*** -0.288*** -0.287*** -0.295*** -0.428** -0.423** -0.427** -0.674***(0.06) (0.06) (0.06) (0.06) (0.19) (0.19) (0.19) (0.22)0.45*** 0.44*** 0.451*** 0.508*** 0.441*** 0.428*** 0.441*** 0.312**(0.05) (0.05) (0.05) (0.05) (0.11) (0.11) (0.11) (0.12)2.142 1.874 2.05 3.527** 4.901* 4.746* 4.847* 2.48(1.43) (1.42) (1.43) (1.65) (2.61) (2.58) (2.6) (3.14)

0.076*** 0.076*** 0.076*** 0.07*** 0.075** 0.075** 0.074** 0.065**(0.02) (0.02) (0.02) (0.01) (0.03) (0.03) (0.03) (0.03)

R-squared 0.76 0.76 0.76 0.77 0.79 0.79 0.79 0.82Number of obs. 640 640 640 586 144 144 144 131

NOTES: The proxy for tax decentralization is subnational tax revenue . The period analyzed is 1986-2010. The estimations includeperiod and country dummies, which are omitted for space considerations. All regressions are estimated with the two-way fixed effectsestimator. Robust standard errors are in parentheses. *, ** and *** denote significance at the 10, 5 and 1% levels, respectively. Thedefinitions of the variables can be found in Appendix I. The list of countries included in the sample is shown in Table 2.

Within-country variations.

Dependent variable is public deficit over GDP

Annual data 5-year average data

Tax decentralization

Ln GDP pc

Population

Tax decent x Interm. corruption

Tax decent x High corruption

Total government revenue

Growth GDP pc

Unemployment rate

Australia

Austria

Belgium

Canada

Switzerland

Czech Republic

Germany

Denmark

Spain

EstoniaFinland

FranceUnited Kingdom

Greece

Hungary

IrelandIceland

IsraelItaly

Japan

Korea, Rep.Luxembourg

Netherlands

Norway

New Zealand

PolandPortugal

Slovak Republic

Slovenia

Sweden

United States

Public deficit = -0.01 + 1.65·Corruption

(t-stat) (-0.01) (2.91)

R-squared = 0.23-7-3.5

03.5

7Pub

lic deficit (% GDP)

0 .5 1 1.5 2 2.5 3

ICRG Corruption

Notes: The variables represent the average of available data over the period 1986-2010.The definitions and sources of the variables can be found in Appendix I.

FIGURE 1 The relationship between public deficit and corruption

34

FIGURE 2 The marginal effect of corruption on public deficit

-8-6

-4-2

02

4Marginal effect o

f corrup

tion on public deficit

0 5 10 15 20 25 30 35 40 45 50Tax decentralization

The figure corresponds to regression 3 of Table 3.90% confidence interval.

Panel A: Tax decentralization

-4-2

02

46

Marginal effect o

f corrup

tion on public deficit

0 5 10 15 20 25 30 35 40 45 50 55 60Expenditure decentralization

The figure corresponds to regression 6 of Table 3.90% confidence interval.

Panel B: Expenditure decentralization

FIGURE 3 The marginal effect of fiscal decentralization on public deficit

-.6

-.4

-.2

0.2

Marginal effect o

f tax decent. on pu

blic deficit

0 .5 1 1.5 2 2.5 3Corruption

The figure corresponds to regression 3 of Table 3.90% confidence interval.

Panel A: Tax decentralization

-.4

-.2

0.2

Marginal effect o

f expend. d

ecent. on pub

lic deficit

0 .5 1 1.5 2 2.5 3Corruption

The figure corresponds to regression 6 of Table 3.90% confidence interval.

Panel B: Expenditure decentralization

35

Appendix I. Description of variables.

VARIABLE DESCRIPTION SOURCE

Banking crisis - fiscal cost

Fiscal costs associated with systemic banking crises.The total fiscal cost is divided by the years of duration of each crisis.

Laeven and Valencia (2012).

Banking crisis - dummyDummy variable indicating whether the country experiences a systemic banking crisis.

Laeven and Valencia (2012).

Bribe paymentsExperience-based indicator obtained from the question: “In the past 12 months, have you or anyone living in your household paid a bribe in any

form? ”

Global Corruption Barometer, Transparency International (several years).

Control of CorruptionPerception-based indicator that measures the level of corruption within the political system. Values range between -2.5 (highly clean) and 2.5 (highly corrupt).

World Bank Governance Indicators (Kaufmann et al. , 2010), from Teorell et al. (2011).

Corruption Perception Index

Perception-based indicator that measures the level of corruption within the political system. Values range between 10 (highly clean) and 0 (highly corrupt).

Transparency International, from Teorell et al. (2011).

Debt service Net government interest payments, as a percentage of GDP. OECD (2012b).

Democracy (polity2)“Revised Combined Polity Score” measures the degree of democracy vs. autocracy in the political system. The scale ranges from -10 to 20, where a higher score means higher democracy.

Polity IV, from Teorell et al. (2011).

Federal Dummy whether the country is a federation. Treisman (2002).

Freedom of pressDegree of freedom of press considering legislative regulation, political pressure, economic influences and repressive actions. The scale ranges from 0 to 100, where a higher score means lower freedom.

Freedom House, from Teorell et al. (2011).

Government fragment.“Probability that two randomly chosen deputies from among the government parties will be of different parties”

Database of Political Institutions, from Teorell et al. (2011).

Growth GDP pc Real per capita GDP growth (annual %).World Development Indicators (WDI), 2011 (World Bank).

i bond Government bond yield at a 10-year maturity. IMF (2012).

ICRG CorruptionPerception-based indicator that assesses the presence of corruption within the political system. We rescale the variable so that higher score means more corruption. The scale ranges from 0 to 6.

ICRG (PRSGroup).

Inflation Ln (1+ inflation), where inflation is consumer prices (annual %). WDI, 2011 (World Bank).Ln GDP pc Logarithm of real per capita GDP (constant 2000 US$). WDI, 2011 (World Bank).Openness Trade (exports and imports) (% of GDP). WDI, 2011 (World Bank).

Parliament fragment.“Probability that two randomly chosen deputies in the legislature belong to different parties.”

Database of Political Institutions, from Teorell et al. (2011).

Population Total population, in millions. WDI, 2011 (World Bank).

Public deficit Government deficit (net borrowing) as a percentage of GDP. OECD (2012b).

Subnational govern. expend.

Subnational (state/regional and local) government expenditure as percentage of total general government expenditure.

OECD (2012a).

Subnational tax revenueSubnational (state/regional and local) tax revenue as percentage of total general government tax revenue.

OECD (2012a).

TD1Subnational (state/regional and local) tax revenue as percentage of total general government tax revenue, considering only those taxes in which subnational units have powers over tax rates and/or tax bases.

Stegarescu (2005) and OECD (2012a).

Total government revenue

Total receipts of general government, as a percentage of GDP. OECD (2012b).

Unemployment rateShare of the labor force without employment but available to work and seeking employment.

WDI, 2011 (World Bank).

36

UNPUBLISHED APPENDICES

FOR THE PAPER:

DOES FISCAL DECENTRALIZATION MITIGATE THE ADVERSE EFFECTS OF CORRUPTION ON PUBLIC DEFICIT?

