Over the past two decades, an international anti-corruption movement has emerged with the shared understanding that malfeasance has negative implications for governance and development. According to the United States Agency for International Development (USAID), there is “an emerging global consensus that fighting corruption and building good governance are essential for the development of people, markets, and nations” (USAID 2005: 1). This connection has also been recognized and adopted by the United Nations Development Programme (UNDP), specifically by its Global Thematic Programme on Anti-Corruption for Development Effectiveness (PACDE) (2008–2013) and its Global Anti-Corruption Initiative (GAIN) (2014–2017) (UNDP 2014). As a result, most countries in the world maintain explicit anti-corruption laws and regulations. Today, the fight against public malfeasance is as ubiquitous as that against poverty and other societal ills.

However, the outcomes of these efforts have not only lagged behind the investments made, but are, in many cases, almost nonexistent. This situation does not necessarily challenge our current understanding of the problem, but rather reflects the challenges of policy implementation. While corruption remains pervasive in the developing world, political rhetoric commonly hijacks anti-corruption discourse, resulting in its instrumentalization and exploitation for the benefit of national leaders. Thus, instead of leading to economic and social development, the anti-corruption movement has been assimilated into the regular political cycles of nations. The result is that reform is often discussed but scarcely carried out. When it does take place, interventions are usually piecemeal and followed by counter-reform efforts, a lack of implementation, and/or sabotage.

This article provides a logical and empirical reference for the reasons why national governments engage in anti-corruption efforts and the ways in which the anti-corruption discourse enters national politics, becomes assimilated into the regular political cycle, and eventually devolves to suit the status quo. To fulfil this objective, the article is structured as follows: the first section reviews the literature on anti-corruption and its depiction of national political incentives, proposing in response a model centered on political capital and corruption profits as crucial elements for empirical analysis. The final sections review and explain the Peruvian experience through one specific type of anti-corruption effort, the National Anti-Corruption Commission (known as CNA, its Spanish acronym), and present conclusions.

Incentives Behind Anti-Corruption Efforts

Literature Review

Though it is a relatively young field of study, corruption is the focus of an important body of literature that examines its core elements, from which the discussion of anti-corruption extends to a consideration of actors, strategies, processes, and problems. Traditionally speaking, the normative approach to anti-corruption studies is best exemplified by the principal-agent model, with the works of Robert Klitgaard (1988), Jeremy Pope (1999), Daniel Kaufmann (1997), and Rose-Ackerman (1998) being among the most prominent. Succinctly put, this model sees corruption as being due to the limited information and actions available to leaders to control the behavior of public officials, thus resulting in abuses of the public trust. Presented in such a way, the model allows for three combinations of principals and agents depending on the structure of power favored: the top-down approach, represented by political leaders being in charge of overseeing state bureaucrats (Doig 1995; Khan 2006; Aron 2007; Man 2009); the bottom-up approach, represented by civil society and their monitoring of public officials (Kisubi 1999; Brunetti and Weder 2003; Kpundeh 2005; Shelley 2005); and the international anti-corruption movement, which, as its name suggests, focuses on the monitoring role and capabilities of foreign agents in fighting local corruption (Martin 1999; Marong 2002; Carr 2007; Wouters et al. 2012). Each of these approaches has been crucially important in creating potential avenues to curb corruption, and, thus, a vast array of normative arrangements is now available from which to select the most appropriate (UNODC 2003, 2004a, 2004b, 2005, 2011; OECD et al. 2013; Transparency International 2014; Richard Holloway 2011).

The bottom-up approach to the principal-agent model would seem particularly promising as society, being the disadvantaged party in a corruption transaction (Spengler 2010; Karklins 2005), should have an inherent motivation to fight malfeasance in the public sphere. However, the argument continues working under the paradigm that there is an actor in the domestic system willing and able to take the role of the principal. Such a premise gives way to the problem of collective action. Bo Rothstein has given much attention to this issue (2005, 2011). He describes this position as questioning “the underlying assumption [in the principal-agent theory] that all societies hold at least one group of actors willing to act like ‘principals’ and, as such, enforce such regimes” (Persson et al. 2013: 8). The collective action problem posits that, in societies ravaged by systemic corruption or simply in those were the issue of corruption does not stay restricted to the higher levels of government but can be found in everyday life (ubiquitous petty corruption), there may not be any actor willing to take the role of the principal, as it is always more profitable to partake in corruption rather than spend private resources to fight it (Del Castillo and Guerrero 2003; Karklins 2005; Uslaner 2008). It becomes the common formula: ‘if everyone is corrupt, then nobody is.’ Consequently, as high levels of corruption allow it to become self-sustaining through its impact on public perceptions, principal-agent strategies fail to gain traction, growing irrelevant or, even worse, detrimental. The real problem then emerges: who is to lead the change when the argument is based on the absence of willing actors? In response, collective action tends to propose solutions that invariably require authoritative leadership, thus falling prey to circular logic and demonstrating an inherent weakness in its range of applications.

