Corruption Control in the Developed World

Conventionally considered a developmental trait that would tend to disappear with the increase of wealth and the stabilization of democracy, corruption is rampant not just among developing countries and recent democracies, but also in mature democracies and developed countries. This editorial introduces the thematic issue and considers what the contributions tell us about new approaches to corruption control in the developed world. It also outlines avenues for future research in the field of corruption control.


Introduction
Since the early 1990s, protests around the globe against corrupt officials have stimulated academic research on the topic of corruption, as revealed by new books being published on a never-ending basis (Bauhr, Bågenholm, Grimes, & Rothstein, in press;Mungiu-Pippidi & Heywood, 2020). Corruption control strategies have drawn on this expanding body of research, but they have proved ineffective in achieving significant and lasting improvement in quality of government. The discrepancy between the high interest in corruption and the low capacity to curb it reveals that the phenomenon is broader and more diversified than the conventional literature and policy recommendations would lead us to think. Whilst previous work on causes and consequences of corruption has helped us to understand broad patterns of corrupt practice, and where it is most deeply embedded, it has been less helpful for identifying what can be done.
For a growing number of researchers, the implementation gap of corruption control efforts lies in the inappropriate theoretical foundations of the standard solutions to the 'principal-agent' problem (Persson, Rothstein, & Teorell, 2019). According to these researchers, the solutions to the 'principal-agent' problem consider corruption as a problem of individual deviance from the system, implicitly assuming that corruption can be tackled if control instruments affect individual agents' motivations to engage in corrupt behavior. They underline that this assumption is flawed in those contexts where corruption is systemic, meaning that corruption is widely perceived as the norm, and those principals monitoring the agents are themselves corruptible if they cannot trust that others will resist corruption.
Viewing corruption as a collective action problem has made an important contribution to the literature by highlighting the very difficult challenge that institutional reforms face in changing levels of distrust in society. However, this vision has little to say about what to do differently, or how (Marquette & Peiffer, 2019). The call for 'big bangs' constitutes the theoretical answer, whereby a multifaceted attack on corruption is applied in an integrated manner to transform the system (Rothstein, 2011).
Yet, the idea of rolling out a broad set of reforms predicated on comprehensive institutional transformation is simply not feasible in most OECD countries where state traditions are so entrenched that 'remaking' the state is unlikely. We thus need better ways of thinking about the design and sequencing of targeted reforms in conditions that are not hospitable for policies that change the basic social contract. In other words, we need greater sensitivity to multiple reform combinations that unfold within different contexts (Fritzen & Dobel, 2018).
Recently, two strands of research have emerged that do not depend on the occurrence of a crisis to implement governance reforms. Drawing on policy making literature (Heclo, 1974), one set of scholars disagree with the 'powering' thesis that underscores those approaches that focus on abrupt and wholesale change. Rather, they understand anticorruption reform as 'problem-solving,' meaning that reformers take advantage of small opportunities and use the complexity of policy areas to advance individually minor but often cumulatively significant changes (Bersch, 2016, p. 206). In contrast to approaches that separate the question of stability from the question of change, the problem-solving take on corruption control echoes the sequencing approach that is well established in the historical-institutionalist literature (Mahoney & Thelen, 2010). It also allows to account for the implications of the two-level governance structure of corruption control. Work on corruption control is characterized by a strong focus on top-down conformance with policy recommendations issued by international organizations. Governments are encouraged to adopt a global menu of tools that are proposed as universal cure for corruption (Rotberg, 2017). Conversely, the problem-solving perspective emphasizes the role of policy implementers within domestic policy arenas. These actors should have flexibility for context-sensitive adjustments in order to address specific policy problems. The problem-solving perspective is all the more relevant in developed settings where compliance is not driven by external pressure through policy conditionality. In these settings, policy actors should first aim at improving those structural factors (fiscal transparency, administrative simplification, professional bureaucracies, etc.) that are prerequisites for the success of the global menu of corruption control tools (Mungiu-Pippidi & Dadašov, 2017). Second, they should select specific corruption control tools that can be incrementally sequenced over time by taking capacity constraints into account.