37

VARIABLE OBS MEAN STD. DEV. MIN MAX

Banking crisis - dummy 775 0.13 0.34 0.00 1.00

Banking crisis - fiscal cost 770 0.36 1.73 0.00 15.60Bribe payments 134 4.19 5.21 0.00 27.00Control of Corruption 341 1.45 0.71 -0.03 2.62Corruption Perception Index 479 7.22 1.80 2.99 10.00

Debt service 680 2.28 2.61 -3.23 12.08Federal 775 0.26 0.44 0.00 1.00

Freedom of press 527 17.94 7.66 5.00 55.00

Government fragment. 714 0.32 0.27 0.00 0.83Growth GDP pc 759 2.01 3.24 -19.73 12.07

i bond 652 6.72 3.50 1.01 40.63ICRG corruption 749 1.41 1.12 0.00 4.00

Inflation 706 1.44 0.82 -2.26 6.32Ln GDP pc 762 9.72 0.66 7.92 10.94

Openness 723 82.29 47.11 16.01 326.76

Parliament fragment. 704 0.69 0.13 0.00 0.90Population 775 31.04 52.93 0.24 309.71Public deficit 681 2.18 4.52 -18.79 31.16Subnational govern. expend. 530 32.11 14.53 4.91 66.45Subnational tax revenue 714 15.83 12.79 0.00 49.45

TD1 403 19.62 16.35 0.05 58.67Total government revenue 681 42.60 8.03 19.66 63.20Unemployment rate 695 7.35 3.91 0.60 23.88

Table 1. Descriptive statistics.

38

ICRG (Refe-rence)

CPI (TI)Control of Corruption (WGI)

GCB (TI)

ICRG (Refe-rence)

CPI (TI)Control of Corruption (WGI)

GCB (TI)

Korea, Rep. 1 1 1 3 3.09 4.65 0.42 1.75Slovenia 1 1 1 1 3.24 6.09 0.98 4.00Italy 1 1 1 1 3.25 4.81 0.43 13.00Estonia 1 1 1 n.a. 3.45 6.05 0.77 n.a.

Slovak Republic 1 1 1 n.a. 3.53 4.13 0.33 n.a.

Poland 1 1 1 1 3.63 4.26 0.37 8.33Czech Republic 1 1 1 1 3.63 4.60 0.41 13.75Greece 1 1 1 1 3.86 4.56 0.39 20.00Hungary 1 1 1 1 4.03 5.07 0.64 19.00Israel 1 2 1 2 4.04 6.78 1.00 3.00

Ireland 2 2 2 2 4.05 7.66 1.61 3.00Japan 2 2 2 1 4.09 6.95 1.21 3.50Spain 2 2 2 2 4.12 6.51 1.25 3.00Portugal 2 1 2 2 4.40 6.44 1.20 2.25Belgium 2 2 2 n.a. 4.45 6.71 1.43 n.a.

France 2 2 2 2 4.50 6.90 1.41 3.33United States 2 2 2 2 4.56 7.52 1.58 2.75Austria 2 2 2 1 4.85 7.96 1.92 3.50Australia 2 3 2 2 4.87 8.67 1.94 2.00United Kingdom 2 2 2 2 4.89 8.43 1.95 2.00

Germany 3 2 2 2 4.96 7.89 1.89 2.00Switzerland 3 3 3 3 5.29 8.84 2.14 1.25Norway 3 3 3 3 5.46 8.75 2.04 1.67Luxembourg 3 3 3 1 5.47 8.58 2.00 8.00Canada 3 3 3 2 5.59 8.84 2.00 2.50New Zealand 3 3 3 1 5.63 9.44 2.30 4.00Netherlands 3 3 3 3 5.66 8.87 2.14 1.75Sweden 3 3 3 3 5.70 9.27 2.22 1.00Iceland 3 3 3 2 5.74 9.28 2.28 2.00Denmark 3 3 3 3 5.81 9.58 2.28 1.25Finland 3 3 3 3 6.00 9.53 2.38 1.75

Matches with ICRG -- 87.1% 96.8% 67.9%Correlations with ICRG -- 0.92 0.94 -0.48

Table 2. Classification of countries into corruption groups (used in Section 5.2)Corruption groups Scores of corruption indicators

Notes: Corruption groups: 1: High corruption; 2: Intermediate corruption; 3: Low corruption. ICRG (Reference) representsthe average over the period 1984-2010. CPI (TI) and Control of Corruption (WGI) represent the average over the period1996-2009. GCB (TI) represents the average over the period 2006-2010. All indicators imply the absence of corruption,except GCB (TI) that means the opposite (note that in this case ICRG corruption is not rescaled).

39

Reference

(T.4-C.4)

Debt service

i bond

Banking

crisis - fiscal

cost

Banking

crisis

dummy

Federal

Gov

ernm

ent

fragment.

Parliament

fragment.

Inflation

Openness

Freedom

of

press

Dem

ocracy

(polity2)

Tax decent.

squared

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

1.41

3***

0.899**

1.247*

**1.467***

1.43

3***

1.4***

1.382*

**1.523***

1.31

6***

1.441*

**1.38

3***

1.565*

**1.25**

*(0.47)

(0.39)

(0.48)

(0.47)

(0.44)

(0.47)

(0.46)

(0.49)

(0.48)

(0.44)

(0.43)

(0.53)

(0.47)

0.02

70.03

50.019

0.026

0.02

80.00

40.029

0.037

0.02

80.02

0.05

40.027

-0.226**

(0.04)

(0.04)

(0.04)

(0.04)

(0.04)

(0.04)

(0.04)

(0.04)

(0.04)

(0.05)

(0.06)

(0.05)

(0.11)

-0.055*

-0.043

-0.051*

-0.051

-0.056

*-0.054

-0.051

-0.059

*-0.06**

-0.056*

-0.067

*-0.04

-0.03

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.03)

(0.04)

(0.03)

(0.03)

-0.06

-0.094**

-0.076

-0.068

-0.089

-0.036

-0.062

-0.046

-0.068

-0.056

-0.004

-0.055

-0.022

(0.06)

(0.05)

(0.06)

(0.07)

(0.07)

(0.06)

(0.06)

(0.06)

(0.06)

(0.06)

(0.06)

(0.06)

(0.06)

-0.536**

*-0.366

***

-0.539**

*-0.498***

-0.477

***

-0.522

***

-0.49*

**-0.496***

-0.537

***

-0.493**

*-0.502**

*-0.591**

*-0.466***

(0.11)

(0.11)

(0.12)

(0.12)

(0.11)

(0.12)

(0.1)

(0.11)

(0.11)

(0.11)

(0.13)

(0.11)

(0.13)

0.36

3***

0.19

3***

0.353*

**0.367***

0.37

8***

0.32

8***

0.344*

**0.298***

0.31

4***

0.333*

**0.29

9***

0.28

1**

0.335***

(0.11)

(0.07)

(0.12)

(0.13)

(0.13)

(0.11)

(0.1)

(0.1)

(0.11)

(0.11)

(0.1)

(0.11)

(0.09)

-1.08

-1.118*

-1.588

-0.925

-0.88

-1.292

-1.308

-1.373

*-1.863*

-1.128

-0.611

-2.292**

*-1.021

(0.75)

