Why Civil Society Cannot Battle it All Alone: The Roles of Civil Society Environment, Transparent Laws and Quality of Public Administration in Political Corruption Mitigation

ABSTRACT

Utilizing a large-N data that covers about 20000 observations from about 200 countries from 1789 to 2018 from the Varieties of Democracy (V-Dem) project, and anchored on institutionalism as an overarching theory, and the nascent literature on civil-society corruption nexus, the paper looks at the predictive capacity of civil society environment, transparency of laws and predictability of enforcement, and rigorousness and impartiality of public administration in political corruption. Using a four-step hierarchical multiple regression, results show that while civil society and its structure is a significant determinant of the level of political corruption, the introduction of transparency of laws and predictability of enforcement, rigorousness, and impartiality of public administration, and civil society environment in the regression model accounted for additional variance in political corruption. Practical and theoretical implications, particularly on civil society-corruption nexus and the broader corruption-democracy linkage, are discussed.

KEYWORDS:

• Civil society
• corruption
• public administration
• transparency
• institutionalism

Introduction

That corruption, defined by the World Bank and the extant academic literature (see Amundsen, 1999; Chang & Chu, 2006; Desta, 2006; Gerring & Thacker, 2004; Manzetti & Wilson, 2007; Rose-Ackerman, 2008; Shleifer & Vishny, 1993; Sung, 2002;) as the abuse of public office for private gain, is a pervasive and an endemic problem around the world (United Nations Convention Against Corruption) is non-debatable. For Transparency International, no one country in the world is corruption free, including those in the G20 and even those considered as long-lived democracies. True enough, as Stevens (2016) argued, “the formal institutions that are most likely to be involved in the causes of corruption include the institutions of democracy” (p. 185). Given its serious socio-economic and political effects, leading development organizations have emphasized the need to curb corruption (Gans-Morse et al., 2018). While there is a unanimity that corruption is detrimental to the interests of society in general (Brown, 2007, Foreword), the search for the ways to combat corruption has led to not only varied but also unclear results of what works, or as Gans-Morse et al. (2018) aptly put, “ … a clear sense of which anti-corruption policies are likely to succeed” (p. 172). This is expected given the multidimensional and multilayered nature of corruption.

For the most part, the economistic perspective of corruption (see Rose-Ackerman, 1999) has become influential in the scholarly literature. However, as critics note, this narrowly technical view has downplayed other important perspectives. As Hindess (2013) posited, “to treat the problem of corruption as if it were really amenable to technical solution is also to ignore the fundamentally contentious character of political life” (p. 10). It seems therefore that a broader political perspective is needed to shed light into some of the important questions both policy-makers and scholars raise about what works to combat corruption, apart from those offered by an economistic view. One equally important body of work that looks at corruption from a contentious, political perspective is that offered by the civil society literature. The extant literature on civil society-corruption nexus stresses on the import of civil society organizations on corruption mitigation. However, little research has been done about the conditions under which civil society organizations impact corruption despite calls from scholars in the field to probe into these. Specifically, much less has been done on the quality of public administration, that is the rigorousness and impartiality of public administration, as a condition for civil society’s anticorruption effects. The paper pursues this direction and hopes to contribute to this field.

The article is organized as follows. A review of the literature on civil society-corruption nexus followed by the relationship between public administration, institutions and civil society is presented. The hypotheses of the article follow. The next section describes the data, variables, and methods. Results are then presented. The last section concludes and provides the limitations of the study and prospects for future research.

