1. Introduction
Corruption has been identified as one of the primary challenges to be faced by contemporary states. As a result of this process, there has been a gradual increase in societal distrust towards the State and a dissemination of the idea that it is a pervasive phenomenon, contaminating the public sector and the political class, and being inadequately addressed and penalized.
This distrust is a hallmark of contemporary democracies and stems from an inherent dysfunction of the representative system, caused by citizens’ fear that state agents do not fulfill the commitment to promote the common good. Furthermore, corruption, through the utilization of state powers for private interests, is undoubtedly one of the main factors amplifying this sentiment even further.
There is consistent empirical evidence that corruption undermines economic development. In this regard, Wei (1998) records that studies conducted by various authors conclude that the more corrupt a country is, the slower its growth. The author further highlights that corruption undermines economic development in various ways, reducing domestic investment, hindering foreign investment, increasing government expenditure, and distorting the allocation of public expenditures.
Mo’s analysis (2001) concluded that a 1% increase in the level of corruption would be capable of reducing the growth rate by approximately 0.72%. In other words, a one-unit increase in the corruption index would decrease the growth rate by 0.545 percentage points.
Considering this scenario, corruption and its combat assume a significant role in the political agenda of countries integrated into a global economy. Therefore, the examination of aspects related to the phenomenon, its characteristics, and manifestations proves crucial for a thorough understanding of the role of democratic governments and their relationship with society.
With the purpose of providing better conditions for the adequate understanding of corruption, this work seeks to systematize some of the main theoretical approach, highlighting attempts to conceptualize and classify it, as well as some of the studies developed to understand its causes and to propose strategies for its control.
There is an extensive body of theory aimed at studying the factors that influence corruption. These factors can be of various natures and dimensions. In this article, we will primarily highlight approaches that prioritize the determinants of corruption as being of individual and institutional order. It is worth noting, however, that corruption is a complex, multifaceted, and multicausal phenomenon, and therefore is not determined by a limited range of factors.
2. Theoretical Aspects of Corruption
The conceptualization of corruption from the perspective of social sciences is not straightforward. This is because the term “corruption” can encompass a range of behaviors and practices that, depending on a particular set of values or cultural factors, may or may not be deemed morally or even legally reprehensible.
In this sense, as pointed out by Andvig et al. (2000) , the complexity of corruption lies in the fact that it has been viewed not only as a structural problem of politics or economics but also as a cultural and individual issue.
Given these peculiarities surrounding corruption, numerous authors have sought to establish a more suitable concept for the phenomenon and its complexity. Holmes (2015) suggests that corruption, in its most traditional sense, is associated with moral impurity. However, the author acknowledges that, although the concept of corruption has changed over the centuries and varies according to each culture, it has been used to describe deviations from norms that may be considered improper.
Treisman (2000) asserts that corruption can be defined as the misuse of public office for private gain. Meanwhile, Heidenheimer et al. (as cited in Andvig et al, 2000 ) argue that corruption refers to transactions between the private and public sectors, through which collective goods are illegitimately converted into private benefits.
Addressing the challenges of arriving at a precise concept of corruption, Jain (2001) , in a work organizing and systematizing empirical and theoretical studies on the subject, emphasizes that although defining corruption is not trivial, there is almost a consensus in the literature that it refers to acts in which public power is used for personal benefits, contrary to what norms and regulations establish.
All these concepts, quite similar, focus on corruption from a perspective more oriented towards the actions of the public sector. Perhaps for this reason, the non-governmental organization Transparency International (2015) addresses corruption in a broader sense, relating it to the abuse of entrusted power for private gain.
Regarding the literature on corruption, Filgueiras (2008) highlights that, in the 20th century, studies on the subject can be defined based on two main agendas: first, developed from the 1950s onwards, from a functionalist perspective, which views corruption as an obstacle to modernization; and second, hegemonic in the social sciences from the 1990s onwards, based on rational choice theory, which suggests that corruption is related to rent-seeking behavior, where the individual preferences of political agents weigh more heavily as they seek to maximize their private income, within or outside the rules of conduct.
