Why Big Government is Good Government
How Direct Taxation Shapes Citizen Demand for Political Accountability
Anna Persson and Bo Rothstein
The Quality of Government Institute
Department of Political Science
University of Gothenburg
FOR BETTER READING DOWNLOAD THE PDF TO THIS PAPER FROM:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1902612
[email protected], [email protected]
Abstract
This paper asks how we can understand the fact that – quite contrary to what
dominant economic theories of corruption predict – bigger governments generally
tend to be less corrupt than smaller ones.
The results, based on an interview study
conducted in Uganda, reveal that the unexpected, comparatively high levels of
corruption in countries with relatively small governments can at least partly be
understood as a consequence of that the size of government does not only affect the
opportunities and incentives for so-called agents to engage in corrupt activities, but it
also affects the incentives for so-called principals to actually hold corrupt officials
accountable.
In particular, the study reveals that the level of direct taxation – which is
a key indicator of the size of government, smaller governments generally having
lower levels of direct taxation than larger governments – seems to play a decisive
role in shaping the incentives for ordinary citizens to hold corrupt officials
accountable, low levels of direct taxation leading to a significantly decreased sense
of “ownership” of the state, and consequently a decreased demand for “good
government” on behalf of citizens.
Since, in the end, the existence of “principals”,
willing to hold public officials accountable, is fundamental for successfully
constraining corruption, the lack of such principals in states with smaller
governments can to a significant extent explain why bigger governments are often
also better governments.
Paper prepared for presentation at the 2011 Annual Meeting of the American Political Science
Association, September 1-4, Seattle.
Preliminary version, please handle with care.
2
Introduction
Economic theories of corruption predict “big government” to be “bad government” (Rose-
Ackerman 1978; Klitgaard 1998; Becker 1995, 1997; Shleifer and Vishny 1998; Gael and
Nelson 1998; Scully 1991; Myrdal 1968; Tanzi 2000; Alesina and Angeletos 2005). This view
follows the logic of the well-known principal-agent theory, which is today the dominant
analytical framework for understanding the drivers of corruption and other forms of low quality
of government (henceforth QoG). Within this framework, corrupt acts are conceived of as the
result of an information and interest asymmetry between an agent (either in the form of a
bureaucrat or a ruler) – assumed to act in his or her own self-interest – and a principal (either in
the form of a ruler or citizens), typically assumed to embody the public interest and hence being
“a highly principled principal” (Klitgaard 1988).
In the end, if we follow the logic of this
perspective, we should expect the opportunities and incentives of agents in general – and of
rulers in particular – to engage in corrupt behavior to be highly correlated with the size of
government in line with Robert Klitgaard’s (1998) conviction that “discretion plus monopoly
minus accountability equals corruption” (Shah 2007; Lawson 2009). More specifically, in line
with this logic, big government should be expected to result in bad government through mainly
two different mechanisms. To begin with, a larger government is likely to increase the monopoly
and discretion agents have over resources to sell in exchange for bribes.
Moreover, a larger
government is likely to increase the size of rents (Kreuger 1974; Klitgaard 1988; de Soto 1989;
Shleifer and Vishny 1998; Gael and Nelson 1998; Rose-Ackerman 1978; Andvig and Fjeldstad
2001). This economic wisdom has been encapsulated by, among others, Gary Becker – the 1992
Nobel laureate in economics – in his Business Week columns under titles such as “To Root out
Corruption, Boot Out Big Government” (1994). For Becker (1995), as well as for many other
3
economists, “the source of corruption is the same everywhere; large governments with the power
to dispense many goodies to different groups”. Therefore, smaller government is “the only
surefire way to reduce corruption” (Becker 1995). A similar view is also put forward by Alberto
Alesina and George-Marios Angeletos (2005, 1234) who conclude their deductive analysis by
stating that “a large government increases corruption and rent-seeking”. In line with this
conviction, contemporary anti-corruption efforts have commonly involved the radical downsizing
of governments in countries ridden with widespread corruption, policies on corruption
thereby being deeply embedded within the wider construction of global agenda of neoliberalism
(Ivanov 2007; Szeftel 2000; Marquette 2003; Brown and Cloke 2004, 2005).
Yet, despite the – at least – theoretical appeal and coherence of the argument holding that
big government should be expected to be bad government, reality seems to tell a very different
story. That is, if we take a deeper look into the empirics of the real world, the relationship
between government size and corruption seems to run in quite the opposite direction than the one
predicted by neo-classical economic theory.
Thus, the comparatively least corrupt countries – to
a significant extent situated in the northern parts of Europe – have generally much larger
governments than the most corrupt ones (generally situated in the developing world, and
particularly in sub-Saharan Africa). In fact, the ten least corrupt countries in the world according
to Transparency International’s ranking (i.e. Finland, Denmark, New Zealand, Iceland
Singapore, Sweden, Canada, Luxembourg, Netherlands, and the United Kingdom) on average
collect 23.64 percent of their GDP in taxes, while the ten most corrupt countries for which there
is data available collect only 14.5 of their GDP in taxes (World Bank 2010). Similarly, central
government expenditure as a percentage of GDP is twice as high in the ten least corrupt countries
compared to in the ten most corrupt countries for which there is data (20.75 percent of GDP
4
compared to 10.45). Even if we take the whole world as a sample we find a positive relationship
between government size and institutional quality, the correlation between total tax revenues as a
share of GDP and institutional quality being 0.34, and the correlation between total government
consumption as a share of GDP and institutional quality being 0.39.