(0.67)

(1.15)

(0.83)

(0.77)

(0.8)

(0.8)

(0.8)

(0.99)

(0.89)

(0.67)

(0.83)

(0.76)

0.01

4***

0.011**

0.014*

**0.013***

0.012**

0.01

4***

0.017*

**0.018***

0.01

6***

0.01

5**

0.01

5***

0.014*

**0.016***

(0.0)

(0.0)

(0)

(0)

(0)

(0.01)

(0.01)

(0.01)

(0.0)

(0.01)

(0)

(0.01)

(0.0)

0.67

3***

-0.28

0.187

2.42

2***

1.24

21.586

1.576

-0.946

-0.002

0.12

1**

1.508*

**0.006*

*(0.11)

(0.21)

(0.26)

(0.85)

(1.04)

(1.19)

(2.96)

(0.65)

(0.01)

(0.06)

(0.33)

(0)

Arellano-Bond test for AR(2)

0.10

90.25

60.255

0.092

0.05

50.10

70.106

0.11

0.03

20.076

0.26

70.034

0.104

Hansen test of overid. rest

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

1.00

11.00

1.00

Num

ber of obs

632

631

591

627

632

632

632

622

629

630

501

564

632

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2

.1)

1.3***

0.81**

1.14**

1.36**

*1.32***

1.29***

1.27

***

1.4*

**1.19***

1.32

***

1.24

***

1.48

***

1.19**

*(0.442)

(0.377)

(0.452)

(0.437

)(0.413)

(0.446)

(0.423)

(0.45)

(0.457)

(0.419)

(0.398)

(0.49)

(0.445

)Pct 20 (Tax: 4

.8)

1.15***

0.69

*1**

1.22**

*1.16***

1.14***

1.13

***

1.24**

*1.02**

1.17

***

1.06

***

1.37

***

1.11**

*(0.415)

(0.371)

(0.425)

(0.401

)(0.385)

(0.425)

(0.396)

(0.413

)(0.431)

(0.398)

(0.371)

(0.45)

(0.423

)Pct 30 (Tax: 5

.8)

1.1***

0.65

*0.95**

1.17**

*1.11***

1.09**

1.09

***

1.18**

*0.97**

1.12

***

1***

1.33

***

1.08

**(0.41)

(0.372)

(0.419)

(0.393

)(0.38)

(0.421)

(0.391)

(0.404

)(0.425)

(0.395)

(0.368)

(0.439)

(0.419

)Pct 40 (Tax: 7

.7)

0.99**

0.56

0.86**

1.07**

*1***

0.98**

0.99**

1.07**

*0.85**

1.01**

0.86**

1.25

***

1.02

**(0.405)

(0.381)

(0.412)

(0.384

)(0.375)

(0.421)

(0.388)

(0.392

)(0.417)

(0.396)

(0.37)

(0.42)

(0.416

)Pct 50 (Tax: 1

0.5)

0.84**

0.44

0.71*

0.93

**0.85

**0.83

*0.84**

0.9**

0.68

0.85**

0.67

*1.14

***

0.94

**(0.415)

(0.406)

(0.416)

(0.39)

(0.386)

(0.439)

(0.403)

(0.393

)(0.416)

(0.411)

(0.394)

(0.404)

(0.425

)Pct 60 (Tax: 1

7.1)

0.48

0.16

0.38

0.59

0.48

0.47

0.51

0.51

0.28

0.47

0.23

0.87**

0.74

(0.505)

(0.508)

(0.484)

(0.481

)(0.477)

(0.542)

(0.507)

(0.471

)(0.466)

(0.508)

(0.528)

(0.428)

(0.502

)Pct 70 (Tax: 2

0.8)

0.28

00.19

0.4

0.27

0.27

0.32

0.29

0.06

0.26

-0.02

0.72

0.63

(0.584)

(0.583)

(0.549)

(0.565

)(0.556)

(0.627)

(0.593)

(0.548

)(0.52)

(0.586)

(0.629)

(0.474)

(0.57)

Pct 80 (Tax: 2

8.3)

-0.13

-0.33

-0.19

0.02

-0.14

-0.13

-0.06

-0.15

-0.39

-0.15

-0.52

0.42

0.41

(0.771)

(0.752)

(0.712)

(0.765

)(0.742)

(0.823)

(0.794)

(0.738

)(0.661)

(0.769)

(0.856)

(0.612)

(0.735

)Pct 90 (Tax: 3

3.1)

-0.4

-0.54

-0.44

-0.22

-0.41

-0.39

-0.31

-0.44

-0.69

-0.43

-0.85

0.23

0.26

(0.908)

(0.872)

(0.833)

(0.909

)(0.876)

(0.965)

(0.937)

(0.877

)(0.768)

(0.901)

(1.016)

(0.722)

(0.857

)

Notes:The

proxyfortaxdecentralization

issubnationaltaxrevenue.The

period

analyzed

is19

86-2010.

The

estimations

includeaconstant

term

andperiod

dummies,which

areom

ittedforspace

considerations.A

llregressionsareestimated

withtheone-step

system

GMM

estimator

usingtheSTATAprogram

xtabond(R

oodm

an,2006),withtheorthogonal

deviations

transform

applied.

For

thefirst-difference

equation

,thesecond

lagof

explanatoryvariables(alltreatedas

endogenous)areused

asinstruments.For

thelevelequation

,thelagged

first-difference

ofexplanatoryvariables

areused

asinstruments.The

exogenou

svariablesaretheperiod

dummies.Robuststandard

errors

arein

parentheses.*,

**and***denote

significance

atthe10,5and1%

levels,respectively.

Percentiles

oftaxdecentralization

arecalculated

foraveragevalues

over

theperiod

1986-201

0.The

definition

sof

thevariablescanbe

foundin

AppendixI.The

listof

countriesincluded

inthe

sample is shown in Table 2.

Total governm

ent revenue

Growth GDP pc

Unemployment rate

Ln GDP pc

Pop

ulation

Add

itional con

trols

Corruption x Tax decent.

Table 3. R

obustness checks: A

dditional con

trol variables. R

eference regression: Table 4- Colum

n 4. Annual d

ata.

