Exploring Performance-Related Pay as an Anticorruption Tool

In recent decades, PRP reforms have increasingly been implemented in public administrations (OECD 2005). Estimates hold that PRP programs support nearly 300,000 federal employees in the USA alone (USOPM 2007, 3). One may also find such schemes in transitioning and developing countries. For instance, the performance-share of a salary for public employees in Ukraine was as high as 80% in 2006 (OECD 2014).

The occurrence of PRP reforms is a part of the trend in public administration, stemming from the New Public Management (NPM) doctrine (Peters 2009).² As such, these reforms are one of the several examples of “market mimicking” mechanisms. PRP programs build on the premise that financial incentives are a cost-efficient way to guide self-interested subordinates’ behavior in line with managerial expectations (Miller and Whitford 2007). Given its influence, it is not surprising that scholars have devoted significant attention to the effects from PRP reforms. While being outside the scope of this study, it can be mentioned that this literature highlights the complexity of implementing such schemes and outcomes may be contingent on historical and cultural factors (Schay and Fisher 2013). Overviews describe PRP programs being implemented “often with mixed or downright disappointing results” (Pollit and Bouckaert 2011, 92). Perry et al. (2009) review 57 studies that evaluate PRP schemes in (US) government during the years 1977–2008 and collect several lessons regarding their implementation. Evaluating the performance of employees is a sensitive process. Also, in societies with well-functioning government institutions—such as the USA—employees perceive that factors unrelated to merits, such as “office politics,” influence performance ratings in PRP schemes (Kellough and Nigro 2002, 163).

Anticorruption reforms have targeted the incentives of public officials by linking rewards to their performance. The idea is to make it rational bureaucrats to refrain from corrupt acts by rewarding honest behavior: “Indeed, sufficiently high rewards would eliminate all incentives to accept bribes” (Polinsky and Shavell 2000, 3). This notion is similar to the thought that points to the importance of the fixed salary paid to public servants. Accordingly, the size of the salary, in relation to the gains from bribery and the risks of losing benefits from the employment if one gets caught, is a strong determinant for bureaucrats’ incentives to engage in bribery (Becker and Stigler 1974). Authors have therefore suggested that competitive salaries would “[reduce] the relative attractiveness of the quick returns available from corrupt individual practices” (Evans and Rauch 1999, 752). While widely investigated, it can be noted that the link between salaries of civil servants and corruption across countries is not uniformly supported (Van Rijckeghem and Weder 2001; Dahlström et al. 2012).³

The incentive-focused reasoning likewise underpins PRP reforms: “pay for performance draws on standard economics, particularly the economic principal-agent view” (Weibel et al. 2010, 388). This is also the thinking behind the suggestion that PRP programs may reduce bureaucratic corruption (Elliott 1997). In line with the reasoning, “a powerful and determined outside monitor (principal) can reduce corruption and improve delivery of services by his bureaucratic agents” (Rauch and Evans 2000, 51), and accordingly, there are different “strategies a principal can use to elicit better performance from his agents, such as performance-based pay” (ibid, 51). Influential authors have argued that rather than focusing on the fixed part of the salary, it is the varying part that activates employees’ incentives and therefore should curb corruption: “a more effective strategy is to change the specific rewards that accrue to specific actions” (Klitgaard 1988, 77). Ades and Di Tella (1997) argue that corruption may be reduced through “providing the right incentives to bureaucrats” (504), such as PRP reforms. However, the authors note that measureable performance indicators for reduced corruption are difficult to construct. Similarly, Rose-Ackerman (1999) posits that “individualized incentive pay systems can be a tool to break up circles of corruption” (80). This reasoning is further developed by Skladany (2009), arguing that PRP programs targeting bureaucrats in developing countries would “substitute for the money that corrupt public servants usually steal from the government” (7). This reasoning is again underpinned by principal-agent theory: “to motivate an agent to not shirk [i.e. avoid duties], the principal can also offer inducements or incentives to the agent that are linked to her performance. … These performance-based incentives are meant to reinforce the existing incentive that an agent has to act in the best interests of the principal—keeping her job” (23). While acknowledging the risk of capture of such reforms in developing countries, the article defends the merit of PRP reforms in such countries and argues that these programs are quite difficult to manipulate. It should be noted that the article defends a PRP scheme where “outsiders” evaluate bureaucrats’ performance.