Reacting to similar issues in other areas of research, the Thinking and Working Politically (TWP) approach has recently coalesced to openly point to the crucial role that local political contexts and actors play, and at the need to address them at every stage of policy formulation and implementation. In the case of anti-corruption efforts, the logic of TWP can be found in Hassid and Brass (2015), who discuss the difference between China and Kenya – with respect to their handling of corruption cases – by referring to the levels of pressure and danger each government experienced. Bueno de Mesquita and Smith (2011: 252) make a similar point by pointing out that if the individual harmed by change is the ruler or the CEO, the same person who has to initiate the change, then we can be confident that change is not going to happen. However, in the 1980s, this argument was already being offered by Chubb and Vannicelli (1988: 145), who, in discussing the handling of political scandals in Italy, realized that the paradox lies in the fact that the initiative for the reform of public institutions and the moralization of public life must come from precisely those political actors who are among the prime causes of the degeneration. The TWP builds on this reasoning to address corruption-related issues from a mainly political perspective, seemingly acknowledging the basic conundrum behind the topic: because anti-corruption measures usually target the actors responsible for their implementation, political will cannot be presumed to exist in any single case.

Theoretical Discussion

In the words of Sahr J. Kpundeh, the study of political will for anti-corruption reform brings into the scholarly discussion the role of “the actors, their motives and the choices they make to promote and implement anti-corruption reforms” (1998: 92). As government activities are never free, the simple idea of performing an action intended to fight corruption requires us to consider the inherent costs of that action as a starting point. For example, Klitgaard (1988: 195) highlights the tradeoff in terms of efficiency by noting that the level of anticorruption effort will be short of the maximum and the optimal level of corruption will not, in practice, be zero. This type of cost, however, is not the only (or even the most important) variable when it comes to real-life politics. If the leadership is engaged in illegal acts, the anti-corruption drive will not only stop short of its maximum level, but much earlier, exactly at the point where future corruption profits are potentially threatened.

On the other hand, anti-corruption efforts, just as any other government activity, do not only translate into costs (financial or otherwise) but, because they positively impact society, also create benefits for the government in the form of political capital, a crucial type of resource that describes the amount of support received from international and societal groups in response to specific circumstances. With this support (which can take such forms as citizen compliance, popular approval, public demonstrations of endorsement, political and financial backing, electoral preference, and many others), authorities can exert control over the political system and carry out their duties without relying solely on coercion.

Although concerns about political capital are maintained with respect to most government activities, anti-corruption initiatives present a peculiar situation as they specifically target governmental actors, including national leaders. Going back to the case of a leadership engaged in corruption, the situation quickly takes a turn for the worse when we consider that political capital can be completely forsaken in favor of larger rewards in the form of illegal proceeds. In any case, the relative weight of political capital as compared to illegal income depends on the subjective preferences of the political actors and when the latter are prioritized, we should expect anti-corruption reform to completely stagnate.

What this line of reasoning represents for the pursuit of anti-corruption efforts can be summarized as follows: For an honest government, anti-corruption policies should be attractive only in direct relation to the political capital they can generate; for a corrupt government, anti-corruption policies should be avoided in direct relation to the interests threatened. As the preferences and interests of the leadership are many and change over time, we can (and should) expect government initiatives to be highly reactive to political circumstances. In other words, anti-corruption initiatives may exhibit their own political cycle based on their potential to garner political capital and threaten corrupt interests.

Characterization and evidence of the type of institutional behavior stated above can be found in the body of literature specifically focused on anti-corruption agencies (ACAs). De Sousa (2010: 13) appropriately posits that they are vulnerable to be used as “window dressing institutions or at best they function as a governmental anti-corruption discourse mechanism,” and that consequently they “can bring considerable symbolic gains to the incumbent” (De Sousa 2010: 17). This position, of course, shares the centrality of political capital as the argument offered by authors dealing with political cleanups, who agree that the impetus to address corruption can be provided by political exigency, rather than by real interest in the efficient functioning of the political and economic institutions (Mbaku 1996: 108). Indeed, discussing China’s approach to fighting corruption, Cheung (2007: 66) concluded that “so long as there is the perceived threat to the leadership’s political security (i.e. a corruption-induced crisis), there should be a correspondingly strong political will to take mitigating or remedial action.”