The first of these currents, based on the so-called “modernization theory,” relates corruption to underdevelopment. It argues that the phenomenon represents the malfunctioning of political system organizations, responsible for creating incentive systems that may lead to deviation. According to this approach, corruption would be an accepted practice in underdeveloped societies, given the low level of political institutionalization (HUNTINGTON, as cited in Filgueiras, 2009 ).
The author further emphasizes that, for the functionalist current, corruption could even be useful for development since, if kept under control, it could be an alternative to promote modernization, as it could streamline bureaucracy, expedite the issuance of licenses and documents by the State, and improve the relationship between the public and private sectors, establishing an informal bond between bureaucrats and private investors (LEFF, as cited in Filgueiras, 2009 ).
Additionally, Filgueiras identifies the existence of a certain gap between the two currents, as from the 1970s onwards, literature on corruption takes a different direction, emphasizing the cultural aspect and relating it to “interactions constructed by social actors, reflecting experiences and values that allow individuals to accept or reject a scheme of corruption” (Filgueiras, 2009: p. 396) . In this approach, development would depend on cultural changes that establish a value system in society to prevent the practice of corruption.
Despite this shift in the focus of research on corruption, it is worth noting that, to this day, a wide literature is produced that relates corruption to cultural aspects. Studies by La Porta et al. (1997) , Treisman (2000) , and Serra (2006) , for example, sought, among other aspects, to assess potential impacts that religion could have on corruption. La Porta et al. (1997) identified that, controlling for the variable of per capita income, countries with more hierarchical religions had less efficient judicial systems, poorer quality bureaucracies, lower rates of participation in civic and associative activities, less trust, and higher corruption. For the authors, religion may have discouraged the formation of horizontal networks of cooperation in these countries. Treisman (2000) , on the other hand, found that the higher the proportion of Protestants in a country’s population, the lower the indicators of corruption perception were. As for Muslims, Catholics, Anglicans, and others, no significant impacts of these religions on levels of perception were identified. Serra (2006) included in his analysis the relationships of corruption with five sociocultural variables, such as religion and ethnic-linguistic fragmentation, and identified that different religions have different effects on corruption and that there are indications, as identified by Treisman (2000) , that Protestant countries are less corrupt.
From the late 1980s onwards, the second major current pointed out by Filgueiras (2008) became present. Research on corruption began to consider the economic factor as fundamental. The work of Susan Rose-Ackerman is one of the landmarks of this new approach. Although recognizing that cultural and moral factors have an impact on corruption, the author asserts that what determines it the most is related to an economic approach, based on the premise that agents act based on their individual preferences, seeking to increase their private gain. This conception is based on a behavior called by economists rent-seeking, in which agents use resources not only for productive purposes but also to receive advantages resulting from economic benefits (Bhagwati, 1974, Krueger, 1974, as cited in Rose-Ackerman, 1999 ).
This new way of understanding what determines corruption has led to a fresh reflection on the strategies that should be used to reduce its incidence. Previously, as already observed, corruption was often perceived as a problem more closely related to values, traditions, and culture. It is now seen as a rational phenomenon, in which the agent weighs the cost, represented, for example, by the possibility of sanction, against the economic benefits it may bring.
In addition to these two conceptions, another approach widely used in the literature on corruption is based on agency theory. This is a classic issue in organization theories, known as the principal-agent problem, which deals with the situation where there is asymmetric information between two actors, meaning that the “agent,” who is tasked with carrying out a particular action, has more information than the “principal,” the authority in question. In this scenario, there may be an incentive for the occurrence of the phenomenon called moral hazard, which refers to inappropriate behavior by the “agent,” contrary to the wishes of the “principal.” According to Barro (as cited in Alt & Lassen, 2003 ), under this conception, voters, the principals, choose politicians, the agents, to govern them. However, the interests between the two parties are not always aligned because the authority delegated to agents by principals allows them to act according to their own interests, often in disagreement with the desires of the principals. In this context, from the perspective of agency theory, corruption would basically result from the difficulties the principal faces in controlling the agent’s conduct. This control, in turn, becomes even more complex due to the discretion that often characterizes the agent’s actions. This is because, given the informational asymmetry between the agent and the principal, the more flexible the agent’s actions are in relation to what norms and regulations dictate, the more complex the task of controlling and evaluating their actions becomes for the principal.