This relationship holds true
also if we compare countries within regions with comparatively very high levels of corruption –
such as the sub-Saharan African ones. For example, in neither of two of the least corrupt
countries in sub-Saharan Africa – Botswana and Mauritius – there is any reason to believe that
the accomplishments are due to adherence to an orthodox neoliberal creed of state minimalism.
Botswana has one of the largest public sectors in Africa with central government expenditure
being about 20 percent of GDP.
In addition, there has been substantial government intervention
in the economy and detailed planning. Similarly, Mauritius’ three main parties are socialist in
ideological orientation and the country maintains a significant welfare state (Goldsmith 2004). In
sum, if we consider existing empirical data, there seems to be very limited support in favor of the
argument that “big government” is generally “bad government” such as standard neo-classical
economic theory predicts. Instead, empirical data seems to indicate that, insofar as good
government is the objective, policy makers should aim for bigger rather than smaller
government. This finding is also supported by the 1992 Nobel Laureate in economics –Douglass
North – and his colleagues, who in a recent book share the view that there has to be something
profoundly wrong with the argument that high public spending is bad for economic growth and
social well-being (North, Wallis and Weingast 2009).
This paper asks how we can understand the fact that – quite contrary to what powerful
economic theory predicts – bigger governments generally tend to be less corrupt than smaller
ones. The results, based on an interview study conducted in Uganda, reveal that the unexpected,
5
comparatively high levels of corruption in countries with relatively small governments can at
least partly be understood as a consequence of that the size of government does not only affect
the opportunities and incentives for so-called agents to engage in corrupt activities, but it also
affects the incentives for so-called principals to actually hold corrupt officials accountable. In
particular, the study reveals that the level of direct taxation seems to play a decisive role in
shaping the incentives for ordinary citizens to hold corrupt officials accountable, low levels of
direct taxation leading to a significantly decreased sense of “ownership” of the state, and
consequently a decreased willingness on behalf of citizens to report and punish corrupt behavior.
Since, in the end, any anti-corruption framework based on the principal-agent model relies on the
existence of actors willing to act like principals (Klitgaard 1988; Mookherje 1997; Galtung and
Pope 1999; Andvig and Fjeldstad 2001), this lack of willingness of citizens in countries with
small governments to control public officials has serious implications for the level of corruption.
We expect this to be the case both if citizens act as individuals – i.e. in the sense that they vote
for parties that have “clean government” on their agenda or force corrupt politicians to step down
(Bågenholm 2009, Weghorst and Lindberg 2010) – or if they act as a collective in organized
forms.
In the end, the results presented in the paper by and large support the tenets of fiscal
sociology, according to which we should in fact expect especially direct taxation to constitute a
significant and important part of the social contract between citizens and the state (Moore 2004;
Lieberman 2003; Persson 2008). More specifically, according to this framework, the level of
corruption should be expected to increase rather than decrease with the down-sizing of
government, precisely as we have empirically witnessed. However, by demonstrating how the
level of direct taxation affect citizens’ demand for good government, this paper still contributes
6
with new important insights to the fiscal sociological perspective, since previous studies
adopting this view have almost exclusively focused on how the incentives of leaders’ to supply
high quality institutions are shaped by the degree to which they depend on direct taxation.
In the remainder of this article, we explore the argument further.
As such, in the next
section we outline the standard economic view of government size and corruption, which is also
reflected in contemporary efforts to curb corruption. Next, on the basis of in-depth interviews
with more than 30 Ugandan respondents, ranging from high-level public officials (including
officials working in anti-corruption agencies) to local NGO representatives and journalists, we
demonstrate that one potential explanation to why small government is often associated with
high levels of corruption (rather than low such as economic theory predicts) is that the low
degree of direct taxation characterizing the majority of countries with small governments seems
to reduce the likelihood that citizens will be motivated to engage in politics through a right to
influence the use of “their” own money.1
Hence, citizens living in countries with low levels of
direct taxation are not likely to hold corrupt public officials accountable such as economic theory
– as well as the international community involved in the fight against corruption – expect them to
do. We conclude the paper by summarizing our argument and spelling out some potential policy
implications.
1 The field work was conducted by Maria Jacobsson in April – May 2007. The transcripts of the interviews cover more than 200
single-spaced pages. The interviewees were selected as informants who share a professional commitment to working on
corruption and anti-corruption issues. In order not to steer the respondents in any specific direction, and hence better be able to
capture the “true” character of corruption, we let the respondents describe the problem of corruption in their respective country
contexts without reservation or guidance. The research projects of which this paper is part, are titled Anti-Corruption Strategies
in Africa. What Works and Why (financed by the research council at the Swedish International Development Agency) and All for
One or One for All? The Determinants of Institutional Quality (financed by the Swedish Research Council).
7
An economic perspective on government size and corruption
Most contemporary work on corruption takes its point of departure in an understanding of
corruption as “the misuse of public office for private gain” (Kaufmann 1997; Rose-Ackerman
1999; Lambsdorff 2007). As such, it rests on the logic of so-called principal-agent theory. Within
this framework, corrupt acts are conceived of as the result of an information and interest
asymmetry between an agent – assumed to act in his or her own self-interest – and a principal,
typically assumed to embody the public interest and hence being “a highly principled principal”
(Klitgaard 1988). In the end, as implied by the standard definition of corruption, corruption
hence occurs when the agent betrays the principal’s interest in the pursuit of his or her own selfinterest.
This betrayal is in turn made possible by the information asymmetry between the two
groups of actors.