But just as political capital becomes a strong incentive to create ACAs, on the opposite side of the alley, political leaders shun away from them out of fear of seeing themselves exposed to prosecution. These seemingly dissonant reactions to ACAs are exemplified by the governmental attacks on Slovenia’s Commission for the Prevention of Corruption (Batory 2010), or in the treatment received by anti-corruption initiatives in Georgia before the Rose Revolution (Kupatadze 2012). The ebb and flow of this cycle of political will for anti-corruption efforts is, thus, summarized by Maor (2004: 22–23):

‘Given that the origin of [special anti-corruption prosecutors] comes from profound public disquiet over the integrity of senior political executives, that corruption scandals are mass media events, and that there is a widespread cynicism in the general public regarding the ethical standards of politicians… political executives are forced to act boldly in order to root out corruption. Consequently, they often resort to overkill, that is, they set up mechanisms with a level of independence higher and a range of powers larger than they are willing to tolerate. Later when investigations come close to home, they retract and try to undermine the credibility of anticorruption mechanisms, and when deemed necessary, to terminate their operation.’

The Political Cycle of Fighting Corruption

Following the preceding arguments, the political cycle of an anti-corruption initiative is proposed here to have five stages, as depicted in Figure 1 below.

Figure 1 

Political Cycle of an Anti-Corruption Initiative.

The process begins when internal and/or external actors express a demand for the improvement of anti-corruption systems in the country, with the offer of providing political capital in return. Although this might be considered as an overall negative scenario, the situation represents an opportunity for new political actors to strengthen their position. Particularly when leaders need political capital, the administration will proceed to adopt an anti-corruption initiative, which may or may not be designed to realistically reform the system, but which nonetheless will prove successful in stimulating support. This ‘payment’ for having heeded demands represents the peak and central point of the cycle, when both sides of the issue feel satisfied with the outcome of their exchange.

Content with the apparent resolution, internal and/or external actors then turn their attention away from anti-corruption demands and towards more salient issues, consequently making the political value of further anti-corruption efforts decline. When this decline is joined by the involvement of members of the administration in corrupt activities, or when anti-corruption becomes an active threat to the stability of the leadership, the complete exhaustion of the previously adopted initiative will follow the expected decline in political will. As a result, and depending on the degree of government involvement in illicit affairs, the lack of incentives and rise of disincentives may even open the door to the devolution of anti-corruption structures, which will eventually make the problem of corruption salient again and restart the whole cycle.

To test the above model, the next section adopts a single case study approach to guarantee scientific rigor in the identification of elite interests from policy decisions. This approach is selected for its potential to provide a detailed and highly contextualized analysis of events that would be otherwise unfeasible (Lewis 2003). Furthermore, the focus on a single case will allow for the employment of a process-tracing strategy, which offers the possibility to assess not only the main cause-effect relationship of interest – political conditions and public policy – but also the intervening mechanisms that connect them both. As explained by George and Bennett, “(t)he process-tracing method attempts to identify the intervening causal process—the causal chain and causal mechanism—between an independent variable (or variables) and the outcome of the dependent variable” (George & Bennett 2005: 206). In other words, process tracing is an analytic tool for drawing descriptive and causal inferences from diagnostic pieces of evidence often as part of a temporal sequence of events or phenomena (Collier 2011: 824). Therefore, by choosing a single case study approach, an attempt will be made to reproduce the chain of events as closely as possible, tracing the adoption and outcomes of policy measures back to the contextual factors that gave birth to them.

For the above reasons, the study now turns to a review and analysis of Peru’s first experiment with an ACA, the National Anti-Corruption Commission (CNA), between 2001 and 2005.

The National Commission to Fight Against Corruption

Immediate antecedents

After the collapse of President Alberto Fujimori’s government in late 2000 due to the discovery of a vast network of corruption throughout the state, the post-Fujimori transitional government, led by President Valentín Paniagua, was tasked with organizing a new electoral process, while expected at the same time to establish the basis for what would become the first national anti-corruption structure in the country.

On April 11, 2001, as one of the actions adopted by the Paniagua administration to stabilize the country and regain political support for the system, the government created the National Anti-Corruption Initiative1 (INA, its acronym in Spanish), an official forum for the discussion and proposal of anti-corruption policies that included actors from the public and private spheres. Additionally, and perhaps more importantly, the INA was established to address the extent and nature of corruption in the country and to produce a national dialogue upon which future standards could be based. Three months were given for the execution of these tasks as this was the time left before the transitional government had to transfer power to the new constitutionally-elected leadership. Nonetheless, thanks to the political circumstances, the support of international actors (and that of President Paniagua and his minister of Justice, Diego García-Sayán Larrabure), and the heterogeneity of its members, the INA not only succeeded in producing the first coherent and comprehensive set of recommendations (INA 2001), but also became the common point of reference for future initiatives on the subject.