Jain (2001) , in examining the types of corruption observable in democratic societies, identified three major groups: grand corruption, bureaucratic corruption, and legislative corruption. The first typically refers to acts perpetrated by the political elite, who wield their power to gain economic advantages through the allocation of funds to projects of their own interest, directing spending to areas where gains from corruption may be greater. Bureaucratic corruption, also known as petty corruption, is practiced by bureaucrats in their dealings with their superiors (the political elite) or the public. Lastly, legislative corruption is related to the power to influence legislators’ votes, who may, for example, be bribed by interest groups to pass legislative measures beneficial to a particular class or category.
Holmes (2015) , on the other hand, distinguishes between petty and grand corruption. While the former is related to the types of corruption affecting citizens in their daily lives, the latter, as its name suggests, refers to corruption at the level of the governmental elite, such as situations where politicians receive bribes to pass laws benefiting certain groups.
Rose-Ackerman and Truex (2012) note that there are studies suggesting that various types of corruption are highly correlated, enabling the identification of more or less corrupt countries. However, the authors emphasize that reform proposals should distinguish between petty and grand corruption, as well as bureaucratic corruption from political corruption, so that the specific characteristics of each sector are considered in these processes.
Building on these definitions and typologies, the following sections will present two approaches to what determines corruption, one focusing on agent behavior and another concentrating on institutional issues.
3. Personalistic Approach to Corruption
One way to observe the determining factors of corruption is through an approach that analyzes it from the perspective of the agent who commits the act of corruption; that is, examining what causes led or will lead them to engage in delinquency.
According to Holmes (2015) , an important theory in the field of criminology that can contribute to a better understanding of corruption is the opportunity theory, which, as the name suggests, refers to acts committed by individuals who frequently seize opportunities that arise to gain personal advantages. Thus, this theory is connected with rational choice theory, as the individual weighs the cost versus the benefit obtained from the act committed.
One of the analyses frequently used to approach corruption from an individual standpoint is one that draws an analogy with the so-called fraud triangle, developed in the classic approach by Donald Cressey (1953) to identify the determining factors of fraud in organizations. In the model’s conception and in his doctoral studies in criminology, Cressey established three conditions that would be determinants for fraud to occur: pressure, opportunity, and rationalization.
In 2004, Wolfe and Hermanson proposed a redesign of Cressey’s model, transforming it from a triangle to a diamond, by including a new variable: the individual agent’s capacity, their ability to perpetrate fraud.
Another perspective that can be used to analyze corruption from the viewpoint of the agents has been developed in research conducted mainly in studies led by Dan Ariely, which focus on the issue from the standpoint of behavioral psychology.
These studies, in addition to concluding that morality is a dynamic and malleable concept, also pointed out that a large part of unethical behaviors results from the actions of various individuals who, although valuing morality and wishing to be seen as ethical people, end up committing offenses, as long as these acts are not sufficient to affect their self-image of honesty. Their research also indicates that dishonesty is “contagious,” meaning that people are more likely to engage in dishonest acts when they perceive that their peers are doing the same (Ariely & Gino, 2016) .
These approaches, while important, are inadequate for a comprehensive understanding of the corruption problem, as they only focus on aspects related to the individual behavior of the agents themselves. This is precisely what Holmes (2015: p. 69) points out. According to the author, “explanations of corruption that focus just in individuals are incomplete, we are all subject to and conditioned by the context in which we live and work.”
In this sense, Mungiu-Pipidi (2024) highlights that individual choices largely depend on the social context, which is why, in her view, in corruption control, policies are much more necessary than judicial processes.