Although such reforms do not lack proponents, few studies have examined the PRP-corruption relationship empirically. Some scholars have investigated the effect of PRP reforms on outcomes that are close to, or may be said to be indirect measures of, bureaucrats’ corrupt behavior. An observational study found that an incentive program in Brazil increased tax collection (an indication of decreased bribery among collectors) (Kahn et al. 2001). Relatedly, Duflo et al. (2012) studied a randomized implementation of an incentive program in Indian schools. They found that when teachers are paid according to their attendance, absenteeism—a misuse of public funds that benefits individual officials—among teachers was reduced. Notably, this scheme was bolstered by a mechanism of monitoring, where pupils were instructed to take a photograph of the teacher and the other students at the start and end of each school day to document actual rates of absenteeism.⁴ Kwon (2014) uses a survey to investigate corruption perceptions among South Korean bureaucrats. The results suggest that officials who believe that their promotion is contingent on their performance are more likely to state that taking substantial gifts from contractors are considered corruption. Only one study has tested claims of the impact from PRP reforms on corruption empirically in a comparative perspective. Dahlström and Lapuente (2012) examine a measure of the degree of PRP in public administrations across a large number of countries. When analyzing a multivariate model with controls for rivaling explanations, they do not find any association between the degree of PRP and corruption levels. Illustrative, a review concludes: “It is striking that few studies directly examine the impact of improved incentives [among bureaucrats] on corruption outcomes” (Olken and Pande 2012, 501).

There are examples suggesting that the success of administrative reforms is partially dependent on institutional quality in the context where they are implemented. Scholars have argued that performance-based management practices for public organizations (i.e., not individuals) may be less suitable in transitioning countries (Anderson et al. 2003).⁵ Similarly, Sundell (2014) proposes that the suitable type of administrative reforms—focusing on meritocratic recruitment—may depend on the context of patronage risk. The study argues that when patronage risks are high, the amount of discretion in recruitment needs to be low in order to prevent abuse. Therefore, while flexible recruitment may be the suitable in countries with low patronage, formal exams are the better option when patronage is widespread. There are hence reasons to suspect that the difference in levels of corruption is crucial to understand also why PRP reforms may be captured.

When corruption is widespread, conventional policy thinking tends to meet resistance because authorities cannot simply use exogenous policies to improve the situation (Andvig and Fjeldstad 2008). In a context of widespread bribery, it is said that corruption resembles a collective action problem where it becomes individually rational to engage in such behavior. If corruption is endemic, the bureaucrat not taking bribes becomes the “sucker” and actors will therefore tend to act corruptly since most other colleagues are perceived as doing it as well (Rothstein 2011). In such circumstances, the underlying foundations behind PRP schemes—principal-agent theory—are altered substantially.

This reasoning has been used to understand delegation in a bureaucracy where senior managers are the principals and bureaucrats are modeled as the agent. Yet, this thinking has been criticized when applied to situations of high corruption: “It assumes away the problem, especially in [a low developing country] context, because the political will to engage in vigorous monitoring and implement appropriate strategies is lacking, or worse yet the principal is himself corrupt” (Rauch and Evans 2000, 529). The postulation of an honest principal—for instance, a senior manager—is not necessarily true under widespread corruption. And if this assumption is no longer valid, it has consequences: “If the supposed principal is also corrupt and does not act in the interest of the public good, the principal–agent framework becomes useless as an analytical tool since there will simply be no actors willing to monitor and punish corrupt behavior” (Persson et al. 2013, 542). Moreover, they argue that this may lead to a breakdown of any anticorruption reform that builds on the principal-agent framework.