The task given to the INA – to develop a framework for future anti-corruption actions (going beyond the transformation already underway) – was described by Minister of Justice García-Sayán as consisting in the development of guidelines for a national plan against corruption that could be adopted by the government and Peruvian society once new authorities assumed leadership in July 2001 (Zileri 2001).2 Starting at 56% in January 2001, by May of that year, 69% of people surveyed in the country supported Paniagua, with 42% expressing their approval of his staying until 2005, if needed (Los Andes 2001). By the end of the transitional government, in July, as many as 83% of Peruvians supported the president (La República 2001a).

In dealing with the severe political crisis left by his predecessor, President Paniagua adopted and implemented measures to address the spread of corruption in the country – measures that effectively gave rise to a new set of national anti-corruption standards. The work of the INA, and the new norms for transparency and access to information, created the basis for efforts focusing on corruption prevention. The fall of Fujimori had exposed the political system to unprecedented levels of public scrutiny, creating radically higher levels of corruption awareness. In addition, the international anti-corruption movement, which was crucial to the prosecution of Fujimori’s cronies and the recovery of stolen assets, had finally made its first real impression on Peruvians; from then on, foreign actors would continue increasing their level of participation.


The 2001 congressional and presidential elections saw the victory of Alejandro Toledo and his party Perú Posible. The newly elected president took office when anti-corruption efforts were highly regarded and expected by many groups, both in the country and abroad. Such groups were inclined to offer their support to a government invested in such policies, but were also ready to withdraw their support, if their demands and expectations went unfulfilled. Aware of these circumstances, the Toledo administration was quick to act and promised the appointment of a so-called ‘anti-corruption czar’ (Valenzuela 2001), a measure that was not only politically promising, but also had become increasingly necessary. In the short span of four months, the 59% of popular approval enjoyed by Toledo in late July eroded at an impressive rate, dropping to 32% in November of the same year (Tanaka 2005). Facing plummeting levels of political capital, the government was in desperate need of renewed legitimacy.

Through a decree3 published in El Peruano on October 12, 2001, the government appointed Martín Belaunde Moreyra, who was at the time dean of the Association of Lawyers of Lima, as ‘High Level Advisor of the President of the Republic for the Fight Against Corruption and the Promotion of Ethics and Transparency in Public Management’. Although the title was rather elaborate, the decree itself did not include any information about the functions that Belaunde Moreyra would perform, nor did it provide any other information. However, a day earlier President Toledo (with Prime Minister Roberto Dañino Zapata) had given a press conference to publicly present his promised anti-corruption czar and state that some of the czar’s duties would be the prevention of irregular acts and the monitoring of activities that could turn into corruption. The latter task was considered a rather delicate subject due to the interests with which it could clash. The Public Procurator’s Office, for example, was already facing ‘institutional jealousy’ (Ugaz 2014: 66) on the part of the Attorney General, who stated that Procurator José Ugaz and his team involved themselves in areas of competency that were the exclusive domain of the Attorney General. Therefore, with the alleged intent of addressing beliefs that the appointment of an anti-corruption czar would interfere with other offices, Toledo reassured members of those offices that Belaunde Moreyra’s duties would not interfere with the activities of the judiciary, the Public Ministry, the Public Procurator’s Office or the Congress, particularly in relation to the Fujimori-Montesinos case (La República 2001b).

However, despite public discourse, the government was not as interested in creating a powerful (or even relevant) anti-corruption agency, as it was in creating the impression of doing so. To acquire political capital, symbolic measures were considered as good as real actions. While this anti-corruption initiative represented a good opportunity to recover from, or at least stabilize, the loss of popular approval suffered over the previous months, a fully implemented office would clearly represent a more taxing and unnecessary endeavor. Hence, the initiative was designed with political capital – and not the fight against corruption – at its core.

To provide evidence of the above assertion, we must first discuss the circumstances under which Belaunde Moreyra was selected. According to a staff member of the CNA,4 the relationship between Belaunde Moreyra and Toledo started during the election campaign. The staff explained that Juan Paz, then political operator and later executive secretary of the CNA, arranged for Toledo to speak at the Association of Lawyers of Lima, headed at the time by Belaunde Moreyra. Many opposition politicians criticized the move as Toledo had been given a prominent political platform and people have since assumed that this meant Toledo owed a favor to the man who would go on to become the anti-corruption czar. This account closely follows the description provided by a former senior official of the Association of Lawyers of Lima.5 The official described a “direct line” between Belaunde Moreyra and Toledo because of the election support and added that Belaunde Moreyra had good relationships with other senior political leaders.