Review of related literature

Civil society and corruption

The civil society-corruption nexus literature is currently divided between those who believe that civil society’s impact in mitigating corruption is undeniable (the optimists) (see for instance Grimes, 2013; Mungiu-Pippidi & Dusu, 2011; Tusalem, 2007), and those who claim that they have by themselves become corrupt or conduits for corruption (the skeptics) (see Gibelman & Gelman, 2004; Greenlee, Fischer, Gordon, & Keating, 2007). Within the first group, development agencies and scholars contend that civil society plays a key role in fighting corruption. No less than the Organization for Economic Co-operation and Development (OECD) and the United Nations Development Programme (UNDP) asserted that the civil society is important in the global fight against corruption (OECD, 2003; UNDP, 2008). This anticorruption function is manifested in a number of respects. First, not only does civil society provide information about or raise awareness on corruption and report governmental malfeasance (Grimes, 2013); it also serves as a vehicle for the mass public to articulate their grievances and associate or organize to call for and promote transparent and accountable state institutions (Tusalem, 2007). Second, civil society diagnostically assesses and monitors the performance of public institutions (Mungiu-Pippidi & Dusu, 2011). Civil society, in the words of Grimes (2013), thus occupies an important role in a polity’s meta-system of checks and balances. Third, civil society partakes in anticorruption policy advocacy and redesign of anticorruption institutions in states (Setiyono & McLeod, 2010; Wampler & Avritzer, 2004).

Despite these optimistic views, there are those who cast doubt on the decisive role of civil society in anticorruption efforts. The first group of scholars in this strand point to challenges to the internal accountability mechanisms of civil society organizations themselves. That civil society organizations are often not very transparent to the public or held accountable is a recurring theme within this second strand (Ebrahim, 2003; Townsend & Townsend, 2004). Gibelman and Gelman (2004), for example, advanced that among many other characteristics, the failure to institute internal controls, lack of oversight and absence of checks and balances in procedures and practices as well in nongovernmental organizations breed corruption. Greenlee et al. (2007) also added that nonprofit organizations suffer from occupational fraud and thereby experience financial losses due to their limited financial management capacity where “essential tasks are undertaken by individuals with little financial expertise and no training in the design of appropriate controls against errors and fraud” (p. 690) coupled by an atmosphere of trust that discourages monitoring.

The second group of scholars within this strand questions the autonomy of civil society organizations, both from the state and from their donors. For example, with reference to CSOs in the Middle East, Wiktorowicz (2000) claimed that when CSOs are embedded in a web of bureaucratic practices and legal codes which allows those in power to monitor and regulate their collective activities, the civil society becomes more an instrument of state social and political control than a mechanism for empowerment. Similarly, with reference to the case of Turkey, Doyle (2018) posited that government co-option of civil society transpires and thus CSOs function to disseminate government ideas in society and to provide a cloak of democratic legitimacy to policy decisions. Indeed, as Lewis (2010) provided, while non-governmental organizations (NGOs) may act as organization spaces for activism, they also present spaces into which governmental power can be projected. The extent to which the state or government does this is also evident in the literature and it runs from the idea of transformation of CSOs into “public utilities” and “semi-public agencies” (van Biezen, 2004 in Saglie & Sivesind, 2018) and “state agencies” (Dreher, Molders, & Nunnenkamp, 2007) to states’ colonization of voluntary organizations (Lorentzen, 2004 in Saglie & Sivesind, 2018).

CSOs’ sources of funding, be it the state or external donors, also have serious implications on their autonomy. For instance, it is claimed that NGOs in developing countries rely heavily on foreign donor funding and donor dominance is evident (AbouAssi, 2012) to the extent that NGOs re-align their priorities with donor interests (Parks, 2008) and appropriate activities to funders’ priorities and demands (AbouAssi, 2014; Mosley, 2012). In this case, CSOs may thus have limited possibilities of making ethically consistent decisions in view of their sources of funding (Egerer, Kankainen, & Hellman, 2018). Similarly, the heavy reliance on external donor’s funding by CSOs resulted in its disconnect from local publics (Dupuy, Ron, & Prakash, 2015) or as Ishkanian (2014) noted, with particular reference to post-Soviet societies, to its reduction to professionalized NGOs that were engaged in advocacy which supported liberal Western values, thereby losing its diversity and authenticity – a stab on the very legitimacy of CSOs. Indeed, as in recent years, studies about and calls for NGO accountability are continuously mounting (see for example, Ebrahim, 2003; Jepson, 2005; O’Dwyer & Boomsma, 2015; O’Leary, 2017).