With the insights in mind that such principals may also become corrupted, the reasoning underpinning PRP reform may be questionable in a context of widespread bribery. In particular, such reforms have certain inherent features that possibly make them particularly ill-suited for corrupt contexts. The evaluation of performance is more complicated than the process of raising the fixed part of the salary to all employees (Dahlström and Lapuente 2010, 581). Moreover, a recent report from the OECD states: “Performance-based pay requires clear rules and transparency to avoid abuses. In [transition] countries, there are no clear and transparent criteria for the allocation of the variable part. This part provides a very broad, sometimes full, discretion to heads of public institutions in allocating bonuses and other additional payments to his or her subordinates” (OECD 2014, 110).⁶ This point is important as the evaluation part of PRP programs inherently results in managers having discretion and room to maneuver.⁷ Intuitively, large amounts of discretion in a high-corruption context raise the concern that opportunities for misbehavior may be created.

Following this notion, it therefore seems that rather convincing reasons exist to believe that (1) the line of thinking that underpins PRP schemes may be less suitable when imposed in corrupt contexts and (2) that such schemes may become captured due to their large share of managerial discretion. However, current research does not provide a detailed understanding of such processes. The purpose of this article is hence to explore the potential of PRP reforms to curb incentives for bribery among civil servants in a context of high corruption. The article investigates this issue empirically through an in-depth study of reforms in the South African civil service.

One of South Africa’s challenges is corruption. While the country places in the middle tier of global rankings (number 64 of 176 in Transparency International’s 2016 classification), corruption is definitely prevalent. Although officials are legally prohibited to “favor relatives and friends in work-related activities,” or to take “decisions which may result in improper personal gain,” (Republic of South Africa 1999), such behavior is not eradicated. In fact, 47% of South Africans say they have paid a bribe in the last year to secure an essential service—far higher than the global average of 27% (Corruption Watch 2013).

South Africa has witnessed numerous anticorruption programs. To reduce officials’ propensity to take bribes, programs of PRP were installed in the 1990s. When the Public Service Regulation was enacted in 1999, one of its features was to link rewards to performance (Public Service Commission 2007, 2). The legislation spells out that “an employee shall be eligible for salary range progression only if … she or he has received consistently satisfactory performance assessment” (Republic of South Africa 1999). Van der Vyver and Bussin (2013) describes the South African PRP system as having three actors in this assessment: the employee, a senior supervisor and a review committee: “[This committee] receives all provisional ratings of employees, to review, compare and validate ratings. Managers and supervisors should be asked to explain and defend ratings that do not appear to be justified” (16). This process may result in rewards for employees on a biannual basis.⁸

The DAFF administration has for decades had challenges of corruption among its Compliance Directorate staff (Hauck and Kroese 2006). This directorate is responsible for ensuring that fishermen abide by regulations and its staff enforces the legislation from local substations along the coast. In practice, this means that such officials control that fishermen have not violated regulations, such as fishing in protected areas or without a license. Graycar (2015) notes the importance of defining the type, activities, sector, and place of corruption, to understand the nuances of this behavior in a certain locality. In this particular context of South African fisheries, the opportunities for bribery occur when inspectors control fishermen, often at local harbors. A vessel or a smaller boat is inspected when entering the shore at designated landing sites and declares its catches to control officers. To illustrate how common such inspections are, there was over 5000 reported inspections in the year 2015 (DAFF 2016). Yet, in practice, these controls do not work as efficiently as intended. Inspectors are known to establish relationships with local fishermen where they receive bribes and, in return, benefit these fishermen by granting noncompliance to rules, i.e., allow them to catch more than their quota, to take juvenile lobsters, or other violations which serve as a source of additional income (Raemaekers et al. 2011; Gallardo Fernández and Hauck 2013). In detail, such bribes may consist of payments of money at the harbor, during the inspection, but could often be the practice of delivering some fish or a box of lobsters to an inspector’s house (Sundström 2013, 2016). Corrupt inspectors also help fishermen who give such bribes by engaging in illegal poaching activities themselves (Hauck and Sweijd 1999) or by lending government vehicles to transport illegal catches unnoticed (Raemaekers 2009). Another example of how these bribes may function are situations where an inspector actually hands out a fine to a rule violating fisherman, but then arranges so that this document is “lost” by clerks at the local magistrate’s office before the fine enters the judicial process (Sundström 2015). Thus, pertaining to the typology of corruption outlined by Graycar (2015), the type of small-scale corruption often found here is collusive and these transactions also take nonmonetary shapes.