Dependent variable is public deficit over GDP

Add

itional con

trol variables

Corruption

Tax decentralization

40

Reference (T.4-C.4)

AUS AUT BEL CAN CHE CZE DEU DNK

(1) (2) (3) (4) (5) (6) (7) (8) (9)1.413*** 1.443*** 1.338*** 1.536*** 1.359*** 1.457*** 1.368*** 1.482*** 1.45***(0.47) (0.48) (0.48) (0.46) (0.51) (0.48) (0.49) (0.48) (0.47)0.027 0.036 0.034 0.036 0.017 0.018 0.024 0.029 0.032(0.04) (0.04) (0.05) (0.04) (0.05) (0.05) (0.04) (0.05) (0.04)-0.055* -0.057* -0.057* -0.06* -0.051 -0.06* -0.054* -0.058* -0.06*(0.03) (0.03) (0.03) (0.03) (0.04) (0.03) (0.03) (0.03) (0.03)-0.06 -0.075 -0.082 -0.053 -0.055 -0.041 -0.064 -0.06 -0.06(0.06) (0.07) (0.06) (0.06) (0.06) (0.07) (0.06) (0.06) (0.07)

-0.536*** -0.531*** -0.528*** -0.545*** -0.521*** -0.517*** -0.556*** -0.521*** -0.575***(0.11) (0.11) (0.11) (0.11) (0.12) (0.12) (0.13) (0.11) (0.12)

0.363*** 0.366*** 0.407*** 0.328*** 0.353*** 0.352*** 0.363*** 0.341*** 0.339***(0.11) (0.12) (0.12) (0.11) (0.11) (0.11) (0.12) (0.11) (0.1)-1.08 -0.98 -1.087 -1.196 -1.071 -1.222 -1.097 -1.123 -1.193(0.75) (0.77) (0.78) (0.78) (0.77) (0.76) (0.87) (0.76) (0.75)

0.014*** 0.013*** 0.015*** 0.016*** 0.015*** 0.018*** 0.014*** 0.015*** 0.015***(0.0) (0.01) (0.01) (0.01) (0.01) (0.01) (0.0) (0.0) (0.0)

Arellano-Bond test for AR(2) 0.109 0.121 0.12 0.109 0.119 0.123 0.113 0.15 0.11Hansen test of overid. rest 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Number of obs 632 611 608 608 608 612 617 613 608

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2.1) 1.3*** 1.32*** 1.22*** 1.41*** 1.25*** 1.33*** 1.25*** 1.36*** 1.32***(0.442) (0.45) (0.447) (0.432) (0.465) (0.449) (0.459) (0.45) (0.438)

Pct 20 (Tax: 4.8) 1.15*** 1.17*** 1.06** 1.24*** 1.11*** 1.17*** 1.11** 1.2*** 1.16***(0.415) (0.419) (0.421) (0.409) (0.427) (0.421) (0.428) (0.421) (0.411)

Pct 30 (Tax: 5.8) 1.1*** 1.11*** 1.01** 1.19*** 1.06** 1.11*** 1.06** 1.15*** 1.1***(0.41) (0.413) (0.417) (0.405) (0.419) (0.416) (0.422) (0.415) (0.406)

Pct 40 (Tax: 7.7) 0.99** 1** 0.9** 1.07*** 0.96** 1** 0.95** 1.03** 0.98**(0.405) (0.405) (0.414) (0.403) (0.413) (0.414) (0.414) (0.409) (0.402)

Pct 50 (Tax: 10.5) 0.84** 0.84** 0.74* 0.9** 0.82* 0.83* 0.8* 0.87** 0.81**(0.415) (0.411) (0.426) (0.418) (0.426) (0.43) (0.42) (0.419) (0.414)

Pct 60 (Tax: 17.1) 0.48 0.46 0.36 0.5 0.48 0.43 0.45 0.48 0.42(0.505) (0.493) (0.523) (0.516) (0.546) (0.539) (0.501) (0.513) (0.509)

Pct 70 (Tax: 20.8) 0.28 0.25 0.15 0.28 0.29 0.21 0.25 0.27 0.19(0.584) (0.568) (0.605) (0.597) (0.647) (0.629) (0.576) (0.594) (0.59)

Pct 80 (Tax: 28.3) -0.13 -0.18 -0.27 -0.17 -0.09 -0.24 -0.15 -0.17 -0.25(0.771) (0.751) (0.8) (0.787) (0.882) (0.84) (0.761) (0.79) (0.782)

Pct 90 (Tax: 33.1) -0.4 -0.46 -0.55 -0.47 -0.34 -0.53 -0.41 -0.45 -0.55(0.908) (0.885) (0.94) (0.925) (1.05) (0.991) (0.895) (0.932) (0.921)

Notes: The proxy for tax decentralization is subnational tax revenue. The period analyzed is 1986-2010. The estimations include a constant termand period dummies, which are omitted for space considerations. All regressions are estimated with the one-step system GMM estimator usingthe STATA program xtabond (Roodman, 2006), with the orthogonal deviations transform applied. For the first-difference equation, the secondlag of explanatory variables (all treated as endogenous) are used as instruments. For the level equation, the lagged first-difference of explanatoryvariables are used as instruments. The exogenous variables are the period dummies. Robust standard errors are in parentheses. *, ** and ***denote significance at the 10, 5 and 1% levels, respectively. Percentiles of tax decentralization are calculated for average values over the period1986-2010. The definitions of the variables can be found in Appendix I. The list of countries included in the sample is shown in Table 2.

Table 4.1. Robustness checks: Outliers. Reference regression: Table 4- Column 4. Annual data.Dependent variable is public deficit over GDP

Excluded country

Corruption

Tax decentralization

Corruption x Tax decent.

Total government revenue

Growth GDP pc

Unemployment rate

Ln GDP pc

Population

41

ESP EST FIN FRA GBR GRC HUN IRL

(10) (11) (12) (13) (14) (15) (16) (17)1.379*** 1.401*** 1.477*** 1.475*** 1.44*** 1.177** 1.467*** 1.881***(0.47) (0.47) (0.51) (0.46) (0.48) (0.47) (0.48) (0.48)0.018 0.027 0.028 0.018 0.02 0.024 0.026 0.028(0.04) (0.04) (0.04) (0.05) (0.04) (0.04) (0.04) (0.05)-0.052 -0.049 -0.06* -0.054* -0.053* -0.043 -0.054* -0.071**(0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03)-0.057 -0.063 -0.065 -0.053 -0.06 -0.055 -0.072 -0.064(0.06) (0.06) (0.07) (0.07) (0.06) (0.06) (0.06) (0.06)

-0.509*** -0.624*** -0.527*** -0.544*** -0.525*** -0.517*** -0.523*** -0.418***(0.11) (0.11) (0.13) (0.12) (0.12) (0.11) (0.12) (0.1)

0.401*** 0.357*** 0.365*** 0.339*** 0.369*** 0.357*** 0.39*** 0.374***(0.11) (0.12) (0.12) (0.11) (0.11) (0.11) (0.11) (0.12)-0.929 -1.359* -0.959 -1.052 -0.928 -1.21 -0.533 -0.343(0.79) (0.74) (0.76) (0.78) (0.76) (0.77) (0.79) (0.81)

0.015*** 0.013*** 0.014*** 0.015*** 0.014*** 0.014*** 0.013*** 0.014***(0.01) (0.0) (0.0) (0.01) (0.0) (0.0) (0.0) (0.0)

Arellano-Bond test for AR(2) 0.13 0.09 0.082 0.103 0.107 0.143 0.118 0.119Hansen test of overid. rest 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Number of obs 608 621 608 608 608 616 617 612

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2.1) 1.27*** 1.3*** 1.35*** 1.36*** 1.33*** 1.09** 1.35*** 1.73***(0.435) (0.439) (0.475) (0.433) (0.452) (0.441) (0.447) (0.44)

Pct 20 (Tax: 4.8) 1.13*** 1.16*** 1.19*** 1.22*** 1.18*** 0.97** 1.2*** 1.54***(0.409) (0.414) (0.441) (0.408) (0.422) (0.417) (0.414) (0.406)