As an emerging body of literature, and which narrows the supposed divide between the optimists and skeptics, a possible third strand looks at the conditions under which civil society may affect control of corruption, including media, government transparency, political competition, and important legal, political and socioeconomic contexts partly shaped by national governments (see Donaghy, 2011; Marinova, 2011; Uhlin, 2009, 2010; Widojoko, 2017). What is common among the last group of scholars above is the belief that civil society cannot constitute a single, independent force in the anticorruption movement and reform. The current study is directed towards an exploration of these conditions.

Linking public administration, civil society, and corruption

The public administration and policy literature are of course not silent on the issue. In particular for instance, the literature on policy failure in corruption control stresses on the problem of agency (see Fritzen, 2005; Mungiu-Pippidi, 2010); insufficiency of existing laws (Anechiarico & Jacobs, 1996); weakness of law enforcement agencies (Batalla, 2015); and policy transfer (Minogue, 2002). Similarly, in the policy implementation body, for instance target compliance, among the factors why those whose behavior the anticorruption policy seeks to change fail to act as expected are: lenient penalties coupled by low detection and conviction rate; autonomy problems; and information deficits (Batory, 2012). Brown (2007) succinctly puts that the success of anticorruption initiatives will depend partly on answers to three major questions: (1) who owns the anticorruption efforts being undertaken?; (2) is leadership being provided from the highest levels of government?; and (3) does the state have the capacity to implement anticorruption strategies and campaigns through effective detection, investigation, and prosecution? (Foreword, xi).

Interestingly, the scholarship on state-civil society and bureaucracy/administration-civil society (policy) partnerships bridges these two preceding bodies of literature, and these raise important points in the study of corruption and anticorruption policy, as well. Initially, as Anechiarico (1998) noted, the scholarship produced by the civil society literature and those of public administration were separated due to increasing professional and academic specialization compounded by other factors. He cautioned that the civil society scholarship neglects those aspects of public administration, which are important in the civil society literature’s central element, civic engagement. Conversely, he continues, “the movement to “reinvent government” in public administration, based on a model of the citizen-consumer, neglects the importance of civic engagement to its central element, government effectiveness” (p. 13). In the broader development literature, this symbiosis between civic engagement and government effectiveness is heralded as well. Serageldin (1996) for instance contends that strong, effective, and efficient governments are essential to development because they have the capacity to create an environment conducive for the private sector and civil society to flourish (as cited in Birner & Wittmer, 2006). Control of corruption, together with regulatory quality and effective service delivery, is not only an aspect of the agenda of good governance (Kaufmann, Kray, & Mastruzzi, 2003) but also of the bigger agenda of development. Needless to say, it is in these contexts that effective and capable public administration is material.

Peters (2001) suggested that the interaction of administration with both formal and informal political actors in the society has a profound impact on the behavior of administrators and on their decisions. The import of this relationship between bureaucracy and civil society in policy is highlighted in the extant literature. For instance, Rashid (2014) posited that the quality of policy inputs of a bureaucracy is affected by three important factors: (1) political influence in bureaucratic functions; (2) decline of bureaucratic capacity as to policy support and management; and (3) weak engagement of bureaucracy with civil society and nongovernmental organizations. As to administration, the literature on collaborative governance and participatory governance also emphasizes this relationship. For instance, looking at the case of Guatemala’s forest administration, Birner and Wittmer (2006) advanced that the success of Instituto Nacional de Bosque (INAB) was based on its institutional design anchored on the two principles of delegation and partnership. That delegation of authority to an autonomous body characterized by partnership with the private sector and civil society proved successful even more so in a sector where mismanagement and corruption were widespread. Several other landmark cases where the partnership proved challenging but successful were the Philippines’ Department of Education’s (DepEd) Textbook Count that sought to address concerns on the transparency, accountability, and efficiency of the DepEd through the help of the civil society organization Government Watch (G-Watch) (Leung, 2005); and Brazil’s Porto Alegre case where participatory budgeting has reduced opportunities for favoritism and challenged the infrastructure of clientelism (Abers, 1998). However, some scholars point to the downsides of participatory policy-making and deliberation. For instance, calling it “participatory-deliberative public administration” (PDPA) whose main goal is the promotion of a rejuvenation of democratic institutions and progressive politics by favoring direct civil society involvement in public policy-making, Baccaro and Papadakis (2009) claim that policy developments in South Africa point to yet one important defining element of the relationship mentioned in the preceding discussion which deviates from the central elements of participatory governance: “the state does not just benevolently devolve, as PDPA theory assumes, but rather has clear preferences about the kind of policies it wants participatory fora to adopt, generally as a result of international macroeconomic pressures” (p. 247). The question of whether the state has predetermined options for civil society in the deliberation process thus surmounts. More importantly though, noting the success of participatory budgeting in Porto Alegre as compared to those in South Africa, Baccaro and Papadakis (2009) contend that “participatory institutions may work in particular circumstances, but also that, in the absence of those circumstances, they may fail dramatically” (p.270).