Depending on perspective, who is the agent and who is the principal in the principalagent
model may differ. In the classical treatment, rulers are modelled as the principal and the
bureaucracy as the agent (Becker and Stigler 1974; and Van Rijckeghem and Weder 2001). The
classical model is hence based on the idea of the state as a “helping hand”, in the sense that it is
expected to serve in the interest of the public good (Shleifer and Vishny 1998). However, the
majority of scholars, as well as policy makers, now agree that there is considerable reason to
doubt whether real-life rulers really should be expected to serve as “helping hands” such as the
classical model assumes (Aidt 2003; Bayart et al. 1999; Chabal and Daloz 1999; Teorell 2007;
Persson, Rothstein and Teorell 2010; Andvig and Fjeldstad 2001; Bardhan 1997; Shleifer and
Vishny 1998; and Williams 1999). Consequently, in the less classical principal-agent model
rulers are modelled as agents and citizens as principals (Persson and Tabellini 2000; Adserà et al.
2003; Besley 2006; and Myerson 1993). As such, the less classical model takes its point of
8
departure in the idea of the state as a “grabbing hand”. At the root of this perspective is the
conviction that politicians should not be expected to maximize social welfare by voluntarily
providing good government to the people or punishing corrupt street-level bureaucrats but
instead pursue their own selfish objectives. That is, corruption is conceived of as a government
pathology, resulting from government officials’ intentionally creating regulations and policies
that entrepreneurs will have to pay bribes to get around (Shleifer and Vishny 1998).
In short,
corruption is a mechanism whereby power is converted into income for politicians (De Soto
1989; and Shleifer and Vishny 1998). In the end, important to note is that this less classical
model resembles the classical principal-agent framework in every respect with the exception
that, instead of assuming benevolent principals on the top, they take another attribute for given –
namely the presence of principals in the form of citizens. In other words, in line with this
perspective, if citizens only have the proper means and information, they will do everything in
their power to discourage malfeasance on behalf of the agents, i.e. the ruling elites (Myerson
1993; Persson and Tabellini 2000; Besley 2006).
Regardless of how the principal-agent relationship is modelled, the policy implications
following from this framework hold that, in order to reduce corruption, both the supply of and
the demand for corrupt activities should be restricted. The principal should, in other words, aim
at negatively affecting the agent’s opportunities to engage in corrupt behavior, as well as the
incentives for actually being corrupt when the opportunity occurs (Klitgaard 1988; Rose-
Ackerman 1978; Teorell 2007; and Andvig and Fjeldstad 2001). This could, according to the
majority of scholars, most effectively be done through control instruments that limit the
monopoly of agents; decrease the level of discretion among agents, and; increase the level of
accountability in the system in line with Robert Klitgaard’s (1988, 75) conviction that
9
“corruption equals monopoly plus discretion minus accountability”. The monopoly and
discretion of agents in general – and of rulers in particular – are, in turn, expected to be
intimately linked with the size of government. More specifically, a larger government is
expected to generate corrupt government mainly through two different mechanisms. First, a
larger government is expected to increase the monopoly and discretion agents have over
resources to sell in exchange for bribes.
That is, it increases the opportunities for corruption.
Second, a larger government is expected to increase the size of rents. That is, it is expected to
increase the incentives for corruption by generating more potential rents for agents in the form of
bribes in return for, for example, issuing permits and licenses, for providing passage through
customs, and for prohibiting the entry of competitors (Kreuger 1974; Klitgaard 1988; de Soto
1989; Shleifer and Vishny 1998; Gael and Nelson 1998; Rose-Ackerman 1978; Andvig and
Fjeldstad 2001). In the end, especially the grabbing hand model of corruption immediately
implies that deregulation and liberalization are extremely important tools for fighting corruption
since, within the framework of this model, no public officials can be entrusted to act like
“principals” (Shleifer and Vishny 1998; and Krueger 1974).
In line with this view, the scaling down of governments has been a top priority of the
contemporary anti-corruption agenda (Ivanov 2007; Fitzsimons 2007, 2009). For instance,
restrictions on public sector employment and on state spending, including elimination of
government subsidies, have been introduced in order to reduce the public resources political
leaders can offer to acquire support. Moreover, privatization has been promoted to reduce the
number of state-owned enterprises and also remove from the government the ability to provide
goods and services to favored constituencies (Mwenda and Tangri 2005). Thus, the
contemporary anti-corruption agenda in many respects seems to share the conviction held by
10
economic theory in general – and the principal-agent framework of corruption in particular – that
smaller governments with limited power to dispense “goodies” to different groups should be
expected to result in better governance (Becker 1995). Yet, despite the theoretical appeal of this
argument, empirical data seems to point in a very different direction.
That is, instead of curbing
corruption, it seems like the scaling down of governments has – in the best-case scenarios – left
already thoroughly corrupt governments about as corrupt as they were before the reforms were
initiated. In the worst-case scenarios, on the other hand, the down-sizing of governments even
seems to have increased the problem of corruption (Riley 1998; Tangri and Mwenda 2001,
2006). In other words, the time-series pattern seems to reveal a very similar picture as the crosssection
pattern discussed in the introductory part of this article, i.e. a picture of a real-life context
in which big government seems to constrain corrupt behavior on behalf of public officials rather
than facilitate it such as predicted by economic theory. In the next section, we provide a potential
explanation to why this is the case.
Why big government is good government
How can we understand the comparative advantage of big government over small government
when it comes to producing high-quality institutions? In this section, we provide a potential
explanation to why this is the case that acknowledges the importance of the other key
explanatory factor to variation in corruption levels put forward by Klitgaard and others, namely
accountability.