A second strand of research has highlighted effective subunits that lie hidden within countries known for widespread corruption (McDonnell, 2020). By gauging the workings of high-performing niches that exhibit features distinct from poor-performing peer organizations, this nascent literature shed light on variations and nuances so far overlooked by most studies in the field, which consider the public sector as a monolithic entity (McDonnell, 2017). This literature constitutes the foundation for an approach that promotes change at the micro-level of individual organizations. As more and more organizations improve their performance over time, significant change could spread throughout the public sector (Prasad, da Silva, & Nickow, 2019).
The articles of this thematic issue move the debate forward and point to more targeted interventions for corruption control. All in all, findings support the claim that policies should be underpinned by clearer conceptualization of corruption types, their meanings and functions within specific institutional contexts and policy dynamics (Heath, Richards, & de Graaf, 2016;Heywood, 2017;Jancsics, 2019).

Overview of Contributions
As the title of the thematic issue suggests, our aim was to take stock of the mechanisms through which advanced societies try to control corruption. We were, and still are, particularly struck by the diffusion of corrupt practices in the developed world, that is, in a context in which the layperson would assume that corruption is marginal and exceptional. On the contrary, we know from the literature and from daily news that this is far from being the case and that also developed countries are beset by corruption.
As is commonly the case with phenomena that defy normative expectations, we knew that we would run up against the difficulty of defining corruption. In common language, 'corruption' indicates a negative departure from a normative standard, but what precisely constitutes such a departure heavily depends on the culture, institutions, and procedures of each country. Therefore, we were not surprised to find that some articles tackled also (broadly understood) definitional questions. Bauhr and Charron (2020), for example, distinguish between 'need' and 'greed' corruption, suggesting that the former might be judged less harshly than the latter. Need corruption, moreover, seems to mostly involve women who often carry the burden of caring for the young, the sick and the elderly in the family and are, therefore, particularly sensitive to the urgency of having access to public services which may depend on someone's help in 'cutting the queue. ' Piattoni and Giglioli (2020) similarly suggest that some forms of particularism, which contemplate an exchange between candidates and sectional interests or entire electoral constituencies, may be less serious corruptions of democracy than exchanges that involve individual voters or that imply the exchange of money for selective benefits. They argue that the provision of constituency-level public goods may in fact help wean democracies from graver forms of particularistic exchanges. Although neither article systematically addresses the issue of defining corruption, they both alert us to the dangers of adopting definitions that are too encompassing and that might therefore expand the contours of the phenomenon beyond recognition.
Beyond definition, the core aim of the thematic issue was to canvass the current literature to extract use-ful suggestions on how to control corruption in advanced societies and to find those perhaps less explored mechanisms and tools that can make a real difference. The mirror problem of agreeing on a shared definition is finding a convincing measure of corruption (Heywood & Rose, 2014). The most popular measurement relies on the perceptions of privileged observers (businesspeople, journalists, scholars) which notoriously are sticky and may be influenced as much by hearsay as by direct experience. Perceptions can be shaped by a host of contextual variables that do not necessarily correlate very strongly with more objective measures of corruption based on direct personal experience. One such contextual variable, which has an important effect on the perception of corruption, is the freedom of the press. As the article by Breen and Gillanders (2020) shows a freer press may induce a perception of lesser corruption above and beyond differences in the underlying phenomena. While a free press certainly is an important tool in the fight against corruption, its effect may be more 'cosmetic' than real in that it induces the belief that corruption, if detected and denounced, would be in fact more harshly punished.
A similar reputational effect might be exerted by another contextual variable such as the adoption of a lobby register, one of the standard recommendations of the OECD. De Francesco and Trein (2020) discuss how such a measure may have the effect of curbing the undue influence of business lobbies, by reducing the information asymmetry between public officials and citizens. Nevertheless the stark contrast between the experiences of Slovenia-where a lobby register has been adopted since the 2010s and a shared and regulated notion of lobbying has been promoted-and Italy-where despite several attempts no register has ever been introduced at the national level because of a widespread rejection of the very notion of lobbying and therefore a refusal to regulate it-is very telling.