Pct 30 (Tax: 5.8) 1.08*** 1.12*** 1.13*** 1.17*** 1.14*** 0.93** 1.15*** 1.47***(0.404) (0.409) (0.434) (0.403) (0.416) (0.412) (0.407) (0.398)

Pct 40 (Tax: 7.7) 0.97** 1.02** 1.01** 1.06*** 1.03** 0.85** 1.05*** 1.34***(0.402) (0.405) (0.424) (0.4) (0.409) (0.408) (0.397) (0.39)

Pct 50 (Tax: 10.5) 0.83** 0.88** 0.84** 0.91** 0.88** 0.73* 0.89** 1.14***(0.417) (0.416) (0.428) (0.412) (0.415) (0.418) (0.401) (0.396)

Pct 60 (Tax: 17.1) 0.48 0.56 0.45 0.56 0.53 0.44 0.54 0.67(0.522) (0.503) (0.507) (0.505) (0.497) (0.502) (0.48) (0.487)

Pct 70 (Tax: 20.8) 0.29 0.38 0.23 0.36 0.34 0.28 0.33 0.41(0.608) (0.578) (0.583) (0.583) (0.571) (0.576) (0.555) (0.569)

Pct 80 (Tax: 28.3) -0.1 0.01 -0.22 -0.04 -0.06 -0.03 -0.07 -0.11(0.81) (0.758) (0.772) (0.769) (0.753) (0.752) (0.739) (0.765)

Pct 90 (Tax: 33.1) -0.36 -0.23 -0.51 -0.3 -0.32 -0.24 -0.33 -0.46(0.954) (0.889) (0.911) (0.904) (0.886) (0.881) (0.873) (0.908)

Growth GDP pc

Unemployment rate

Ln GDP pc

Table 4.1 (Continued). Robustness checks: Outliers. Reference regression: Table 4- Column 4. Annual data.Dependent variable is public deficit over GDP

Excluded country

Corruption

Tax decentralization

Population

Corruption x Tax decent.

Total government revenue

42

ISL ISR ITA JPN KOR LUX NLD NOR

(18) (19) (20) (21) (22) (23) (24) (25)1.466*** 1.264*** 1.432*** 1.417*** 1.476*** 1.439*** 1.589*** 1.334***(0.46) (0.47) (0.5) (0.49) (0.49) (0.5) (0.45) (0.46)0.021 0.028 0.028 0.05 0.029 0.02 0.05 0.003(0.05) (0.04) (0.04) (0.04) (0.04) (0.05) (0.04) (0.03)-0.052 -0.053 -0.052 -0.079*** -0.042 -0.056* -0.063* -0.043(0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03)-0.055 -0.063 -0.057 -0.057 -0.09 -0.064 -0.066 -0.004(0.06) (0.06) (0.06) (0.06) (0.06) (0.06) (0.06) (0.04)

-0.559*** -0.521*** -0.545*** -0.469*** -0.472*** -0.525*** -0.52*** -0.461***(0.12) (0.11) (0.11) (0.1) (0.11) (0.12) (0.11) (0.1)

0.386*** 0.348*** 0.35*** 0.367*** 0.259*** 0.375*** 0.36*** 0.308***(0.12) (0.11) (0.11) (0.1) (0.09) (0.12) (0.11) (0.09)-1.141 -1.25* -1.097 -1.415** -1.519** -0.799 -1.188 -0.705(0.73) (0.73) (0.73) (0.7) (0.73) (0.99) (0.74) (0.71)

0.016*** 0.015*** 0.014*** 0.015*** 0.012** 0.014*** 0.015*** 0.014***(0.01) (0.0) (0.0) (0.0) (0.0) (0.01) (0.0) (0.0)

Arellano-Bond test for AR(2) 0.171 0.118 0.091 0.024 0.123 0.111 0.119 0.198Hansen test of overid. rest 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Number of obs 613 620 608 608 612 612 609 608

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2.1) 1.36*** 1.15*** 1.32*** 1.25*** 1.39*** 1.32*** 1.46*** 1.24***(0.428) (0.441) (0.475) (0.455) (0.452) (0.465) (0.42) (0.438)

Pct 20 (Tax: 4.8) 1.21*** 1.01** 1.18*** 1.03** 1.28*** 1.17*** 1.28*** 1.12***(0.397) (0.416) (0.452) (0.421) (0.416) (0.431) (0.394) (0.421)

Pct 30 (Tax: 5.8) 1.17*** 0.96** 1.13** 0.96** 1.24*** 1.12*** 1.23*** 1.09***(0.391) (0.411) (0.448) (0.413) (0.406) (0.423) (0.389) (0.418)

Pct 40 (Tax: 7.7) 1.06*** 0.86** 1.03** 0.8** 1.16*** 1.01** 1.1*** 1**(0.384) (0.408) (0.446) (0.4) (0.392) (0.413) (0.386) (0.417)

Pct 50 (Tax: 10.5) 0.92** 0.71* 0.88* 0.58 1.04*** 0.85** 0.93** 0.88**(0.392) (0.422) (0.458) (0.395) (0.387) (0.415) (0.399) (0.428)

Pct 60 (Tax: 17.1) 0.57 0.36 0.54 0.06 0.76* 0.49 0.51 0.59(0.48) (0.518) (0.549) (0.444) (0.443) (0.491) (0.5) (0.503)

Pct 70 (Tax: 20.8) 0.38 0.16 0.35 -0.24 0.61 0.28 0.28 0.43(0.558) (0.6) (0.626) (0.501) (0.506) (0.565) (0.584) (0.567)

Pct 80 (Tax: 28.3) -0.01 -0.23 -0.04 -0.83 0.3 -0.13 -0.19 0.11(0.747) (0.793) (0.813) (0.653) (0.671) (0.749) (0.78) (0.722)

Pct 90 (Tax: 33.1) -0.26 -0.49 -0.29 -1.22 0.1 -0.4 -0.5 -0.1(0.883) (0.932) (0.95) (0.769) (0.795) (0.885) (0.92) (0.836)

Table 4.1 (Continued). Robustness checks: Outliers. Reference regression: Table 4- Column 4. Annual data.Dependent variable is public deficit over GDP

Excluded country

Corruption

Tax decentralization

Corruption x Tax decent.

Total government revenue

Growth GDP pc

Unemployment rate

Ln GDP pc

Population

43

NZL POL PRT SVK SVN SWE USA

(26) (27) (28) (29) (30) (31) (32) (33)0.972** 1.406*** 1.521*** 1.569*** 1.316*** 1.38*** 1.504*** 1.308***(0.45) (0.47) (0.46) (0.47) (0.46) (0.47) (0.48) (0.48)0.005 0.028 0.035 0.032 0.023 0.012 0.029 0.013(0.04) (0.04) (0.05) (0.05) (0.04) (0.04) (0.04) (0.04)-0.046 -0.051* -0.057* -0.064* -0.054* -0.051 -0.064** -0.05*(0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03)-0.073 -0.064 -0.045 -0.073 -0.067 -0.058 -0.048 -0.069(0.06) (0.06) (0.06) (0.07) (0.06) (0.07) (0.06) (0.07)

-0.563*** -0.55*** -0.518*** -0.524*** -0.554*** -0.531*** -0.526*** -0.586***(0.12) (0.12) (0.12) (0.12) (0.12) (0.12) (0.11) (0.12)