More specifically, on the basis of an interview study conducted in Uganda – a
thoroughly corrupt country with a comparatively very small government even for African
standards – we argue that one potential explanation to why bigger governments are generally
better governments is that citizens in countries with bigger governments are much more likely to
hold corrupt officials accountable. This, in turn, seems to be the result of that citizens in
11
countries with larger governments generally pay a larger amount of their income in taxes and,
hence, have greater incentives to hold corrupt officials accountable through a perceived right to
influence the use of their “own” money.2
In the remainder of this section, we develop this argument further. As such, in the next
sub-section, we explore the – for the argument – key concept of accountability. What does
accountability mean, and how is it related to corruption?
In a case study of Uganda, we then
show that the variation in the level of accountability and, hence, in the level of corruption
between countries with smaller and bigger governments seems to be closely linked to variation in
the level of direct taxation. More specifically the study reveals that, in countries in which citizens
do not pay any significant share of their income in taxes, they do not either expect the
government to deliver much in terms of good governance (or other public goods for that sake) in
the first place. Consequently, they do not demand accountability from public officials such as
economic theories of corruption expect them to do.
Accountability and corruption
Accountability in politics, described as early as 1816 by Jeremy Bentham (1999), refers to the
ability of citizens to use the electoral mechanism to shape the incentives facing politicians (see
also Barro 1973; Ferejohn 1986). More specifically, accountability refers to the degree to which
principals can control agents in a principal-agent relationship (Lassen 2001; Rakner and Gloppen
2002). As such, in essence the concept carries two basic connotations: 1) answerability, i.e. the
obligation of public officials (either in the form of bureaucrats or rulers) to inform about and to
explain what they are doing, and; 2) enforcement, i.e. the capacity of accounting agencies (either
2 The correlation between both direct taxes (in the form of taxes on income, profits and capital gains) and total taxes and direct
taxes and government consumption is 0.99, indicating that direct taxation is indeed a key characteristic of big government (World
Bank 2010).
12
in the form of rulers or citizens) to impose sanctions on powerholders who have violated their
public duties (Schedler 1999). As argued by Schedler (ibid, 14), this two-dimensional structure
of meaning makes the concept a broad and inclusive one that, within its boundaries embraces at
least three different ways of preventing and redressing the abuse of power; it implies subjecting
power to the threat of sanctions, it obliges power to be exercised in transparent ways; and it
enforces powerholders to justify their acts. In sum, the standard definition of accountability
hence implies that “A is accountable to B when A is obliged to inform B about A’s (past or
future) actions and decisions, to justify them, and to suffer punishment in the case of eventual
misconduct” (Ibid, 17).
There is now a general consensus holding that the successful implementation of any anticorruption
effort based on the principal-agent framework ultimately rests on the existence of
actors willing to act as “principals” by holding corrupt officials accountable (Klitgaard 1988;
Mookherje 1997; Rauch and Evans 2000; Andvig and Fjeldstad 2001; Galtung and Pope 1999;
and Mungiu-Pippidi 2006). This conviction is also mirrored in Klitgaard’s well-renowned
equation holding that “monopoly plus discretion minus accountability equals corruption”. In
particular, according to a large number of scholars, how well any government functions
ultimately hinges especially on how good citizens are at holding their politicians accountable for
their actions (Adserà, Boix, and Payne 2003). In fact, according to this literature, political
accountability explains as much as up to two-thirds of the variance in the levels of corruption
(ibid). Hence, if the aim is to understand why bigger government is generally better government,
we should explore more in detail why the citizens in countries with larger governments to a
greater extent are willing to hold corrupt officials accountable than the citizens in countries with
smaller governments. In the section, we explore this issue further.
13
Why demand when you did not pay? Small government as a source of bad government
Acknowledging the great importance of political accountability for constraining corruption, the
international community has during the last two decades – along with policies of deregulation
and liberalization – introduced a variety of formal mechanisms and institutions aimed at
increasing the ability also of citizens in the developing parts of the world to hold public officials
accountable, including monitoring mechanisms, increased transparency, democratic election
mechanisms, a free press, as well as the strengthening of civil society (World Bank 2000;
Stapenhurst and Kpundeh 1999; and Van Rijckeghem and Weder 2001). However, as being
evident from the great discrepancy in the level of corruption between countries with larger
governments and countries with smaller governments, this formal institutional framework has
obviously not been as effective in all countries. Uganda, which is the case in focus of this subsection,
is one of the countries in which the introduction of a large variety of mechanisms of
accountability has not led to any real change in corruption levels. In fact, Uganda can in most
respects be argued to be the stereotype of a country with a comparatively small government.
Despite the introduction of a number of instruments and institutions for regulating and
monitoring the ethical standards of public officials from “below”, the management of public
affairs and institutions by those who are entrusted with positions of authority in the country has
not improved at all. According to Global Integrity (2008), the gap between the institutional
framework and its actual implementation is “huge”. In fact, since Transparency International
introduced its “Corruption Perception Index” (CPI) in 1998, the level of corruption in Uganda
has been nothing else than rampant. While in 1998 Uganda had a CPI of 2.6, today it has an
index of 2.5, placing it as one of the 30 most corrupt countries in the world. As existing theories
about the relationship between corruption and the degree to which citizens hold corrupt officials
14
accountable should make us expect, this comparatively very high level of corruption is to a
significant extent mirrored in the behavior of citizens. Similar to the situation in most other
African countries, the reporting and conviction of corrupt cases in Uganda remain low (Doig and
Riley 1998; Riley 1998). Moreover, vote-buying – i.e. when citizens sell their votes in exchange
for different goods – is extremely widespread, and corrupt politicians often stand good chances
for re-election (Mwenda and Tangri 2005). In fact, until today, not a single top leader in Uganda
has been prosecuted in a corruption case (Tangri and Mwenda 2006).