In 2001, bribery in this authority got media attention through a high-level court case that sentenced a large fishing company for poaching and having inspectors on a pay roll. As a result, numerous reforms were instated during the early 2000s to reduce corruption. The department increased salaries for enforcement officers and introduced randomized “lifestyle audits” to monitor illegal payments, in which expenses and assets of an official are reviewed in relation to incomes. They started working with informants and installed a telephone hotline for reporting corruption (Hauck and Kroese 2006, 79). Importantly, a PRP scheme was introduced. DAFF now grants rewards to “encourage good performance” (DAFF 2016, 132) aimed at increasing inspectors’ propensity to report violations and thus become less inclined to take bribes. The assumption is that inspectors commonly taking bribes to refrain from report violations tend to have a poor record of fined poachers and confiscated goods. Conversely, inspectors refusing such bribes will have more to show in terms of enforcement actions. This reward is generally 10% of the salary and paid biannualy. While figures on these payments are hard to come by, for instance, to whom exactly they are paid, the annual report of DAFF for the budget year 2015–2016 describes that 23.1% of employees qualified for performance rewards (DAFF 2016).

Choosing the case of the Compliance Directorate of DAFF as the focus of this study has several benefits. The administration has documented problems of bureaucratic corruption and the PRP reform has been introduced with the ambition to reduce bribe taking. Moreover, the presence of substation managers in each station along the coast offers potential insights into the assessment process of the PRP bonuses.

The study reports insights gained from a data collection where the author conducted 43 interviews in early 2014. In total, 34 respondents were inspectors at substations along the western and southern coasts. Moreover, nine carefully selected key informants were interviewed: four former senior managers of DAFF, two leaders of fishermen associations, two journalists, and a scholarly expert. Through local contacts, the author was introduced to senior managers holding high-ranking administrative positions in the organization. These actors ensured that a message was sent out to the staff employed at substations, giving the inspectors permission to talk with the author on the problems facing their enforcement. Without revealing their names to management, the author then visited such inspectors. Therefore, their identities were not disclosed to higher-ranking officers.

The authority employs about 300 fishery control officers, distributed across 22 stations within the two regions Northern Cape and Western Cape, stretching the western and southern coastline (DAFF 2014). Yet, information about individual features of these officers, such as gender, race, or the number of years employed, is not publically available in a systematic overview (personal communication with responsible authority). Care was taken to visit stations in varying contexts along the coast. Likewise, the selection of the 34 inspectors was conducted purposively to maximize the variation of individual features thought to affect their attitudes to corruption. Also, recently retired inspectors were sampled as they face fewer risks when speaking about these issues. To hide the identities of respondents, this section refrains from reporting individual combinations of such features. Respondents’ years of experience vary widely—from recent employment to 39 working years (the average was approximately 12 years). Five respondents were women, although they are quite scarce in the general pool of inspectors. Most of the persons interviewed are black or colored, nine persons were white.⁹ Seven persons were senior managers. Finally, the respondents are not very well educated: two had a bachelor’s degree. The perception is that, besides the high amount of women, the sample is roughly representative of the general pool of inspectors.

When asked to participate, the respondents were told that sensitive issues were to be discussed; yet, these talks would be confidential. On no occasion did a respondent opt out from such a meeting. Interviews lasted between 30 and 60 min and were conducted in English, the first language for half of the group of respondents. Therefore, no interpreter was present. To maintain confidentiality, no recording device was used. Instead, great care was spent in taking notes of these talks, transcribed within the following hours. Interviews took place in settings of respondents’ choice and with no third party present.