Such anecdotal evidence suggests that this may have been a patronage appointment and points to a potential lack of independence on the part of the czar. This problem was brought to light only a couple of months later. In January 2002, local newspapers6 reported that Belaunde Moreyra was avoiding the investigation of members of the government regarding allegations of corruption (La República 2002) and even acted as in their defence (Cavero 2002). Indeed, one staff member of the CNA believed that Belaunde Moreyra was trying to obstruct the work of the commissioners when issues affecting the government were brought forth.7

To appreciate the true nature of the government initiative, we must also consider the CNA design and function. This anti-corruption commission, the first of its kind at the national level, was created by Supreme Decree No. 120-2001-PCM on November 17, 2001, roughly a month after Belaunde’s appointment. It was to be presided over by a representative of the president, filled by members of the government and civil society, and included the participation of the comptroller general and the ombudsman (or their representatives). Among its functions were the development of anti-corruption plans and recommendations, the promotion of ethical behavior and transparency in the public sphere, the promotion of international cooperation, conducting periodic surveys, and other activities aimed at supporting the fight against corruption. These functions gave the CNA a leading position as the Peruvian anti-corruption body in charge of prevention. However, the most notable tasks of the CNA involved not the prevention but the control of corruption as they were also tasked to: (1) evaluate and denounce acts of corruption to the Public Prosecutor; (2) provide information about corruption involving public officials or private citizens; and (3) conduct public hearings to analyze cases of administrative corruption.

One of the key aspects regarding the creation of the CNA was its timing. On October 9 (two days before the appointment of Belaunde Moreyra as czar), Congressmember Ana Elena Townsend Diez Canseco introduced a legislative proposal to create an ‘Office Against Corruption’ with normative, technical, economic, financial, and administrative autonomy and whose chair would be chosen by Congress.8 However, four days before the ruling on Townsend’s proposal, the government created the CNA, which, according to a member of the anti-corruption commission,9 the government had decided upon only two weeks after Belaunde Moreyra was appointed czar. The sudden existence of a commission with similar duties as the one proposed by Congress member Townsend, albeit with different types of power, provided grounds for the national parliament to reject her proposal.

Furthermore, Supreme Decree No. 120-2001-PCM established that the CNA would be presided over by the representative of the president of the republic; to be precise, art. 7 of the decree reads: ‘The High Level Advisor of the President of the Republic for the Fight Against Corruption and the Promotion of Ethics and Transparency in Public Management is to be the President of the National Commission of Prevention and Fight against Corruption and Ethics and Transparency in Public Management’.10 In other words, deep-seated in the norm was the presence of Martín Belaunde Moreyra as the CNA’s president. Under these circumstances, the CNA could only be as politically independent as the anti-corruption czar himself.


There were early signs that the creation of an anti-corruption commission could be at least partially credited to the administration’s pursuit of political capital. A member of the CNA11 remembers: ‘When he [Belaunde Moreyra] was appointed they made a big circus out of it. They went all out. He went to Congress, and every day he appeared on the news’. Belaunde Moreyra also approached various opposition leaders, such as Alan García Pérez from the American Popular Revolutionary Alliance party (APRA) and Lourdes Flores Nano from National Unity (Caretas 2001), who had the second- and third-largest parties in parliament, respectively.’

When the CNA was created, it was clear for some that Toledo’s approach to the prevention of corruption was markedly different from the approach that was expected. According to Peña-Mancillas (2011), NGO Proética criticized the creation of the CNA for not addressing the suggestions produced by the INA, which were clearly stated in its final report, A Perú Without Corruption (INA 2001). The format of the anti-corruption body as recommended by the INA was closer to that proposed by Congress member Ana Elena Townsend: its chair should be appointed by a majority of parliament, and its political and administrative autonomy guaranteed. It was evident that, notwithstanding the amount of formal power originally given to the CNA, its effectiveness would be impaired if a level of independence was not maintained. However, as described earlier, instead of being empowered to pursue any incidence of alleged corruption in the government, the CNA was tightly linked to the political will of President Toledo. A member of the INA12 observed how the idea of an anti-corruption commission evolved from requiring the leadership of relevant and prestigious members of civil society who could ensure impartiality, to one controlled almost entirely by government agencies under a czar. According to the interviewee, when they observed the selection process, they predicted this would not be effective. Another member of the INA13 was even more vocal, observing that ‘in the government of Toledo, there was never a real interest in getting involved in the issue of anti-corruption’.

The situation was not improved by Belaunde’s selection of close – though ultimately critical – friends to fill some of the commissioners’ seats, particularly those crucial to the CNA’s objective: Agustín Figueroa Benza, Francisco Diez-Canseco Távara, and Enrique Obando Arbulú, three of the four representatives from civil society.14 By the middle of 2001, Figueroa quit the CNA due to his dissent15 with some of the activities Belaunde Moreyra had carried out, and was replaced by Rafael Villegas Cerro, who was also brought in by the czar. Guillermo Benavente Ercilla, representative from the Catholic Church, quit for similar reasons. In August, Carlos Morelli Zavala, representative from the Ministry of Justice, followed suit and was replaced a month later by Alberto Ygor Martínez Llanos, former congressional candidate for the list of Perú Posible (President Toledo’s party) in 2000.