It follows from the preceding discussion that the environment within which civil society operates, not only its structure or other internal characteristics, condition its influence on corruption. The broader institutionalist framework is informative in this regard. The significance of institutions in addressing the corruption problem is widely highlighted in the extant literature. Open and transparent political institutions (Alt & Lassen, 2003; Lindstedt & Naurin, 2010), strong rule of law and well-established political-legal structures (Zhan, 2012) including the justice system institutions (Rios-Figueroa, 2012), electoral rules and constitutional framework (Kunicova & Rose-Ackerman, 2005), the design and structure of government institutions and political processes (Shleifer & Vishny, 1993) and anticorruption commitment rules (Collier, 2002) and agencies and enforcement organizations (Yang, 2009) inhibit corruption.

Hypotheses

From the above discussion then, and following the third strand in the civil society-corruption nexus literature, I argue that while civil society and its structure affects corruption mitigation, such effect is conditioned by a number of factors. Firstly, the environment that regulates specifically the entry and exit of CSOs and the extent to which the government attempts to repress CSOs is important in the CSO’s anticorruption effects (H1). Secondly, anchored on institutionalism as an overarching theory, I also argue that the broader politico-legal institutional environment is as well significant. I hypothesize, therefore, that the presence or absence of transparent laws with predictable enforcement is material in accounting for civil society’s supposed effect in curbing corruption (H2). Lastly, gleaning from public administration and the bureaucracy-civil society relationship body, I hypothesize that rigorous and impartial public administration conditions not only the effect of civil society structure but also the effects of civil society environment and transparent laws and predictable enforcement in political corruption (H3).

Data, variables, and methods

This paper relies on the data from Varieties of Democracy (V-Dem),¹ which covers about 200 countries with a time-series from as early as 1789 to present. The data used here are measured in an ordinal scale and calculated into interval scale by the measurement model of the V-Dem, except for political corruption, which was already in interval scale.²

Independent variables

The key independent variables are civil society structure, civil society environment, transparency of laws and predictability of enforcement, and rigorousness and impartiality of public administration.

Civil society structure

This variable measures the strength of CSOs. It is argued that the strength of CSOs is significant in its anticorruption effects but it is conditioned by three factors: civil society environment, transparency of laws, and rigorousness of public administration. Two (2) variables from the V-Dem are used to indicate civil society structure: CSO participatory environment (v2csprtcpt), which describes the involvement of people in civil society organization; and CSO consultation (v2cscnsult), which describes whether major civil society organizations are routinely consulted by policymakers on policies relevant to their members. Similar to the succeeding variables, the preceding indicators flow from less democratic to more democratic, unlike in the dependent variable, political corruption index, which flows from less corrupt to more corrupt.

Civil society environment

This is divided into two (2) main variables from the V-Dem, which are CSO repression (v2csreprss), which looks at whether the government attempts to repress civil society organizations; and CSO entry and exit (v2cseeorgs), which assesses the extent to which the government achieves control over entry and exit by civil society organizations in to public life (see Bernhard, Tzelgov, Jung, Coppedge & Lindberg, 2015; Coppedge et al., 2018).

Transparency of laws and predictability of enforcement (v2cltrnslw)

This variable measures the extent to which the laws are transparent and predictable. Specifically, it asks the question: are the laws of the land clear, well publicized, coherent (consistent with each other), relatively stable from year to year, and enforced in a predictable manner? (see Pemstein, Marquardt, Tzelgov, Wang, Krusell & Miri, 2018; Coppedge et al., 2018).