The interviews first touched upon the tasks of inspectors and when describing the practice of bribery, follow-up questions explored such behavior. The talk then became freer, using open-ended questions to explore how the situation of bribery could be improved. At such occasions, the respondents often brought up performance bonuses. Since the sampling strategy was purposive, the interpretation of this material sought to find themes and illustrations of insights rather than to analyze answers with a statistical approach suitable for probabilistic sampling (Bennett and Elman 2006).

Conducting interviews on controversial issues, the researcher may in some occasions benefit from taking the ambivalent position of being an “outsider” (Eriksson and Stern 2009). In this context, the perception is that the position of the author—neither being a government employee nor affiliated to a political party—allowed respondents to air discontent without risks of repercussions. Respondents would often, without being asked, describe how they had received boxes of fish to be blind to rule violations.

A key feature in the respondents’ accounts is that a senior manager (often referred to as “the supervisor”) runs each substation. These managers are generally thought to be the most involved in established networks in the coastal communities and are perceived as being more corrupt than subordinates and newcomers:

When I came here the supervisor had worked here for 20–30 years. He lived by the harbor. He didn’t perform well. He knew people. And they knew him well. I noted that when I started you want to perform. But the supervisor said, “You should only have given this guy [a violator] a warning.” (I:4)

My supervisor fought for this inspector doing this [taking bribes]. And I told management what I was witnessing. But no one did anything. … So there is not really anything you can do. Each station is run by the substation manager. They are the bosses of their own station. (I:5)

Several descriptions suggest that having a manager that is involved in corrupt practices will affect the work of the subordinates. A previous general director of the department explains this reasoning:

Officers know that their bosses are corrupt. … They know there are no consequences. If you’re trying to do anything right you will lose your job. And then it is easier to put your head under the bed cover and pretend you did not see it. Or, better, be in it yourself. … The implication of having corrupt bosses at a station is that you know you can get away with anything. Because you will actually be able to. (K:5)

Regarding the PRP system—most often called the “merit-system” or the “merit-bonus” by respondents—the accounts describe in different ways how the introduction of this system was implemented, with it generally perceived as being a distorted outcome. A widespread belief seems to be that these bonuses are not based on professional accomplishments. Rather, they are seen as rewards from managers that are given on contrastingly different grounds. A general notion is that the third party in this assessment process, the review committee, seldom has a real say in which employees will receive bonuses. Instead, this decision is taken earlier in the process by the senior manager. This manager evaluates the performance of a subordinate before the assessment is handed to the committee for verification. This process often contains discretion and constitutes possibilities for the manager to influence who is (not) rewarded. Importantly, a fundamental aspect of this decision seems to be that personal relationships influence which inspector gets this reward:

Even when you meet your targets you never get a bonus. And a colleague who didn’t meet his target got it. … So the previous managers used favoritism. Then, the same person would get the merit every year and year. They were buddies. (I:17)

There are only a few persons who get it [the biannual merit bonus]. So I ask myself, “Why is this so?” Something like a favorite of top managers. (I:20)

The accounts highlight whom these managers reward. If managers themselves are involved in corruption, they will not benefit from rewarding non-corrupt subordinates. Rather, they will seek to assist bureaucrats that collude in corrupt practices. The rationale for such behavior is that managers involved in bribery will have an incentive to reward colleagues that (1) are interested to continue the practice of payments from citizens and (2) will not blow the whistle to outsiders of such practices. Therefore, the personal relations previously described, in turn, have corrupt connotations as they are used selectively to benefit inspectors that collude with managers:

The wrong person gets the merit. It is on an agreement with the supervisor. So he must like you, agree [with your behavior] when he represents you. Otherwise he will talk badly about you [in front of the review committee]. Now, the system is not based on how much compliance you have ensured [i.e. reported violations]. (I:12)