While this lack of political accountability in Uganda can to some extent certainly be
attributed even to other factors, the informants interviewed for the purpose of this study to a
significant extent share a similar understanding of the problem, arguing that the unwillingness of
ordinary citizens to hold corrupt officials accountable largely stems from the fact that they do not
pay any direct taxes. In the absence of direct taxation, the informants argue, people do not feel
any real sense of “ownership” of the state. Hence, the likelihood that citizens will be motivated
to demand good governance – or, for that sake, at all engage in politics – is severely reduced.
Why care about whether public officials are corrupt or not if you are not the one paying the bill?
Or, as argued by the informants, the limited “connection” to the money make people indifferent
to the problem of corruption:
They don’t feel connected. They don’t feel connected because there is limited taxation here. [If they see
corruption,] they will tell you ‘that money is government money, it is not ours’.
…the weak state, the lack of taxation, and the donor influence make people feel no direct connection to
the money. And since most of the money is coming from the donors people think ‘that is donor money
and therefore not ours’. So they don’t feel that attachment. It would be different if people felt that they
were being taxed.
…the graduated tax has been abolished so people don’t [see the government as their property]. The other
taxes you pay are indirect, like on commodities. You don’t feel it. So you don’t think you are
contributing. So it is making citizens a little divorced from the government.
People here do not see it as the property that belongs to the government actually belongs to them. They
think that government and themselves are different, but [the government] is theirs. It is their property; it is
their wealth as a country. But people have distanced themselves from government. So when they see
15
someone abusing a government thing they don’t mind. They say ‘after all it is government’, not knowing
it is themselves who are going to suffer… They don’t see the connection. For them the government is a
mystery, something that is not part of them.
The problem of detachment from the state – and the resulting indifference to corruption – is,
according to the informants, often even worse in the rural areas where the majority (87 percent)
of the Ugandan population live, since very few people in these areas are at all taking part in the
formal economy.
People are passive and I don’t think it is because they have taken bribes. I think it is because, especially in
the rural areas people are detached from government... These are people that are not in the mainstream.
They don’t even relate to government in a formal way; they don’t pay taxes so they are kind of delinked.
So when they are passive and don’t care what the leaders do it is not because they have taken bribes. I
think it is largely because they do not feel that they have a relationship with the government. So I think
when people are sensitised it might be wrong to assume that they are passive because they are taking
bribes and that they therefore cannot make demands on their leaders. I think that people are passive
because they don’t feel part of the government process. So people just need to be mobilised and feel that
although they don’t pay taxes still this money spent is there money.
Another informant agrees that a sense of “ownership” of the state is likely to be missing to a
greater extent in the rural areas:
…you have a situation whereby the state is supported by donor funds and indirect taxes so somebody in a
village does not really care what the government does because they will say ‘it is not my money; I did not
pay tax’. Now recently when graduated tax was abolished, people have even less stake in government.
People do not realise that there are indirect taxes. When you drink a bottle of mineral water or take a taxi,
people don’t see that they are paying taxes.
In the end, according to the informants, the reason why citizens in Uganda do not hold corrupt
officials accountable is that, since they did not pay any taxes in exchange for public goods in the
first place, they do not expect the government to deliver anything. In other words, what Margaret
Levi (1988) has called the “quid pro quo of taxation” simply seems to be lacking. One of the
informants describes the lack of a reciprocal relationship between citizens and the state, within
the framework of which taxes are paid in exchange for public goods:
[In the end, there] must be a form of responsibility that is created between what you pay with your taxes
and what you demand from government. If more people were paying taxes here and it was biting on them
they would demand something in return.
16
Another informant fills in, arguing that – quite contrary to what one would expect with the point
of departure in the principal-agent relationship described in the literature, within the framework
of which government officials are expected to dance to the tunes of the people in Uganda it in
fact seems to be the other way around because of the close to absence of direct taxation:
Here it is the party that says ‘we shall have universal secondary education and the people will embrace it.’
It should be the other way around. People should say ‘we are paying taxes and we want universal
secondary education’. Then parties must tune themselves to the dance, to the tunes of the people. I think
that is the solution.
In fact, according to many of the informants, the absence of direct taxation does not only lead to
a decreased demand for good governance, but to some extent it even strengthens already very
bad government. This is because, since people do not pay anything directly for the services
delivered by the government, there is a widespread perception that the government is in fact
doing citizens a favor when it delivers a service, even if this service happens to be paid by
indirect taxes such as, for example, value-added tax. In short, as a result of the absence of direct
taxation citizens do not only abstain from demanding good governance, but to some extent they
even pay tribute to the corrupt:
Whatever they receive from government, even when they have a right to receive it, they see it as a favour
and therefore they are all appreciative. They say ‘after all it is coming to me free; I have not paid for it so
why should I make noise’. That is the thinking of the people.
[Since] they have removed graduated tax, people no longer pay. So they feel disempowered. Whatever
they receive from government, even when they have a right to receive, they see it as a favour. And
therefore they are all appreciative.
People have taken it to their mind that they (i.e. government officials) are helping them, not knowing that
there are taxes and that they are entitled to these services. So that mentality is what we must change.