Rigorousness and impartiality of public administration (v2clfmove)

It focuses on the extent to which public officials generally abide by the law and treat like cases alike, or conversely, the extent to which public administration is characterized by arbitrariness and biases. It asks whether public officials are rigorous and impartial in the performance of their duties (see Pemstein et al., 2018; Coppedge et al., 2018).

Dependent variable

Political corruption index

This is an aggregate of four different variables in the V-Dem, which reflect how pervasive political corruption is in the countries. The political corruption index (v2x_corr) includes six distinct types of corruption that covers both different areas and levels of the polity realm, distinguishing between executive, legislative and judicial corruption. The index includes two other indices, namely, public sector corruption index (v2x_pubcorr) and executive corruption index (v2x_execorr), and two other variables, namely, legislature corrupt activities (v2lgcrrpt) and judicial corruption decision (v2jucorrdc) (see McMann, Pemstein, Teorell & Seim 2015; Coppedge et al., 2018).

A four-step hierarchical multiple regression was performed to investigate the predictive capacity of civil society environment, transparency and predictability of laws, and rigorousness and impartiality of public administration factors on the level of political corruption (N = 23652) while controlling for the effect of civil society structure. Preliminary analyses were conducted to test whether assumptions of normality, linearity, and homoscedasticity were violated. Results indicate that multicollinearity was unlikely to be a problem. Missing data were handled using listwise deletion for better comparability.

Results

The summary results of the regression are presented in Table 1. Model 1 presents the variation on political corruption based on the civil society structure. In order to look at the supposed effect of civil society environment on political corruption while controlling for the effect of civil society structure, Model 2 is provided (H1). Model 3 presents the effect of transparency of laws and predictability of enforcement while controlling for the effects of the civil society structure and civil society environment (H2). The fourth and last model presents the effect of rigorousness and impartiality of public administration on political corruption (H3).

Table 1. Summary of hierarchical regression analysis for variables predicting political corruption (N = 23652).

Variables	Model 1			Model 2			Model 3			Model 4
	B	SE B	β	B	SE B	β	B	SE B	β	B	SE B	β
Civil Society Structure
CSO participatory environment	−.039	.002	−.199**	−.038	.002	−.193**	−.025	.002	−.127**	−.025	.002	−.125**
CSO consultation	−.038	.002	−.188**	−.032	.002	−.157**	.011	.002	.052**	.022	.002	.108**
Civil Society Environment
CSO entry and exit				.037	.003	.197**	.062	.002	.330**	.049	.002	.262**
CSO repression				−.048	.002	−.248**	−.017	.002	−.085**	−.006	.002	−.030*
Transparency of Laws and Predictability of Enforcement							−.135	.002	−.700**	−.054	.002	−.280**
Rigorousness and Impartiality of Public Administration										−.111	.002	−.567**
R	0.360			0.378			0.600			0.689
R^2	0.129			0.143			0.360			0.475
R^2change	0.129			0.014			0.217			0.115
F for change in R^2	1757.102**			189.378**			8009.485**			5178.903**

Note. Statistical significance: *p< .05. **p< .001

In the first step of the hierarchical multiple regression (Model 1), civil society structure, a measure of civil society strength indicated by two variables from the V-Dem, namely CSO participatory environment and CSO consultation were entered. Civil society structure accounts for 36% of the variance in political corruption and the model was statistically significant (F (2, 23649) = 1757.102, p < .001). Both indicators were significant predictors of reduction in political corruption: CSO participatory environment (β = −0.199) and CSO consultation (β = −0.188).

In the second step (Model 2), the predictor civil society environment which has the following indicators was entered: CSO entry and exit and CSO repression. This model was statistically significant (F (4, 23647) = 987.236, p < .001) and explained 37.8% of the variance in political corruption. All civil society environment factors made a significant unique contribution to the model. The best predictor of reduction in political corruption in this model is CSO repression (β = −0.248) followed by CSO participatory environment (β = −0.193) and then CSO consultation (β = −0.157). It is important to note that CSO entry and exit did not lead to reduction in political corruption, and this result was statistically significant (β = 0.197). Overall, the entry of civil society environment resulted in only 1.4% additional variance in the dependent variable.