Another important element underpinning the CNA’s symbolic nature was the provision (or rather absence) of material resources. When the CNA was established, there were no resources for operations; Belaunde Moreyra was not even given an office during the first months of his appointment and had to perform his duties from the building of the Association of Lawyers. According to a senior official of the association,16 Belaunde Moreyra was forced to work there due to the absence of a budget and carried out necessary duties with the help of the association’s employees. Even as he recruited staff, there was no office for them until January 10, when Belaunde left the position of dean and moved with his staff to a small office in the building of Petroperú (the state-owned petroleum company).

According to the audit report (PCM 2006), in 2002 the CNA was allocated S/. 3 million (approximately US $1 million), an amount that, as the document states, was below the amount required to fulfill its mandate. The first part of this budget – S/. 1 million (about US $330 thousand) – was not transferred until late March.17 Belaunde Moreyra himself was on a monthly contract basis,18 a fact that suggests a level of informality in the management of the agency. Furthermore, throughout the thirteen months the commissioners were appointed to their posts, and despite repeated official requests to the prime minister, the budget for allowances was never allocated, an issue which highlighted the absence of executive support.

Regardless of the formal limitations the commission faced in pursuing the fulfillment of its anti-corruption duties, it still proved effective as a core element in the government’s efforts to show its commitment to anti-corruption. A study carried out in February 2002 showed that 10% of respondents thought that the fight against corruption (independent of the prosecution of the Fujimori-Montesinos case) was being prioritized by the Toledo administration above everything else (APOYO 2002). Another study, conducted in March, showed that 46.1% of surveyed people were aware of the existence of the CNA, and that almost half of them believed that it would accomplish its mission (University of Lima 2002). Furthermore, while a study sponsored by the World Bank (2001) during the Paniagua administration had found that 85% of respondents identified corruption as the second most important obstacle in the country (outdone only by ‘unemployment’), a national survey conducted by Proética (2002) and published in November 2002 found that corruption had fallen to third place at 63%, behind ‘economic crisis’ and ‘unemployment’, and even to fourth place (behind ‘crime’) when considering spontaneous answers. This data on public perceptions indicates that the creation and political management of the CNA throughout 2002 succeeded in ameliorating the anti-corruption demands inherited from the previous administration and helped to portray the Toledo administration as invested in actively fighting corruption.

From the above account, it is possible to assert that the CNA’s role during the first year of the Toledo administration largely involved the stimulation of political support for the government – that is, to offer a display of official anti-corruption efforts without unnecessary spending of political resources or fostering new sources of pressure. Toledo’s government created an empty structure able to publicly display commitment to anti-corruption efforts, while being politically and financially restricted to address systemic change. Therefore, based on the above anecdotal and contextual information, it seems that the commission that Belaunde Moreyra presided over was only as attractive as the political capital it could generate.

However, early in the second year of the Toledo administration, circumstances in and around the CNA began to change and affect its political value. After generating what little political capital was available from the symbolic implementation of an anti-corruption agency, the CNA reached the peak of its relevance in the political cycle and entered a period of institutional devolution, ultimately ending in its deactivation.

Fall from grace

With political and financial dependence on the executive, the CNA was always going to struggle to be an independent body. The characteristics described earlier made it particularly difficult for the commission to undertake any serious actions to execute its official duties. However, with time, the Commission tried to pursue its own institutional evolution, leading to confrontations with the administration.

The Commission became interested in gaining institutional legitimacy by following two strategies. First, it pursued the improvement of its legal status by seeking support in Congress. The Commissioners’ intention was to institutionalize the fight against corruption and ensure the longevity of the CNA by replacing the executive decree that created it with a proper law. The text of a legislative project was finalized by a subgroup of commissioners on October 4, 2002.

Second, the commissioners (including Belaunde Moreyra) saw the involvement of the CNA in investigations of corruption scandals as a way of attracting popular attention and increased support. While this strategy represented, in principle, a position common with that of the Toledo administration, it also required the Commission’s engagement in activities that would threaten the corrupt interests of powerful people. A member of the Public Procurator’s Office19 confirmed the pressure that was placed on anti-corruption investigators to forgo cases against powerful individuals and the resistance they experienced.

The CNA’s approach to high-profile cases started as early as February 2002, with a request for information from the National Superintendent Agency for Tax Administration (SUNAT, by its Spanish acronym) regarding irregular procurements. However, the real political friction began with the adoption of a critical position by the CNA during a scandal involving First Lady Eliane Karp and her contractual relationship with private bank Wiese Sudameris. As confirmed by a member of the CNA20 President Toledo and the Prime Minister were angered by it and did not provide any further support to the CNA. Another member of the commission21 states: ‘The Commission became a sort of enemy, or an agency that should be feared’. Indeed, as the media echoed the CNA’s position and highlighted its criticism of the whole affair (Núñez 2002), the government began to see the CNA as a source of pressure.