It is evident, therefore, that individual conduct cannot be entirely disregarded. On the other hand, it is essential to consider that, in many situations, institutional conditions also play a significant role alongside individual behavior. These conditions will determine the context in which fraud may occur. For example, the opportunity outlined in the fraud triangle may be generated by the power available to the agent, by the discretion they possess, or by the conviction that they will remain unpunished, given the weakness of institutional controls or the absence of penalization mechanisms.
4. Institutional Approach to Corruption
Another way to identify the factors that determine corruption is to examine it from the perspective of the institutional conditions that either facilitate or inhibit the occurrence of the phenomenon.
According to Jain (2001) , the existence of corruption requires three elements: discretionary power; economic advantages associated with this power, and a legal/judicial system that presents a low probability of detection and/or punishment for the perpetrator of the offense. While the first two factors, economic advantages and discretion, combine to create incentives for corruption, the latter acts as a deterrent to its practice.
It seems evident that economic advantages are clearly a determining factor in corruption. If the behavior of bureaucrats and political agents is analyzed from the perspective of rational choice, it can be concluded that they would tend to act as rent-seekers, seeking to increase their private wealth, as opposed to public interest.
Thus, although there may be situations where corruption is used as a lever to occupy professional positions or public offices (thus providing professional or social status), or, still, in cases where it is used by the political class to remain in power, it is undeniable that, by its very nature, corruption is related, even indirectly, to financial benefits. Based on this approach, the greater the financial advantage obtained by the agent in relation to their expectation, the greater the risk of corruption.
As for the second variable, discretion is the factor that allows the bureaucrat to deviate from norms and regulations to obtain individual benefits. As already mentioned, discretionary power is the one that grants a certain freedom for the action of the public agent, establishing a margin of freedom for the decision to be made (Di Pietro, 2000) . Andvig et al. (2000) describe that the more public agents have discretion, through abundant, complex, and non-transparent rules, the more likely corruption is.
The empirical studies carried out by Kwon (2011) , based on a survey with 800 Korean public servants, indicate that higher levels of discretion (measured by the extent of delegation by supervisors) are positively associated with the susceptibility to corruption of bureaucrats. For the author, discretion is a “double-edged sword”. If, on one hand, higher levels of discretion can improve the performance of public service, on the other hand, they can lead to more corruption.
Susan Rose-Ackerman (apud Anechiarico & Jacobs, 1996: p. 194 ) points out that activities performed by street-level bureaucrats, such as police officers and inspectors, are highly exposed to corruption. She also notes that the bribes they receive may go unnoticed, as this type of employee has broad discretionary power to make case-by-case determinations, which cannot be easily verified by their superiors. She warns, however, that restricting the discretion of these agents may not be an effective remedy for reducing corruption, as it may only result in its transfer to other hierarchical levels.
Pires (2013) highlights that discretion occurs when the norm contains vague or imprecise concepts or when it allows the public agent an alternative conduct. These are some of the characteristics that can create an environment conducive to corruption. The analysis of a sample of 26 African countries by Lambsdorff and Cornelius (apud, Lambsdorff, 2005 ) indicated a positive association between the lack of objectivity and clarity in government norms and corruption.
Regarding the strategies considered most effective in reducing corruption, it is possible to perceive that many authors mention the control of discretion as one of the initiatives that could be important to avoid eventual deviations. Kaufmann (1997) records that evidence clearly indicates that economic reforms are important means to address corruption and that the design of these programs should consider the discretion available to politicians and bureaucrats. According to the author, the main activities that need to be examined in these reforms are those involving a certain degree of discretion, such as issuing licenses, authorizations, and permits, public procurement adjudications, and authorizations for exemptions and tax evasion, among others.
Kwon (2011) emphasizes that imposing limits on bureaucratic discretion, although it has the potential to reduce corruption, can also impair the bureaucracy’s commitment, reducing its productivity in delivering public services.
Andvig et al. (2000) highlight that local agents, in addition to having greater discretion than those responsible for making “national” public policy decisions, are more susceptible to pressure from local interest groups on issues such as taxation.