In sum, the interview study conducted in Uganda indicates that, at least to a certain
extent, the generally higher levels of corruption in countries with small governments can be
understood as a result of the lack of a social contract between citizens and public officials in such
states. More specifically, since citizens in countries with smaller states do not pay any significant
amount of their income in direct taxes, they will have limited incentives to hold corrupt officials
17
accountable through a perceived right to influence the use of their “own” money. In fact, in the
absence of direct taxation, citizens are – according to the informants – to some extent even likely
to show appreciation of corrupt officials, further reinforcing the problem.
Summary and conclusions
“Taxes are what we pay for a civilized society” (Oliver Wendel Holmes, to be read on the front of the
Ministry of Taxation in Washington D.C.)
There is now a general consensus holding that a high-quality government is a necessary
requirement for countries to foster economic development and social well-being more generally
(Holmberg, Rothstein and Nasiritousi 2008; North, Wallis and Weingast 2009; Acemoglu,
Johnson, and Robinson 2002, 2004). Taking this insight seriously, in the last one and a half
decades there has been an explosion in terms of literature aiming to find the causes of bad
government. The winning literature in this race in terms of its influence on the work of scholars
and policy makers has – at least so far – been standard economic theories of corruption, which
hold that big government is one of the main sources of bad government. In fact, so strong is this
conviction that empirical evidence in favor of the opposite hypothesis – i.e. the big government
in general in fact tends to be correlated to good government – has been systematically neglected.
The reasoning in one of the very first empirical articles about “the quality of government”,
published by the well-renowned economists Rafael La Porta, Florencio Lopez-de-Silanes, Andrei
Schleifer, and Robert Vishny (1999), clearly mirrors this general neglect of empirical facts
working against established economic theory. Using a wealth of data from between 60 and 120
countries, they came to what seems to have been for them a very surprising conclusion. In their
own words, they “consistently found that the better performing governments are also larger, and
collect higher taxes. Poorly performing governments, in contrast, are smaller and collect fewer
taxes" (Ibid, 42). That this was a surprise for them is indicated by the fact that they immediately
18
added that their finding should certainly not be used as an argument for increasing the size of
government in a given state.
Theories about dominating paradigms in research could potentially help us understand
this general neglect in the literature of empirical evidence in favor of the argument that big
government in many ways actually seems to be better than suggested by established economic
theory. The theory about dominating paradigms in research says that when a theory is at its
height, empirical anomalies are simply disregarded since scholars have no way of handling this.
However, our empirical results reveal that there is potentially another theoretical framework that
can better account for the fact that big government is generally better than small government,
making La Porta et al.’s finding less of an anomaly. More specifically, revealing the ways in
which direct taxation seems to be the key to shaping the incentives of citizens to demand
accountability from public officials, this paper adds important insights to the theory of fiscal
sociology.
The starting point of the fiscal sociological framework is the conviction that taxation
forms the “sinews” of the state and is, as such, in many respects the prerequisite for any
governance as such (Schumpeter 1991 [1918]; Levi 1988; North 1981; Pierson 1996). More
specifically, states that are able to regularly collect taxes from a wide range of societal actors
should, according to this framework, generally be able to govern effectively also in a wide range
of other areas, while the inability of a state to generate significant revenue through taxation
should be expected to be a precursor to state failure, or even collapse (Alesina and La Ferrara
2000; Bergman 2003; Lieberman 2003; Tilly 1990; Persson 2008). In particular, according to the
fiscal sociological perspective, we should expect direct taxation to play a crucial role in the
process of state development. In particular, scholars adopting a fiscal sociological have
19
demonstrated that rulers that depend on direct taxation for their survival are much more likely to
be willing to supply good governance than rulers that do not rely on direct taxation for their
survival. This is because, from the standpoint of fiscal sociology, taxation is interpreted in terms
of a collective action problem, which is ultimately solved by the establishment of a social
contract between the rulers and the ruled which implies that, in order to be able to generate
significant amounts of tax revenue, the ruling elite must provide a contract or bargain that is
considered fair (Levi 1988). More specifically, from this perspective, in order to be able to
collect taxes, rulers must offer something that the citizens demand – such as, for example,
representation, accountability, or services – in return and in an exchange based on reciprocity
(Bräutigam 2008). Consequently, rulers that depend on taxation for their survival tend to have
much larger incentives to establish high-quality institutions compared to rulers that do not rely
on taxation.
In the end, while the results presented in this paper by and large support the theory of
fiscal sociology, the study still adds important insights to this literature by revealing how
government size in general – and the level of direct taxation in particular – not only affects the
supply of institutional quality, but also the demand. More specifically, the study reveals that, in
the absence of direct taxation, citizens are not very likely to demand good government.
Consequently, both Klitgaard’s now well-known equation and contemporary anti-corruption
efforts seem to stumble on their own logic. In particular, quite contrary to what is generally
assumed, our study reveals that the factors on the left-hand side of Klitgaard’s equation holding
that “monopoly plus discretion minus accountability equals corruption” are not independent of
each other, but rather the level of accountability in a political system should be expected to at
least partly depend on government size.