The last straw that precipitated the fall of the CNA was the interest of the anti-corruption czar Belaunde Moreyra in the case of Minister of Internal Affairs Alberto Sanabria Ortiz, who, after taking the post in January 2003, immediately faced allegations of corruption and favoritism (Chávez 2003). Due to Belaunde’s position, the government took steps to cripple the young anti-corruption commission, which by then had exhausted its role as a source of support and become a potential threat to its interests.

The last official meeting of the CNA took place on February 6, 2003. Exactly one week later, Martín Belaunde Moreyra was appointed as ambassador to Argentina,22 departing the country a day later. The rush was so great that his resignation was only made official two months later, on April 25, which meant that the ministerial decree had to explicitly state that it had ‘retroactive efficacy’. Finally, on April 30, the official newspaper El Peruano published Supreme Decree No. 047-2003-PCM, which modified the norm creating the CNA, rendering it harmless to political interests and justifying the decision to end the CNA based on complaints made by the Office of the Comptroller General (OCG).

From the beginning, and disregarding Toledo’s public discourse, other government agencies had criticized the work of the commission for what they perceived was an unconstitutional invasion of their functions. The most hostile among them was the OCG, which, although it had been included by norm as an observer member of the CNA’s activities, refused to participate in any of its meetings. By early February 2002, the Comptroller General had sent an official communication to the presidency of the Council of Ministers (PCM) requesting the modification of the CNA’s normative role in two key aspects: the removal of certain attributions from the Commission -particularly those related to ex post control duties that the OCG considered to be part of the latter’s prerogatives – and the exclusion of the OCG as a member.23 Soon thereafter, President Toledo and Prime Minister Dañino expressed to members of the Commission their opinion that the CNA should focus on prevention rather than the control of corruption.24

The management audit carried out by the PCM (2006: 11–14) explains that “as a consequence of [the OCG’s] apprehensions and previous coordination between the members of the Commission, the Prime Minister, and the President of the Republic… five of the eight articles of Supreme Decree No. 120-2001-PCM [were] modified.” Through communication dated October 05, 2006,25 the technical secretariat of the CNA reported that “among the most important consequences [of the modification of the norm in 2003], is the fact that the CNA lost the possibility of driving processes of investigation of presumed acts of corruption or lack of transparency in the different levels of government and especially in the higher spheres of political power.”26, 27

Following these measures, the CNA was further denied political relevance by the appointment of a political operator, Juan Paz Espinoza, as the executive secretary of the commission on May 1. This position had been officially created by the amending norm (published on April 30) to supersede the administrative duties of the president of the Commission. By doing so, the government managed to: (1) avoid the appointment of a new president, (2) disable any official meeting of commissioners, (3) guarantee its control of the CNA’s activities, and (4) keep the commission running (at least formally). For the next two years, no meeting of commissioners would be organized and most of the activities would revolve around the harmless and symbolic training of public officials and civil society.

When asked about the fact that no representative of the President of the Republic was appointed after the ‘resignation’ of Belaunde Moreyra, Jaime Reyes Miranda, secretary general of the PCM, explained that the appointment of the president of the CNA was the role of public officials other than the Secretary General of the PCM, noting further that he had fulfilled his duty of warning the successive presidents of the Council of Ministers about the need to designate a new president after the resignation of Martín Belaunde Moreyra (PCM 2006: 9).28

With respect to the same issue, Juan Paz Espinoza asserted that (PCM 2006: 7) coordination was carried out during important meetings with the prime minister.29 The executive seemed to be aware of the situation of the CNA, purposely keeping it without leadership and excluding it from participation in other government agencies and civil society. As a result, according to the audit report of 2006, of the fourteen activities officially planned by the CNA between 2002 and 2004, the Commission focused on only four: the promotion of public ethics; training sessions; media campaigns; and interinstitutional agreements. In addition, the audit process found that in 2004 only 9.4% of the programmed activities could be accounted for, pointing to a serious deficit in management.

Transfer to the Ministry of Justice

The CNA was revisited through Supreme Decree No. 035-2005-PCM, signed on May 6, 2005. This second modification was brought about for a number of reasons, the most important being: (1) Toledo’s government was confronting its second-lowest level of popular support caused by a series of corruption scandals throughout the second half of 2004; (2) the UNCAC had been ratified by the Peruvian parliament on October 19, 2004; and (3) the Agreement between the Government of the Republic of Peru and the Government of the United States regarding the transfer of confiscated assets (signed on June 12, 2004), which saw the return of funds stolen during the Fujimori administration, demanded the setting of institutional arrangements to comply with the requirements included in the agreement. These three circumstances created a new wave of pressure on the government of President Toledo, which in turn put the CNA through a second, albeit shorter, political cycle.