[Inspectors at my station] were taken for misbehavior in this sense [bribery]. Their salary was held in for two or three months for disciplinary causes. But at the same time, at the end of the same year, they received a merit bonus in their salary. This is ten percent extra to your salary. So it sends the signal that crooked behavior is in this sense rewarded. (I:1)

[Managers] base an assessment on favors. Some people are underestimated, others are overestimated. They want to promote people who always say “yes, yes.” But these persons are not hard on poaching. Many of them even had disciplinary cases against them, for corrupt behavior. But these just disappear. And the persons assessing these corrupt guys, of course they get something back. (I:29)

These bonuses may also turn into an exchange between inspectors and managers. Accounts illustrate three different types of exchanges that are somewhat different. The first type of exchange is a situation of mutual collusion where the inspector and supervisor are involved in bribery and use bonuses to reinforce this relationship. Accordingly, the actors will then have an incentive not to speak openly about this practice:

If the station’s supervisor is corrupt, then why wouldn’t you yourself be? And no one will report. Because then they will report on each other. So people are quiet. This is how so many substations work along the coast. And the merit system actually boosts this. (I:13)

A second and different situation is the exchange where officers involved in corruption can “buy” such rewards from the senior manager. In turn, the manager gives positive performance reviews after receiving a gift from the inspector:

The supervisor will say, “Give me a bucket of fish and I will make sure you get the bonus.” (I:13)

Third, this exchange can also be a path for inspectors to blackmail corrupt senior managers. In such a situation, it is an inspector that uses information about a manager’s involvement in corruption to ensure that the performance is positive and the bonus rewarded:

It’s demoralizing. A merit is a large sum of money. So there are people getting the merit year after year. You don’t understand why they got it. … Then you realize it. ... The person getting the merit had information of bribery that the supervisor had been involved in. He used this as leverage. I understood that this person had something on the supervisor. He would spill the beans about this corrupt supervisor. So he used this as a way to get merits, a higher salary. (I:6)

These corrupt relationships also determine who is not given such rewards. Contrary to the blind and corrupt inspectors who benefit from this practice, bureaucrats who do not take bribes become isolated and perceive that this has economic and professional costs:

Some people get boosted, others don’t. And if the supervisor does not like you, then you are stuck. And you know how you have to act to become a favorite: you have to play along. And the guys getting the bonus are the guys doing nothing, looking to the other side. (I:21)

You have to be “one of the boys” to get a merit bonus. Like my supervisor, he will never make me get a bonus. He does not like [inspectors who catch violators]. And he is not an honest inspector. So I am on my own. (I:27)

It is evident that the presence of colleagues involved in this practice—taking bribes and being blind to violations—seems to create frustration among honest inspectors:

If someone does not like you, the supervisor or someone in the reviewing panel, then you won’t get it. And this opens up for giving and taking. So if you don’t have a good relation [did not bribe the manager] they will just say, “This was all part of your job.” So it is demotivating in fact. (I:18)

[Your supervisor] decides who gets a merit. And then you realize it is not conviction rates that give merit. It’s favoritism. … Influence decides who gets merits. So this system of merits becomes the real demotivator. It makes it apparent that it is not the number of cases that matter for promotion. (I:5)

Following the reasoning that honest behavior is associated with an economic cost, this reward structure becomes a disincentive for inspectors to apply themselves and refrain from bribery. Furthermore, this seems to increase shirking behavior, the tendency to avoid performing duties. Respondents describe how having colleagues that are engaged in corruption, yet receive bonuses, has an impact on their own willingness to enforce regulations:

We [inspectors at the substation] have a very poor record of arrests. I had three [colleagues] that did not catch a single guy for two–three years. But it did not cost them anything! … The feeling is that “why should I catch someone? This will not give me merits.” (I:1)

Instead [the merit bonus] will make people to work less. We will think, “Why should I work if I’m not rewarded.” (I:20)

Therefore, the accounts suggest that these selective returns may lead to (1) honest bureaucrats finding that their performance is not rewarded and (2) inspectors therefore considering engaging in behavior that is more beneficial. In this context, collusion with corrupt colleagues is clearly more advantageous.