20
However, while the results of the study point in a rather clear direction, the policy
implications are far from as straightforward. Since taxation in many ways resembles a collective
action problem (Levi 1988; Lieberman 2003; Persson 2008), to increase the level of taxation in a
particular country is far from an easy task. Moreover, one must also be aware of the fact that
taxation can potentially have very different effects depending on how taxes are collected. Insofar
as they are collected coercively by predatory governments as a means to secure control rather
than with the ambition to act in the interest of the public good, there is little reason to expect
taxation to play a positive role in the process of state development (Moore 2007). In fact, in the
hands of the wrong government, taxation clearly runs the risk of turning into an important tool of
repression rather than a tool of accountability. Yet, as this study reveals, to the extent that direct
taxation is connected to a sense of “ownership” of the state, it can obviously play an important
and positive role in the state-building process. On the basis of this argument, scholars adhering to
the fiscal sociological perspective have suggested that one effective way to increase the degree to
which rulers are forced to be accountable to their citizens would be to decrease the level of
foreign aid going into the state budget of poor and corrupt countries (Moore 1998, 2001; Knack
2000; Djankov, Moltalvo, and Reynal-Querol 2008). Inspired by rentier state theories, the basic
argument here is that when states gain substantial incomes from external sources such as oil,
minerals, or foreign aid, they do not need to levy domestic taxes and therefore become less
accountable to the societies they govern. In fact, a quick look at the empirical data in many ways
seems to support such an argument. According to a study conducted by the International
Monetary Fund (IMF) in 1995, seventy-one percent of the African countries receiving more than
10% of GDP in aid were also in the group of countries judged by the IMF to have a lower than
expected tax effort. Moreover, in a more recent study, Deborah Bräutigam and Stephen Knack
21
(2004) present time-series evidence that at least in Africa, higher aid levels are generally
associated both with larger declines in the quality of governance and in tax revenues as a share of
GDP.
While we do not question the statement that withdrawal of foreign aid could potentially
be an important tool for increasing the incentives of rulers not to abuse public office for private
gain, the results of this study do still not entirely support such a proposition. Rather, the results of
the study support the argument put forward by Bräutigam and Knack (2004), holding that foreign
aid could potentially be of great help in building an accountable state apparatus insofar as it is
delivered more selectively and in ways that reinforce a virtuous cycle of development rather than
contribute to a vicious cycle of poor governance and economic decline. More specifically, our
results suggest that if foreign aid was somehow channelled through society, favorably as a kind
of “citizen salary” subject to taxation by the national government, this could potentially have
positive implications for accountability. In particular, while certainly aware of the potential
practical problems inherent in such an approach (for example, in many poor countries it is
difficult to even get in touch with the majority of the citizens since few of them are taking part in
the formal economic and political system), the thought here is that foreign aid channelled in this
way could – at least ideally – would work as a form of “substitute direct tax”, giving citizens a
greater sense of “ownership” of the state and, hence, making them more likely to hold corrupt
officials accountable.
References (incomplete)
Acemoglu, Johnson, and Robinson 2002
Acemoglu, Johnson, and Robinson 2004
22
Adserà, Alícia, Carles Boix, and Mark Payne. 2003. Are You Being Served? Political
Accountability and Quality of Government. The Journal of Law, Economics, &
Organization 19 (2): 445-490.
Aidt, Toke S.. 2003. Economic analysis of corruption: A survey. Economic Journal 113 (491):
F632-F652.
Alesina and Angeletos 2005
Alesina and La Ferrara 2000
Andvig, Jens Christopher, and Odd-Helge Fjeldstad. 2001. Corruption: A Review of
Contemporary Research. Report R 2001: 7. Bergen: Chr. Michelsen Institute.
Barro 1973
Bauhr et al. 2010
Bayart, Jean-François, Stephen Ellis, and Béatrice Hibou. 1999. The criminalization of the state
in Africa. Oxford: International African Institute in association with James Currey.
Bechert, Insa, and Markus Quandt. 2009. "ISSP Data Report: Attitudes towards the Role of
Government." Bonn: GESIS. Liebniz-Institute für Sozialwissenschaften. Working Paper
2009:2.
Becker 1994
Becker 1995,
Becker 1997
Becker, Gary S., and George J. Stigler. 1974. Law Enforcement, Malfeasance, and
Compensation of Enforcers. Journal of Legal Studies 3 (1): 1-18.
Besley, Timothy. 2006. Principled Agents: The Political Economy of Good Government.
Oxford: Oxford University Press.
23
Bentham [1816] 1999
Bergman 2003
Björklund 1989
Blanchflower, Jackman, and Saint-Pauls 1985
Brown, and Cloke 2004
Brown, and Cloke 2005
Bräutigam and Knack 2004
Bågenholm 2009
Chabal, Patrick, and Jean-Pascal Daloz. 1999. Africa Works: Disorder as Political Instrument.
London: International African Institute in association with James Currey, Indiana
University Press.
Charron, Nicholas, Lewis Dijkstra and Victor Lapuente. 2010. Mapping Quality of Government
in the European Union: A Study of National and Sub-National Variation. Working paper
2010:22, The Quality of Government Institute, University of Gothenburg.
Dahlström, Carl, Johannes Lindvall and Bo Rothstein. 2009. Social Protection Strategies in
Efficient and Inefficient States. Working Paper 2009:26. The Quality of Government
Institute, University of Gothenburg.
De Soto 1989
Djankov et al. 2008
Doig, Alan, and Stephen Riley. 1998. Corruption and Anti-Corruption Strategies: Issues and
Case Issues and Case Studies from Developing Countries. In Corruption and Integrity
Improvement Initiatives in Developing Countries, edited by in G. Shabbir Cheema, and
Jean Bonvin, 45-62. Paris: OECD.
24
Donahue, John D. 1989. The privatization decision : public ends, private means. New York:
Basic Books.
Ferejohn 1986
Fitzsimons 2007
Fitzsimons 2009
Gael and Nelson 1998
Galtung, Fredrik, and Jeremy Pope. 1999. The Global Coalition Against Corruption: Evaluating
Transparency International. In The Self-Restraining State: Power and Accountability in
New Democracies, edited by Andreas Schedler, Larry Diamond, and Marc F. Plattner,
257–282. Boulder, CO: Lynne Rienner Publishers.