Both contributions argue that contextual variables that improve the transparency of potentially corrupt deals should make it simpler for businesspeople and citizens to monitor the behavior of politicians and administrators who, in this view, are uniquely interested in extracting unwarranted resources (money or votes) from the members of civil society. Any device that improves the transparency of the dealings between politicians and administrators, on the one hand, and civil society, on the other, should help the latter fight corruption. Businesspeople, voters, and citizens at large may in fact feel relatively powerless in refusing and sanctioning such offers should institutional and structural conditions be perceived as unsurmountable or they may be driven by maximizing calculations to accept them. Transparency measures are certainly important but may also produce frustration rather than resolve (Bauhr & Grimes, 2014). Effective measures should operate also on the supply side of corruption, making it less rewarding for elected and career officials to offer corrupt deals.
The attention gets, therefore, directed to the incentives that may affect the supply side of corruption. Two articles directly address this side of corruption control. Drápalová and Di Mascio (2020) detail how the institutionalization of city managers may drastically improve the quality of municipal governments despite their belonging to regions otherwise affected by widespread corruption and not significantly differing from other, similarly structured surrounding municipalities. They attribute this rather extraordinary result to the professional aspirations of the city managers and to their sensitivity to contextual features of governance. In practice, city managers operate to decouple the promises made in the electoral circuit from the activities performed by the administrative sector and constitute a sort of institutional check that offers elected politicians the possibility of playing a virtuous two-level game with the voters in contexts marked by the longevity of incumbents. This article contributes to the expanding literature dealing with the impact of political competition on corruption control by identifying under which conditions a low level of political competition may support the launch of institutional reforms (Schnell, 2018). It shows that political vulnerability also originates from elections in units different from those under investigation (regional elections and local elections in surrounding municipalities). Mayors in regions dominated by other parties felt constrained and focused on showcasing their ability as good managers. The existence of such pressures implies that political competition occurring in the units of analysis is not capable of telling the entire story and that multi-arena patterns might be important political determinants of institutional reforms at the local level.
A warning against the excessive use of oversight and punishment mechanisms in the public administration comes from the work of Odilla (2020) who draws her empirical material from an innovative dataset of legal prosecutions of administrative wrongdoing in Brazil as well as semi-structured interviews. She examines the effectiveness of the horizontal accountability incentives created within various administrative agencies and discovers that the performance of the 'integrity enforcers' is hampered by reluctance and uncertainty. Unless the investigation of administrative corruption is entrusted to specialized structures and their operations streamlined, the danger of a discretionary pursuit of cases of corruption will act as a deterrent against the diffusion and standardization of these practices. Together these articles drive home the message that the public administration is a crucial intervening variable in most attempts to curb corruption and that the professionalization of bureaucracies is crucial. Most of the literature on corruption still overlooks the management of public officials and this is a significant omission given the role that these actors play in corrupt governmental networks (Della Porta & Vannucci, 1999;Jancsics & Jàvor, 2012). Therefore, there is need for research assessing the effects of a broad set of public personnel management practices to gain a deeper un-derstanding of corruption, and how to curb it (Meyer-Sahling, Mikkelsen, & Schuster, 2018).
Patronage-that particular form of corruption according to which jobs in the public administration are distributed to friends and political supporters rather than according to merit-is not only bad in itself because it deprives the state of much needed professional skills, but is also instrumental in multiplying and spreading to the entire system corrupt practices that have to rely on the complacency of the public administration to be perfected. Sometimes, a perception of widespread corruption is as damaging as its actual diffusion. If corruption is perceived as systemic and if, on the other hand, corruption control is perceived as selective, the effectiveness of integrity enforcement and the rational calculations of all agents involved will be negatively affected. That this might be the case even in otherwise very different countries (also in terms of the corruption perception index that characterizes them) is further confirmed by the article by Gisladottir, Sigurgeirsdottir, Stjernquist, and Ragnarsdottir (2020) who study the corrupt practices that surround the management of the fishing and timber sectors in Iceland and Romania, respectively. The loops that describe the management of these two economically crucial sectors-that hinge upon renewable, but also depletable, resources-are not identical in terms of the incentives and perceptions that they create but lead to surprisingly similar phenomena. If, for whatever reason, corruption control is perceived as ineffective or for going only after the 'small fish,' then the incentives for stepping it up decrease and resignation and cynicism rather take hold. Fortunately, new technology may lend a helping hand by making monitoring of over-fishing and over-harvesting simpler and accessible to a larger pool of concerned individuals that may amplify the enforcement capacity of the institutional inspectors.