After entry of the variable transparency of laws and predictability of enforcement in Model 3, the total variance on political corruption was 60% (F (5, 23646) = 2659.163, p < .001). While controlling for the effect of the civil society structure and civil society environment, transparency of laws and predictability of enforcement accounted for the additional 21.7% variance in political corruption. All the factors entered made a unique significant contribution to the model, but it is important to note that the best predictor of reduction in political corruption in this model is transparency of laws and predictability of enforcement (β = −0.700) followed by CSO participatory environment (β = −0.127) and CSO repression (β = −0.080). It is also noteworthy that with the entry of the variable transparency of laws and predictability of enforcement in this model, CSO consultation did not reduce political corruption (β = 0.52) unlike previously noted in Model 2.

In the final model, when the variable rigorousness and impartiality of public administration was entered, 68.9% (F (6, 23645) = 3564.363, p < .001) of the variance in political corruption was accounted for. The introduction of such a variable in the model, while controlling for all the other variables, accounts for an additional 11.5% variance in political corruption. While all the variables entered made a unique significant contribution to the model, the following were the best predictors in the reduction of political corruption: rigorousness and impartiality of public administration (β = −0.567) followed by transparency of laws and predictability of enforcement (β = −0.280), CSO participatory environment (β = −0.125), and CSO repression (β = −0.030). It is important to point out as well that with the entry of rigorousness and impartiality of public administration in this model, CSO consultation did not lead to reduction in political corruption (β = 0.108) (unlike previously in Model 2) similar to that reported in Model 3 when the variable transparency of laws and predictability of enforcement was entered.

Discussion and conclusion

The results above confirm the argument put forward in this paper: while civil society and its structure is a significant determinant of the level of political corruption, the introduction of civil society environment (Model 2), transparency of laws and predictability of enforcement (Model 3) and rigorousness and impartiality of public administration (Model 4) in the regression model accounted for additional variance in political corruption. However, and more importantly, of the three predictors entered after civil society structure (in Model 1), it was transparency of laws and predictability of enforcement that had the highest additional variance (21.7%), followed by rigorousness and impartiality of public administration (11.5%), and civil society environment (1.4%).

These results point not only to the import of institutional arrangements, transparency of laws and predictability of enforcement in this case but also to the quality of public administration, in curbing corruption. While the results echo the suggestions made previously by scholars that the reduction of corruption is founded on the presence of legal (and political) institutions, the current study points as well to the idea that corruption mitigation is also founded on public administration, and especially to the agents, public officials if we may. Moreover, while the current scholarship on the conditions that affect civil society’s impact on anticorruption point to the relevance of free media (Camaj, 2013; Themudo, 2013), politico-legal institutions (Marinova, 2011), historical antecedents (Baiocchi, Heller, & Silva, 2008), and political parties (Morlino, 2011), the current study highlights the supposed role of the quality of public administration and its administrators in the civil society-corruption nexus, one that has been disregarded in the extant scholarship. While Brown (2007) claimed that the success of any anticorruption initiative will also depend on the answer to the question “Is leadership provided from the highest levels of government?”, the current study asks whether such leadership is also rigorous and impartial, or whether it is characterized by arbitrariness and biases. As such, even the debate on the relationship between civil society and corruption, and the anticorruption effects of civil society to be more specific, is also founded on the discussions on the relationship between structure and agents, one that is also at the heart of much of the social sciences. This leads us back to, for instance, the importance of agency (Fritzen, 2005; Mungiu-Pippidi, 2006) together with sufficiency of existing laws (Anechiarico & Jacobs, 1996) and strength of law enforcement institutions (Batalla, 2015) in corruption mitigation.