0.354*** 0.391*** 0.348*** 0.401*** 0.371*** 0.328*** 0.328*** 0.358***(0.12) (0.12) (0.11) (0.14) (0.12) (0.11) (0.1) (0.12)-1.459* -1.161 -0.997 -0.883 -1.086 -1.137 -1.167 -1.158(0.78) (0.84) (0.85) (0.76) (0.77) (0.74) (0.76) (0.81)

0.015*** 0.014*** 0.015*** 0.014*** 0.015*** 0.016*** 0.024*** 0.014***(0.0) (0.0) (0.0) (0.0) (0.0) (0.01) (0.01) (0.0)

Arellano-Bond test for AR(2) 0.099 0.106 0.166 0.15 0.116 0.122 0.113 0.102Hansen test of overid. rest 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Number of obs 608 617 608 617 621 608 608 613

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2.1) 0.88** 1.3*** 1.4*** 1.43*** 1.2*** 1.27*** 1.37*** 1.2***(0.42) (0.443) (0.429) (0.433) (0.439) (0.44) (0.442) (0.454)

Pct 20 (Tax: 4.8) 0.75* 1.16*** 1.24*** 1.26*** 1.06** 1.13*** 1.2*** 1.07**(0.395) (0.42) (0.399) (0.404) (0.418) (0.412) (0.408) (0.436)

Pct 30 (Tax: 5.8) 0.71* 1.11*** 1.19*** 1.2*** 1.01** 1.09*** 1.14*** 1.02**(0.39) (0.416) (0.393) (0.398) (0.415) (0.406) (0.4) (0.433)

Pct 40 (Tax: 7.7) 0.62 1.01** 1.08*** 1.07*** 0.9** 0.99** 1.01*** 0.92**(0.387) (0.414) (0.387) (0.394) (0.416) (0.401) (0.39) (0.433)

Pct 50 (Tax: 10.5) 0.49 0.87** 0.92** 0.89** 0.75* 0.84** 0.83** 0.78*(0.399) (0.426) (0.397) (0.407) (0.432) (0.411) (0.393) (0.447)

Pct 60 (Tax: 17.1) 0.19 0.53 0.54 0.47 0.4 0.5 0.41 0.45(0.494) (0.515) (0.49) (0.51) (0.53) (0.501) (0.473) (0.534)

Pct 70 (Tax: 20.8) 0.02 0.34 0.33 0.23 0.2 0.31 0.17 0.26(0.573) (0.591) (0.571) (0.596) (0.61) (0.579) (0.549) (0.606)

Pct 80 (Tax: 28.3) -0.32 -0.04 -0.1 -0.25 -0.2 -0.07 -0.3 -0.11(0.761) (0.772) (0.764) (0.798) (0.798) (0.767) (0.735) (0.779)

Pct 90 (Tax: 33.1) -0.55 -0.29 -0.38 -0.56 -0.46 -0.32 -0.61 -0.35(0.897) (0.903) (0.904) (0.944) (0.934) (0.903) (0.872) (0.905)

Table 4.1 (Continued). Robustness checks: Outliers. Reference regression: Table 4- Column 4. Annual data.Dependent variable is public deficit over GDP

Excluded country Observations excluded if stand. resid. >|2|

Corruption

Ln GDP pc

Population

Tax decentralization

Corruption x Tax decent.

Total government revenue

Growth GDP pc

Unemployment rate

44

Reference (T.4-C.8)

AUS AUT BEL CAN CHE CZE DEU DNK

(1) (2) (3) (4) (5) (6) (7) (8) (9)1.516*** 1.447*** 1.593*** 1.593*** 1.496*** 1.474*** 1.855*** 1.631*** 1.53***(0.53) (0.5) (0.57) (0.52) (0.53) (0.52) (0.58) (0.55) (0.57)0.005 0.005 0.014 0.009 -0.023 -0.029 0.024 0.008 0.009(0.04) (0.04) (0.04) (0.04) (0.06) (0.07) (0.04) (0.05) (0.04)-0.051* -0.051* -0.056* -0.058* -0.039 -0.051 -0.065** -0.049 -0.051*(0.03) (0.03) (0.03) (0.03) (0.04) (0.05) (0.03) (0.03) (0.03)0.046 0.007 0.031 0.032 0.054 0.136 0.047 0.1 0.078(0.14) (0.1) (0.13) (0.13) (0.15) (0.13) (0.13) (0.13) (0.15)

-0.642** -0.681*** -0.65** -0.68** -0.646** -0.469 -0.544** -0.572* -0.639**(0.28) (0.26) (0.28) (0.28) (0.29) (0.29) (0.26) (0.3) (0.28)0.318** 0.35** 0.323** 0.329** 0.303** 0.315* 0.292** 0.287** 0.291**(0.14) (0.15) (0.15) (0.14) (0.13) (0.19) (0.14) (0.14) (0.14)-1.095 -0.968 -1.154 -0.991 -1.066 -1.412** -0.841 -1.187 -1.225(0.77) (0.71) (0.77) (0.75) (0.81) (0.65) (0.81) (0.79) (0.79)0.019** 0.016** 0.018** 0.019** 0.019* 0.027*** 0.019** 0.02** 0.02**(0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01)

Arellano-Bond test for AR(2) 0.729 0.679 0.753 0.692 0.686 0.626 0.688 0.79 0.672Hansen test of overid. rest 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Number of obs 142 137 137 137 137 137 138 138 137

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2.1) 1.41*** 1.34*** 1.47*** 1.47*** 1.41*** 1.37*** 1.72*** 1.53*** 1.42**(0.521) (0.486) (0.558) (0.508) (0.518) (0.486) (0.563) (0.544) (0.563)

Pct 20 (Tax: 4.8) 1.27** 1.2** 1.32** 1.31** 1.31** 1.23*** 1.54*** 1.4** 1.28**(0.52) (0.484) (0.56) (0.507) (0.52) (0.472) (0.548) (0.548) (0.562)

Pct 30 (Tax: 5.8) 1.23** 1.15** 1.27** 1.26** 1.27** 1.18** 1.48*** 1.35** 1.24**(0.523) (0.486) (0.563) (0.51) (0.524) (0.476) (0.545) (0.553) (0.564)

Pct 40 (Tax: 7.7) 1.13** 1.05** 1.16** 1.15** 1.19** 1.08** 1.35** 1.26** 1.14**(0.534) (0.497) (0.573) (0.521) (0.541) (0.498) (0.546) (0.568) (0.574)

Pct 50 (Tax: 10.5) 0.98* 0.91* 1* 0.99* 1.08* 0.93* 1.17** 1.12* 0.99*(0.561) (0.525) (0.598) (0.547) (0.578) (0.559) (0.558) (0.6) (0.597)

Pct 60 (Tax: 17.1) 0.65 0.57 0.63 0.6 0.82 0.59 0.74 0.8 0.65(0.664) (0.635) (0.691) (0.649) (0.715) (0.782) (0.638) (0.717) (0.69)

Pct 70 (Tax: 20.8) 0.46 0.38 0.42 0.39 0.67 0.4 0.5 0.62 0.46(0.741) (0.714) (0.759) (0.723) (0.812) (0.933) (0.707) (0.799) (0.76)