Goldsmith 2004
Holmberg, Rothstein, and Nasiritousi 2008
IMF 1995
Ivanov, Kalin. 2007. The Limits of a Global Campaign against Corruption. In Corruption and
Development. The Anti-Corruption Campaigns, edited by Sarah Bracking, 28-45.
Basingstoke, Hampshire: Palgrave Macmillan.
Kaufmann 1997
Klitgaard, Robert. 1988. Controlling Corruption. Berkeley: University of California Press.
Knack 2000
Kreuger 1974
Lambsdorff 2007
La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert Vishny. 1999. "The
Quality of Government." Journal of Law, Economics and Organization 15:222-279.
25
Lassen 2001
Lawson, Letitia. 2009. The Politics of Anti-Corruption Reform in Africa. Journal of Modern
African Studies 47 (1): 73-100.
Lechner, Michael, Ruth Miquel, and Conny Wunsch. 2007. "The curse and blessing of training
the unemployed in a changing economy: The case of East Germany after unification."
German Economic Review 8:468-509.
Levi, Margaret. 1988. Of Rule and Revenue. Berkeley: University of California Press.
Lieberman 2003
Marquette 2003
Milner, Henry, and Eskil Wadensjö (Eds.). 2001. Gösta Rehn, the Swedish model and labour
market policies : International and national perspectives. Aldershot: Ashgate.
Mookherje 1997
Moore 1998
Moore 2001
Moore 2004
Moore 2007
Mucciaroni, Gary. 1990. The Political Failure of Unemployment Policy 1945-1982. Pittsburgh:
University of Pittsburgh Press.
Mungiu-Pippidi, Alina. 2006. Corruption: Diagnosis and Treatment. Journal of Democracy 17
(3): 86-99.
Mwenda, and Tangri 2005
26
Myerson, Roger B.. 1993. Effectiveness of Electoral Systems for Reducing Government
Corruption – A Game-Theoretic Analysis. Games and Economic Behavior 5 (1): 118-
132.
Myrdal, Gunnar. 1968. Asian Drama: An Inquiry into the Poverty of Nations. London: Allen
Lane the Penguin press.
North 1981
North, Douglass C., John Wallis, and Barry R. Weingast. 2009. Violence and Social Orders: A
Conceptual Framework for Interpreting Recorded Human History. Cambridge:
Cambridge University Press.
Persson 2008
Persson, Rothstein, and Teorell 2010
Persson, Torsten, and Guido Tabellini. 2000. Political Economics: Explaining Economic Policy.
Cambridge: The MIT Press.
Pierson 1996
Rakner, and Gloppen 2002
Rauch, and Evans 2000
Rehn 1985
Riley, Stephen P. (1998). ‘The Political Economy of Anti-Corruption Strategies in Africa’, The
European Journal of Development Research, 129-159.
Rose-Ackerman, Susan. 1978. Corruption: A Study in Political Economy. New York: Academic
Press.
Rose-Ackerman, Susan. 1999. Corruption and Government: Causes, Consequences, and Reform.
Cambridge: Cambridge University Press.
27
Rothstein, Bo. 1992. Den korporativa staten. Intresseorganisationer och statsförvaltning i svensk
politik. Stockholm: Norstedts.
Rothstein, Bo. 1996. The Social Democratic State: The Swedish Model and the Bureaucratic
Problem of Social Reforms. Pittsburgh: University of Pittsburgh Press.
Rothstein, Bo, and Jonas Bergström. 1999. Korporatismens fall och den svenska modellens kris.
Stockholm: SNS.
Schedler 1999
Schumpeter 1991 [1918]
Scully 1991
Shah, Anwar. 2007. Tailoring the Fight against Corruption to Country Circumstances. In
Performance Accountability and Combating Corruption, edited by Anwar Shah, 233–254.
Washington D.C.: World Bank.
Shleifer, Andrei, and Robert W. Vishny. 1998. The Grabbing Hand, Government Pathologies
and Their Cures. Cambridge: Harvard University Press.
Stapenhurst Rick, and Sahr J. Kpundeh, eds. 1999. Curbing Corruption: Toward a Model for
Building National Integrity. Washington: World Bank.
Stotsky, J. G. and A.Wolde Mariam in “Tax Effort in Sub-Saharan Africa,”
International Monetary Fund Working Paper WP/97/107 (IMF,Washington, D.C., September
1997).
Streeck, Wolfgang, and Philippe C. Schmitter. 1985 (eds.). Private Interest Government: Beyond
Market and State. London: Sage.
Szeftel, Morris. 1998. Misunderstanding African Politics: Corruption and the Governance
Agenda. Review of African Political Economy 25 (76): 221-240.
28
Tangri, and Mwenda 2001
Tangri, and Mwenda 2006
Tanzi 2000
Teorell 2007
Tilly 1990
Van Rijckeghem, Caroline, and Beatrice Weder. 2001. Bureaucratic Corruption and the Rate of
Temptation: Do Wages in the Civil Service Affect Corruption, and by How Much?.
Journal of Development Economics 65 (2): 307-331.
Weghorst, and Lindberg 2010
Weir, Margaret. 1992. Politics and Jobs: The Boundaries of Employment Policy in the United
States. Princeton: Princeton University Press.
Williams, Robert. 1999. New Concepts for Old?. Third World Quarterly 20 (3): 503-513.
World Bank. 2000. Anti-Corruption in Transition. A Contribution to the Policy Debate.
Washington: World Bank.
World Bank 2010
Comments