The findings of this article do not necessarily imply that PRP reforms meant to reduce bribery will always fail in corrupt contexts. However, designing them in such settings is a great challenge. In criticizing overly optimistic statements of the prospects of principals controlling their subordinates, it has been said that “In an escalating cycle of agents overseeing agents, we must ask: Who monitors the monitors?” (Shapiro 2005, 280). This highlights a feature that was key in distorting the PRP reform investigated here, that “monitors” themselves, the managers, are perceived to be involved in the practices the reforms are supposed to curb. Yet, in search of designing successful PRP schemes to counter corruption, monitoring should be explored further. We have seen—in the study by Duflo et al. (2012), focusing on teacher absenteeism in Indian schools—that there may be examples of incentive programs targeting corruption that can be implemented in high-corruption contexts. In that study, key for success was that the program designed monitoring—where pupils documented teachers’ presence with cameras—that made tampering with the performance measure difficult. Designing programs that have agents with little incentive to collude with officials that monitor performance is of course a challenging task. Nonetheless, a suggestion for policy may possibly be that a crucial condition for the design of PRP schemes targeting bribery incentives in high-corruption contexts is that external agents that have little incentive to collude with the bureaucrats whose performance is being evaluated should be responsible for monitoring. Herein lies a challenge, as evaluating performance is foremost a task suitable for peers in an administration. Yet, if such schemes should be used, insights in this regard need to be advanced.

Moreover, it is sometimes difficult to identify the agents that have little incentive to collude with officials. Citizens have clear incentives to stop non-collusive corruption, that is, being asked for bribes to obtain services that they are entitled to. However, their incentives are not equally clear regarding collusive corruption, actions where law-violating citizens pay officials to let legal infringements go unreported. In fact, many individuals will benefit by this possibility, and it is harder to identify who may function as monitoring agents for this kind of corruption. Following this reasoning, it is predicted that monitoring of PRP schemes that seek to reduce corruption incentives is more likely to succeed when targeting non-collusive corruption, than compared to collusive corruption. Therefore, it may be better to use PRP schemes in administrations where clerks ask for side payments to do their duties than to use these schemes in bureaucracies of law enforcement where officers take bribes to let violations go unreported. Yet, since non-collusive corruption was not studied in this article, this proposition should be further studied in future studies.¹²

There are several aspects of the installation of PRP systems that are not analyzed in this article which could be fruitful themes for further research. First, regarding the specific case of the South African fisheries sector, research has not yet investigated how other stakeholders—such as the affected fishermen—perceive the rewards to corrupt officers and the marginalization of non-corrupt inspectors. Research that builds on interviews with fishermen in this sector suggests that they view inspectors as dishonest when accepting bribes, yet tend to uphold corrupt agreements since they benefit in the short run from this behavior (Sundström 2013). Future studies could build on this work by exploring perceptions of inspectors’ salary incentives also among fishermen. Second, the installment of the PRP system studied in this article most likely altered the relations between inspectors, by increasing the general willingness to engage in corrupt activities and punishing those who refrained from this behavior through the pay check. If future research were to study such a reform with a before-and-after perspective, additional insights could be gained in regards to how this process affected relations between officers. Finally, a majority of the inspectors interviewed for the purpose of this article stated that they took the job mainly because other positions were scarce. Similarly, it seems that very few of the inspectors in this authority are formally trained in the field of marine resource conservation. One respondent, who is exceptional since he had a university degree in conservation, viewed himself as alienated to the corrupt colleagues and suggested that the lack of suitable education among these officers may be a factor resulting in a collective with little motivation to carry out their tasks. No matter if education is the main reason for this outcome, the DAFF inspectors seem to lack the esprit de corps often said to be characteristic of an administration free of corruption (Dahlström 2015). Future research would benefit from investigating if measures to improve a collective motivation to protect marine resources, such as educational interventions, may reduce willingness to take bribes.