Therefore, the discretion available to the bureaucracy is an important factor in the occurrence of corruption. However, limiting it does not seem to be, at least in many cases, the best solution. It is discretion that, in theory, would allow the bureaucrat to flexibly adapt the strict rules of public service to the reality encountered, to what best serves the interest of society. The option remaining, therefore, is to analyze ways to control it without, however, making it inefficient or creating more rigidity in the bureaucratic structure.
5. Corruption Control
Controlling corruption is not trivial. As corruption is a multifaceted and multicausal problem, strategies to reduce its incidence must address a series of dimensions. It is, therefore, a complex problem that consequently requires complex solutions. For this reason, it is not uncommon to encounter a series of dilemmas related to combating corruption.
Jacobs (2000) highlights that one of the issues in combating corruption is precisely a tendency to create an environment characterized by a wave of indignation against government corruption, which transforms the phenomenon into such a politically significant issue that it can lead, as one of its consequences, to political opponents seeking to define each other as corrupt in one way or another, while claiming themselves to be blameless in all respects. Such observations are readily identifiable in various cases occurring in several countries, where corruption has come to be viewed with an often moralistic and opportunistic bias, used with the sole intention of public self-promotion by self-proclaimed incorruptible political agents.
Anechiarico and Jacobs (1996) emphasize that another consequence of the fight against corruption may be that governments become less effective and less efficient because many anti-corruption controls tend to reinforce bureaucratic pathologies.
Mungiu-Pippidi (2006) asserts that the fight against corruption has become a significant industry involving substantial global expenditures. However, such investment has generally not resulted in significant successes. For the author, the failure of anti-corruption policies stems from the fact that the adopted anti-corruption measures are non-political in nature, whereas corruption that undermines trust in democracy is obviously political. Thus, in more developed countries, in stronger democracies, corruption tends to be perceived in an individualized, more isolated manner, whereas in weaker democracies it tends to be seen as what the Mungiu-Pippidi calls particularism, which can be simplified as an uneven distribution of power. Many anti-corruption strategies fail precisely because they do not address this, the inequality of power that is at the root of corruption (Mungiu-Pippidi, 2006) .
Regarding the causes of corruption, in addition to the economic advantages and discretionary power, already mentioned in the previous section, the third determinant of corruption pointed out by Jain (2001) is the low likelihood of detection and sanctions. This variable is therefore closely linked to the mechanisms of control and punishment of potential wrongdoers.
According to Tanzi (1998) , at least in theory, if the other conditions are held constant, corruption could be reduced by increasing the possibility of punishment. Citing Gary Becker’s classic analyses on crime prevention, Tanzi points out that a system of sanctions, combined with the existence of institutional controls, are factors that can reduce the incidence of the phenomenon.
Mishra (2006) highlights the importance of control procedures in limiting or fostering corruption. For the author, the success of anti-corruption measures depends largely on efforts to monitor the behavior of agents.
Mingiu-Pipidi (2024) argues that corruption control should be conceived as a balance between opportunities versus constraints imposed on governments based on an independent judiciary, a free press, and enlightened citizens with the ability to demand good governance.
According to Treisman (2000) , the propensity for corruption by public officials is determined through a balance between the costs and benefits of the action. Thus, the agent evaluates the costs, including psychological, social, and financial aspects, and weighs them against the benefits derived from corruption. For the author, political scientists and economists have been suggesting a series of characteristics of the economic, financial, and political systems of countries that may affect this relationship between costs and benefits. The author emphasizes, however, that “the most obvious cost is the risk of being caught and punished” (Treisman, 2000: p. 402) .
One way to understand the nuances of the process of controlling and holding public agents accountable can be seen from the conception of accountability and the actions taken to make it effective. These mechanisms can prevent distortions in the activities of bureaucrats.
According to the classic distinction proposed by O’Donnell (1998) for forms of accountability, while vertical accountability represents the relationship between the population and public authorities, mainly manifested through elections, horizontal accountability is related to the existence of state agencies with the power to carry out control and oversight actions; a mechanism that the literature usually refers to as checks and balances.