Indeed, civil society cannot battle it all alone. That civil society organizations are an all-powerful actor by themselves maybe an overestimation of civil society’s potential. The contributions of civil society in anticorruption work are considered futile without the preceding important mechanisms discussed above. The success therefore of any actor in anticorruption reform and movement also hinges on the environment national governments partly shape, including the environment within which civil society organizations work (specifically participatory and repressive ones as studied here) and the broader politico-legal institutional framework where governance and administration take place in general. But several notes of caution must be made here, and this has serious implications not only on the supposed civil society-corruption nexus but also on the corruption-democracy linkage in general, and on the import of the quality of public administration as an important condition for civil society’s anticorruption effects. It is important to reflect on the question of whether the state has predetermined options for civil society (Baccaro and Papadakis, 2009; Doyle, 2018), not only in participatory governance or policy-making but also and most especially in the anticorruption agenda of governments. For instance, it was found out in this study that, taken all together in the final regression model, CSO consultation did not lead to reduction of political corruption. It is important thus to not only ask the question of whether CSOs are routinely consulted by policymakers; it might be prudent to ask whether states have already had predetermined options for civil society in its anticorruption agenda, thus making CSO consultation irrelevant in the equation. Similarly, more than the frequency of inclusion of CSOs in the policy-making process by state actors (Schrama & Zhelyazkova, 2018), the quality of engagement with CSOs matter. As Rashid (2014) emphasized, weak engagement of the bureaucracy with civil society and nongovernmental organizations affects the quality of policy inputs of a bureaucracy. Perhaps, it is only when the quality of engagement with civil society is improved that the positive effect, that is reduction of political corruption, of CSO consultation transpires. Moreover, several scholars have warned of the effect of current trends toward democratic recession such as the squeezing of civic space (Buyse, 2018) and on how state elites have increasingly used the NGO legal environment as a ruling strategy to disable dissent (Dupuy et al., 2015; Gilbert & Mohseni, 2018) on the vibrance of civil society and by extension, its contribution to the enhancement of democracy and governance. However, it is puzzling but very important to note yet again that as found out in this study, similar to CSO consultation but unlike CSO repression, CSO entry and exit, defined here as the extent to which the government achieves control over the entry and exit of CSOs into public life, did not lead to a reduction in political corruption as the regression models provide. Regulatory, consultative and repressive environments or perhaps institutions have differing effects on political corruption.

While the current study used a large-N data that covers about 20000 observations from about 200 countries from 1789 to 2018, it is limited in a number of ways and these provide prospects for further research. First, other control variables can be used to assess the robustness of the results presented here. For instance, it might be worth revisiting the economistic view of corruption as put forward in the introduction. How do economic variables factor in this model? Second, and still founded on the overarching theory of institutionalism, one might be prompted to ask how different are democratic polities and autocracies or presidential from parliamentary systems in this regard? Third, given the onset of digital revolution and the ever-growing significance of the Internet in politics, it may also be material to look at how the state of a country’s e-government development and e-participation conditions civil society’s anticorruption effects. Fourth, and most importantly, the regression model offers only one formula for political corruption. Could there be multiple pathways to corruption, given that as emphasized earlier in the introductory part, corruption is multidimensional and multilayered? Several techniques can be used to look at this. Qualitative comparative analysis (QCA), for instance, is informative in this regard.

Acknowledgments

I want to thank the peer reviewers for the constructive feedback. I also would like to thank the editor of this journal, Ali Farazmand, for the helpful comments. Errors are mine alone, of course.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

The present publication is the outcome of the project “From Talent to Young Researcher project aimed at activities supporting the research career model in higher education”, identifier EFOP-3.6.3-VEKOP-16-2017-00007, co-supported by the European Union, Hungary and the European Social Fund. 

Notes

1. It is a new approach to conceptualization and measurement of democracy. It is co-hosted by the University of Gothenburg and University of Notre Dame (Coppedge et al., 2018).

2. See Coppedge, Michael, John Gerring, Carl Henrik Knutsen, Staffan I. Lindberg, Svend-Erik Skaaning, Jan Teorell, Joshua Krusell, Kyle L. Marquardt, Juraj Medzihorsky, Daniel Pemstein, Josefine Pernes, Natalia Stepanova, Eitan Tzelgov, Yi-ting Wang, and Steven Wilson. 2018. “V-Dem Methodology v8”. Varieties of Democracy (V-Dem) Project for discussion on the methodology of the V-Dem.