The government’s response was to modify the CNA to allow its executive secretariat to hold official meetings of commissioners in the event of the president’s absence. In addition, certain preventive and control functions were improved. However, in the process of implementing these seemingly positive steps, the political dependence of the commission was further secured. On January 13, 2005 (less than three months before the modification of the CNA’s statute), the government replaced Juan Paz as executive secretary with Alberto Ygor Martínez Llanos.30 Martínez Llanos held the post of commissioner representing the Ministry of Justice, and like Juan Paz, was another actor the government knew could be counted on to keep the CNA under control. In contrast to his predecessor, who had been brought to the commission by Belaunde Moreyra from the Association of Lawyers of Lima, he was an official member of the incumbent party. According to a staff member of the CNA, Martínez Llanos was appointed because he was another political operator: “He was close to the president, and he belonged to [Toledo’s] party. So, it was a matter of ‘it’s your turn now’”.31 The new executive secretary had unsuccessfully run for Toledo’s Perú Posible party in the congressional elections of 2000.32

Subsequent events revealed the political instrumentalization of the commission for a second time. On May 20, 2005 (exactly two weeks after the modification of the norm), the government decreed33 that the CNA, along with the Ministry of Justice, would be the agency in charge of identifying the anti-corruption initiatives that the repatriated funds34 would be invested in. On August 8, the first meeting of commissioners in two-and-a-half years was finally held under the interim presidency of Martínez Llanos. During the next sessions, the task of identifying anti-corruption initiatives was properly discussed and reported, following government orders. This process was officially completed with the approval of the projects by Ministerial Resolution No. 402-2005-JUS on October 12. Two weeks later, on October 26, the government produced Supreme Decree No. 082-2005-PCM, ordering the transfer of the CNA from the sphere of the prime minister to the Ministry of Justice. In its new sector, the CNA was put under the supervision of Vice-Minister of Justice, Jaime Reyes Miranda, who had been Secretary General of the PCM between 2002 and 2005 – the years during which the CNA was paralyzed – and who became the commission’s new president in early 2006.35

With that decision, the CNA was effectively demoted to the sectorial level and, although, its mandate remained national, it became clear that, in terms of institutional power, it would have a limited impact on the fight against corruption.


The case of Peru’s CNA shows that government efforts to fight corruption (particularly in those countries where malfeasance pervades the public sector) follow a lifecycle highly dependent on political circumstances.

When political capital is needed, leaders are incentivized to propose and adopt measures to fight corruption in direct relation to the benefits available. This was found to be the case in the hurried and improvised way in which President Alejandro Toledo first appointed an anti-corruption czar and soon afterwards created the first national anti-corruption commission. The political and symbolic nature of this office was sufficiently clear in the formal powers granted to it and the limited resources and support provided thereafter.

Furthermore, once the commission fulfilled the goal set by the administration and political priorities changed, the study found that its development followed a pattern consistent with the tipping point of our theorized political cycle. The deliberate devolution of the CNA was evident in the appointment of its president to the post of ambassador to Argentina, the sudden and dramatic downgrade of its functions, and the appointment of a political operator tasked with keeping the commission running. These actions to limit the efficacy of the CNA seems to confirm the logic offered earlier that support is withdrawn and a counter-reform process takes place when the cost-benefit ratio of an anti-corruption measure tips against the interests of the political leadership.

In short, this study shows that the CNA was implemented only to the extent that it could provide political capital without affecting the status quo. In doing so, it supports the logic that anti-corruption policies are only as attractive as the benefits they provide to the government. When the latter favors political capital, measures to curb public malfeasance are not only possible, but realistic. On the other hand, when government is engaged in illegal activities, anti-corruption measures are not allowed to be anything more than symbolic actions. In this respect, the Toledo administration reflected both sets of circumstances between 2001 and 2003 by first focusing on novel anti-corruption demands and creating an image of clean leadership, and then rushing to control the emergence of new corruption-related scandals.

Finally, the above findings suggest that, in acknowledging the mutability of leaders’ preferences and interests, it is possible to determine the real anti-corruption potential and mechanics of an administration by looking at the concrete actions it takes to either support or devolve anti-corruption policies. While this may sound like a self-evident truth, the international anti-corruption movement could greatly improve its approach to national implementation of anti-corruption commitments by dropping any assumptions regarding local political willingness to reform, and instead designing specific approaches using a case-by-case analysis of leaders’ interests and investments. Only in this way can we take advantage of political cycles in the fight against corruption.