Fazekas and Wachs (2020) draw our attention back to the incentives that affect the political class, on whose decisions all other institutional incentives depend. They discover that in public procurement-a classically corruption-prone area of administrative activitycorruption operates to discriminate against certain providers and to hamper the competitive functioning of the market. In other words, corrupt public procurement networks are thinner than non-corrupt ones. Contested political elections and government turnover lead to the renegotiation of the contracts and to the reconfiguration of the networks, which opens up at least the possibility of replacing some favored providers. They conclude that a well-functioning democracy characterized by competitive elections and alternation in government should increase the chances of breaking corrupt networks.
This article brings us back to the political level. We infer from this perusal across many different OECD countries, levels of government, and institutional branches that corruption control in advanced societies can be contained only thanks to context-specific mechanisms that both reduce the incentives to engage in corrupt deals for the actors that lie at the supply end and increase the convenience to monitor and punish for the actors that stand at the demand end.

Future Research
In conclusion, we outline avenues for future research in the field of corruption control. While elections are expected to curb corruption, empirical tests of this expectation have produced inconclusive results. Thus, there is still room for work on factors like information and loyalty that undermine accountability for corruption (De Vries & Solaz, 2017). The surge in populist movements has encouraged corruption control expectations without delivering results, and this has further widened the gap between voters and representative institutions (Mungiu-Pippidi, 2020). Anti-corruption campaign has helped populists to flourish, but populists themselves seemingly have thought less about introducing effective governance mechanisms than about rhetoric (Peters & Pierre, 2019). This underlines the need for a broader research agenda on populism, anti-corruption rhetoric and good governance (Bågenholm & Charron, 2015).
Another avenue of research regards regulatory and institutional innovation that has been a distinctive feature of corruption control policy in the last few decades. While anticorruption agencies have been widely adopted, there is a limited amount of scholarship on such agencies. An emerging literature has explored the impact of organizational factors and leadership on the effectiveness of anticorruption agencies (Di Mascio, Maggetti, & Natalini, 2020;Tomic, 2019). These studies revealed that agencies' effectiveness is not crucially shaped by their statutory independence, but rather by the reputational management of their leaders. This finding calls for wider inquiry into drivers of agency autonomy and performance.
It would also be worth to re-consider the relationship between corruption and regulation. Dunlop and Radaelli (2019) have reviewed the more frequent claims about regulation and corruption: Deregulation hinders corruption; it is the quality of regulation that hinders corruption; specific anti-corruption heavy regulatory frameworks raise the cost of applying for public procurement and funding, while regulatory complexity resulting from the layering of anticorruption measures makes paradoxically non-compliance harder to detect. Dunlop and Radaelli suggest to re-cast this debate by focusing on the combination of policy instruments that affect rulemaking (judicial review, regulatory impact assessment, freedom of information acts, etc.). This opens a new area of inquiry that would benefit from work on data that is needed to examine variations in patterns of rulemaking.
Finally, it is often argued that more of the responsibility for anticorruption should be delegated to local communities, civil society actors, and ordinary people, whose mobilization against corruption might take advantage of digital technologies (Kossow & Kukutschka, 2017;Zinnbauer, 2015). Future research should focus on gathering more evidence on organizational and individual determinants of the decision to report wrongdoing and fight corruption (Su, 2020;Su & Ni, 2018;Taylor, 2018). This would help understand how dissatisfaction with corruption can be channeled to bring about change (Peiffer & Alvarez, 2016).