Pct 80 (Tax: 28.3) 0.09 0 0.01 -0.04 0.38 0.02 0.01 0.26 0.08(0.917) (0.897) (0.919) (0.895) (1.029) (1.261) (0.875) (0.987) (0.924)

Pct 90 (Tax: 33.1) -0.16 -0.25 -0.27 -0.32 0.19 -0.23 -0.31 0.02 -0.17(1.044) (1.028) (1.036) (1.018) (1.182) (1.486) (1.001) (1.121) (1.044)

Notes: The proxy for tax decentralization is subnational tax revenue. The period analyzed is 1986-2010. The estimations include a constant termand period dummies, which are omitted for space considerations. All regressions are estimated with the one-step system GMM estimator usingthe STATA program xtabond (Roodman, 2006), with the orthogonal deviations transform applied. For the first-difference equation, the secondlag of explanatory variables (all treated as endogenous) are used as instruments. For the level equation, the lagged first-difference of explanatoryvariables are used as instruments. The exogenous variables are the period dummies. Robust standard errors are in parentheses. *, ** and ***denote significance at the 10, 5 and 1% levels, respectively. Percentiles of tax decentralization are calculated for average values over the period1986-2010. The definitions of the variables can be found in Appendix I. The list of countries included in the sample is shown in Table 2.

Table 4. 2. Robustness checks: Outliers. Reference regression: Table 4- Column 8. 5-year average data.Dependent variable is public deficit over GDP

Excluded country

Corruption

Tax decentralization

Corruption x Tax decent.

Total government revenue

Growth GDP pc

Unemployment rate

Ln GDP pc

Population

45

ESP EST FIN FRA GBR GRC HUN IRL

(10) (11) (12) (13) (14) (15) (16) (17)1.557** 1.87*** 1.264** 1.511*** 1.494*** 1.268** 1.49*** 1.978***(0.66) (0.6) (0.52) (0.53) (0.52) (0.57) (0.51) (0.6)0.003 0.027 0.002 0.001 0 0.007 0.006 0.033(0.05) (0.04) (0.04) (0.04) (0.05) (0.04) (0.04) (0.04)-0.05 -0.066** -0.051* -0.051* -0.043 -0.047 -0.056* -0.07**(0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03)0.078 0.004 0.053 0.016 0.062 0.013 -0.049 -0.037(0.17) (0.12) (0.14) (0.13) (0.14) (0.12) (0.1) (0.15)-0.618* -0.806*** -0.6* -0.706** -0.613** -0.635** -0.833*** -0.704***(0.33) (0.31) (0.31) (0.28) (0.3) (0.28) (0.26) (0.22)0.294* 0.338** 0.347*** 0.37** 0.313** 0.347*** 0.455** 0.326**(0.16) (0.14) (0.13) (0.14) (0.14) (0.13) (0.19) (0.16)-1.164 -1.039 -1.137 -0.901 -1.157 -0.927 -0.221 -0.606(1) (0.76) (0.81) (0.74) (0.81) (0.76) (0.94) (0.98)

0.02* 0.016** 0.019** 0.017** 0.019** 0.017** 0.013** 0.014(0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01)

Arellano-Bond test for AR(2) 0.811 0.791 0.541 0.64 0.772 0.65 0.223 0.848Hansen test of overid. rest 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Number of obs 137 139 137 137 137 138 138 137

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2.1) 1.45** 1.73*** 1.16** 1.4*** 1.4*** 1.17** 1.37*** 1.83***(0.633) (0.581) (0.516) (0.517) (0.517) (0.549) (0.48) (0.578)

Pct 20 (Tax: 4.8) 1.31** 1.55*** 1.02* 1.26** 1.29** 1.04** 1.22*** 1.64***(0.616) (0.563) (0.521) (0.512) (0.522) (0.527) (0.458) (0.564)

Pct 30 (Tax: 5.8) 1.27** 1.49*** 0.97* 1.22** 1.25** 1* 1.17** 1.58***(0.613) (0.559) (0.526) (0.513) (0.526) (0.523) (0.453) (0.561)

Pct 40 (Tax: 7.7) 1.17* 1.36** 0.87 1.12** 1.16** 0.9* 1.06** 1.44**(0.613) (0.556) (0.541) (0.521) (0.542) (0.519) (0.448) (0.561)

Pct 50 (Tax: 10.5) 1.03 1.17** 0.72 0.97* 1.04* 0.77 0.9** 1.24**(0.625) (0.562) (0.573) (0.542) (0.575) (0.526) (0.453) (0.57)

Pct 60 (Tax: 17.1) 0.7 0.73 0.38 0.64 0.76 0.46 0.53 0.78(0.705) (0.622) (0.687) (0.633) (0.694) (0.598) (0.519) (0.635)

Pct 70 (Tax: 20.8) 0.51 0.49 0.19 0.45 0.6 0.28 0.32 0.52(0.775) (0.679) (0.767) (0.702) (0.778) (0.664) (0.58) (0.693)

Pct 80 (Tax: 28.3) 0.14 0 -0.19 0.07 0.29 -0.07 -0.09 0.01(0.95) (0.827) (0.948) (0.866) (0.968) (0.83) (0.736) (0.839)

Pct 90 (Tax: 33.1) -0.11 -0.33 -0.44 -0.18 0.08 -0.3 -0.37 -0.34(1.082) (0.94) (1.077) (0.985) (1.103) (0.956) (0.853) (0.951)

Growth GDP pc

Unemployment rate

Ln GDP pc

Table 4. 2 (Continued). Robustness checks: Outliers. Reference regression: Table 4- Column 8. 5-year average data.Dependent variable is public deficit over GDP

Excluded country

Corruption

Tax decentralization

Population

Corruption x Tax decent.

Total government revenue

46

ISL ISR ITA JPN KOR LUX NLD NOR

(18) (19) (20) (21) (22) (23) (24) (25)1.484*** 1.526*** 1.4*** 1.619*** 1.386*** 1.578** 1.709*** 1.293**(0.51) (0.53) (0.48) (0.54) (0.53) (0.64) (0.56) (0.51)-0.016 0.001 0.014 0.025 0.007 0.008 0.024 0.001(0.05) (0.05) (0.04) (0.04) (0.04) (0.05) (0.04) (0.04)-0.034 -0.05 -0.055* -0.071*** -0.03 -0.047 -0.058* -0.055**(0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03) (0.03)0.107 0.051 -0.02 0.044 -0.062 0.041 0.026 -0.038(0.15) (0.14) (0.11) (0.13) (0.14) (0.17) (0.14) (0.11)

-0.637** -0.662** -0.696** -0.568** -0.506 -0.656** -0.703** -0.768***(0.3) (0.29) (0.27) (0.26) (0.35) (0.28) (0.27) (0.25)

0.263** 0.317** 0.407*** 0.302** 0.214* 0.328** 0.337*** 0.273**(0.13) (0.13) (0.16) (0.14) (0.12) (0.14) (0.13) (0.11)-1.215 -1.157 -0.831 -1.305* -1.233* -1.165 -1.026 -0.681(0.77) (0.78) (0.76) (0.7) (0.74) (1.05) (0.74) (0.67)0.022** 0.02** 0.015** 0.02*** 0.011 0.018* 0.017** 0.013**(0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.01)