Peruzzotti (2008) highlights that the legal apparatus divides the state into jurisdictions that not only regulate the behavior of bureaucrats to prevent personal interests from prevailing over the public interest but also control the degree of discretion in their actions. For the author, the concept of accountability is fundamental for the realization of democracy, as it establishes that governments must be controlled to ensure that there are no deviations in the management of public affairs.
Filgueiras and Aranha (2011) argue that it is control that ensures that discretion does not result in corruption, and therefore, control is essential for building legitimacy within bureaucratic organizations. Thus, processes of transparency and accountability of public officials, for example, can become essential tools in the effort to prevent the actions of bureaucracy from being distorted due to the discretion attributed to it.
In relating corruption to agency theory, Klitgaard (1994) suggests that corruption can be expressed by an equation in which it is equal to the monopoly of decision-making power, plus discretion, minus accountability. Based on this conception, higher levels of control would reduce corruption without necessarily reducing the discretion of bureaucrats.
Tanzi (1998) emphasizes the importance of controls, especially those existing within the institutions themselves, for the prevention of corruption. Thus, transparent procedures, good audit strategies, and clear rules on ethical behavior, among other aspects, would play an important role in reducing the incidence of the phenomenon.
Rose-Ackerman, when addressing anti-corruption policies, highlights the importance of transparency and accountability practices. These strategies should combine how the government operates with opportunities for citizens to monitor and participate in the decision-making process, considering the regulation of the interface between the public and private sectors.
Klitgaard (2008) further asserts that the fight against corruption should focus on reducing monopoly (through increasing competitiveness), clarifying discretion (aiming to demonstrate to society what the rules of the game are), and increasing transparency and accountability, for example, using performance measurement systems and auditing, monitoring, and evaluating public policies.
Regarding the forms of control of bureaucratic actions and the inherent difficulties in these controls, Przeworski (1998) , for example, notes that democratic institutions, in general, lack mechanisms that allow citizens to directly sanction the actions of bureaucrats. Thus, society’s control over bureaucracy occurs only indirectly, at the time of voting, when voters can either sanction or not the behavior of politicians. It is up to elected politicians, therefore, to promote control of bureaucrats so that their conduct aligns with their government project and with what society desires.
All these approaches focus on controlling corruption by increasing the chances of detection and the costs for the agents, in relation to the benefits it may bring. Undoubtedly, these are important measures to reduce its incidence.
However, it must not be forgotten that, as already noted in this same work, corruption stems from multiple causes, ranging from institutional factors to individual and cultural issues. Thus, while its combat should focus on increasing the cost and chances of detection of those involved, corruption prevention, in addition to institutional measures (such as increasing transparency and accountability), should also include initiatives related to cultural and individual changes that may contribute to shaping the behavior of actors. In this sense, actions aimed at disseminating ethical behaviors in society, encouraging civic participation, or strengthening mutual trust between citizens and between them and the State, have been pointed out as important for reducing the incidence of the phenomenon.
6. Conclusion
Corruption is a global phenomenon that affects contemporary democracies. The analysis of its determining factors is therefore essential for its proper understanding and serves as a contribution to future reflections on the challenges that states must face in implementing public policies.
The theoretical review conducted in this work allows for reflection on the complexity of the study of corruption. Starting from the various attempts by various authors to conceptualize it, also passing through the different typologies developed, until reaching the main approaches that seek to study its causes and ways to control it, it is relatively clear that it is a phenomenon of a multi-causal, multifaceted, and highly complex nature.
From such theoretical reflections, it is possible to understand the difficulties and the importance of evaluating the determining factors of corruption to define adequate strategies to reduce its incidence. Therefore, care must be taken with generalizations, universal diagnoses, and causal interferences. If, on the one hand, corruption is a highly complex phenomenon, on the other hand, there are strong indications that it is related to factors of individual, cultural, and institutional order. Hence the importance of such factors being widely considered in the development of strategies designed to prevent and combat it, to improve its control and thus reduce its incidence.