Arellano-Bond test for AR(2) 0.787 0.717 0.588 0.509 0.992 0.707 0.763 0.861Hansen test of overid. rest 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Number of obs 138 139 137 137 137 137 137 137

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2.1) 1.41*** 1.42*** 1.28*** 1.47*** 1.32** 1.48** 1.59*** 1.18**(0.504) (0.523) (0.47) (0.538) (0.518) (0.627) (0.545) (0.48)

Pct 20 (Tax: 4.8) 1.32** 1.29** 1.13** 1.28** 1.24** 1.35** 1.43*** 1.03**(0.51) (0.527) (0.465) (0.544) (0.514) (0.62) (0.543) (0.456)

Pct 30 (Tax: 5.8) 1.29** 1.24** 1.08** 1.21** 1.21** 1.31** 1.38** 0.98**(0.515) (0.532) (0.466) (0.548) (0.516) (0.62) (0.545) (0.45)

Pct 40 (Tax: 7.7) 1.22** 1.14** 0.98** 1.07* 1.15** 1.21* 1.26** 0.87**(0.532) (0.547) (0.475) (0.559) (0.524) (0.625) (0.554) (0.442)

Pct 50 (Tax: 10.5) 1.12** 1* 0.82 0.87 1.07* 1.08* 1.1* 0.71(0.569) (0.581) (0.5) (0.583) (0.547) (0.644) (0.579) (0.442)

Pct 60 (Tax: 17.1) 0.9 0.67 0.46 0.4 0.87 0.77 0.72 0.35(0.695) (0.7) (0.605) (0.667) (0.642) (0.731) (0.679) (0.489)

Pct 70 (Tax: 20.8) 0.77 0.49 0.25 0.14 0.75 0.59 0.5 0.15(0.783) (0.785) (0.683) (0.727) (0.714) (0.8) (0.754) (0.539)

Pct 80 (Tax: 28.3) 0.51 0.12 -0.15 -0.39 0.53 0.24 0.07 -0.26(0.98) (0.976) (0.863) (0.867) (0.883) (0.967) (0.929) (0.674)

Pct 90 (Tax: 33.1) 0.35 -0.12 -0.42 -0.73 0.38 0.01 -0.21 -0.53(1.119) (1.112) (0.992) (0.968) (1.005) (1.091) (1.056) (0.778)

Table 4. 2 (Continued). Robustness checks: Outliers. Reference regression: Table 4- Column 8. 5-year average data.Dependent variable is public deficit over GDP

Excluded country

Corruption

Tax decentralization

Corruption x Tax decent.

Total government revenue

Growth GDP pc

Unemployment rate

Ln GDP pc

Population

47

NZL POL PRT SVK SVN SWE USA

(26) (27) (28) (29) (30) (31) (32) (33)1.26** 1.461*** 1.702*** 1.343*** 1.462*** 1.558*** 1.471*** 1.075**(0.53) (0.5) (0.54) (0.47) (0.54) (0.5) (0.48) (0.5)-0.014 0.003 0.016 -0.001 0.005 0.002 0.012 -0.002(0.04) (0.04) (0.05) (0.04) (0.04) (0.05) (0.04) (0.04)-0.046 -0.052* -0.05 -0.052 -0.051 -0.048 -0.06** -0.06**(0.03) (0.03) (0.03) (0.04) (0.03) (0.03) (0.03) (0.03)0.029 0.009 0.104 0.004 0.035 0.078 -0.034 -0.097(0.11) (0.13) (0.12) (0.12) (0.14) (0.11) (0.12) (0.1)

-0.754*** -0.775*** -0.601** -0.763*** -0.658** -0.61** -0.707*** -0.865***(0.27) (0.27) (0.29) (0.24) (0.29) (0.29) (0.26) (0.27)0.337** 0.394** 0.274* 0.342** 0.337** 0.305* 0.385** 0.355***(0.14) (0.15) (0.14) (0.16) (0.14) (0.16) (0.17) (0.12)-1.204* -1.119 -1.282* -1.358** -1.037 -1.183 -0.879 -0.635(0.72) (0.88) (0.73) (0.67) (0.8) (0.73) (0.69) (0.71)0.018** 0.018** 0.021** 0.018** 0.018** 0.02*** 0.024 0.013**(0.01) (0.01) (0.01) (0.01) (0.01) (0.01) (0.02) (0.01)

Arellano-Bond test for AR(2) 0.671 0.926 0.913 0.848 0.696 0.975 0.629 0.766Hansen test of overid. rest 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00Number of obs 137 138 137 138 139 137 137 137

Marginal effects of corruption for different percentiles of fiscal decentralization

Pct 10 (Tax: 2.1) 1.16** 1.35*** 1.6*** 1.23*** 1.36** 1.46*** 1.35*** 0.95**(0.528) (0.49) (0.525) (0.444) (0.53) (0.481) (0.464) (0.468)

Pct 20 (Tax: 4.8) 1.04* 1.21** 1.46*** 1.09** 1.22** 1.32*** 1.18*** 0.78*(0.534) (0.485) (0.521) (0.435) (0.53) (0.475) (0.452) (0.441)

Pct 30 (Tax: 5.8) 1* 1.16** 1.42*** 1.04** 1.17** 1.28*** 1.13** 0.73*(0.538) (0.486) (0.523) (0.437) (0.533) (0.476) (0.451) (0.434)

Pct 40 (Tax: 7.7) 0.9 1.06** 1.32** 0.94** 1.07* 1.18** 1.01** 0.61(0.553) (0.495) (0.532) (0.449) (0.545) (0.484) (0.454) (0.425)

Pct 50 (Tax: 10.5) 0.78 0.91* 1.18** 0.79 0.92 1.05** 0.84* 0.44(0.583) (0.518) (0.556) (0.485) (0.573) (0.507) (0.468) (0.424)

Pct 60 (Tax: 17.1) 0.47 0.57 0.85 0.45 0.59 0.73 0.44 0.04(0.69) (0.617) (0.657) (0.626) (0.679) (0.607) (0.544) (0.472)

Pct 70 (Tax: 20.8) 0.3 0.38 0.67 0.25 0.4 0.55 0.22 -0.18(0.766) (0.692) (0.732) (0.727) (0.757) (0.683) (0.607) (0.526)

Pct 80 (Tax: 28.3) -0.04 0 0.3 -0.14 0.02 0.19 -0.23 -0.62(0.938) (0.865) (0.907) (0.953) (0.936) (0.859) (0.758) (0.666)

Pct 90 (Tax: 33.1) -0.27 -0.26 0.06 -0.39 -0.23 -0.04 -0.52 -0.92(1.062) (0.989) (1.034) (1.112) (1.065) (0.986) (0.868) (0.774)

Table 4. 2 (Continued). Robustness checks: Outliers. Reference regression: Table 4- Column 8. 5-year average data.Dependent variable is public deficit over GDP

Excluded country Observations excluded if stand.

resid. >|2|

Corruption

Ln GDP pc

Population

Tax decentralization

Corruption x Tax decent.

Total government revenue

Growth GDP pc

Unemployment rate

* The country abbreviations correspond to the 3-letter country code ISO-3166-1 alpha 3.