Keeping the New Broom Clean Lessons in Human Resource Management from the kpk

Indonesia has been struggling to reform its corrupt bureaucracy for years. In 2002 an independent Corruption Eradication Commission (kpk) was established by law. This article explains how the commission has managed to avoid the usual corrupt practices—theultimatedangertoitsmission—bycircumventingtheflawedincentive-system of the civil service. It compares key features of the kpk’s human resource management system with the practices of the general civil service and argues that autonomy in internal budget allocation, recruitment practices, performance-based promotion, remuneration, and attendance have fostered kpk staff integrity. The ability of the kpk to promote job descriptions and performance-based remuneration in the civil service dissipated once it went beyond select allies, but has ultimately fed into a new civil service law passed in 2014. To sustainably reduce corruption, the new Indonesian government must rigorously follow through with civil service reform.


Introduction
The Komisi Pemberantasan Korupsi (Indonesian Corruption Eradication Commission, kpk) has had a consistently better public image than other law enforcement agencies in Indonesia since it started operations.1 This image can mostly be attributed to its outstanding success in investigating and prosecuting high-profile corruption cases and to its organizational culture of integrity (Schütte 2012).
The Indonesian bureaucracy is infamous for its corruption. It is the result of a deeply entrenched system of misguided incentives, opportunities for abuse, and a lack of accountability. In 2002 the kpk was established by Law No. 30 of 2002 (henceforth: kpk Law) to spearhead anti-corruption efforts. The kpk was given a broad mandate: to coordinate with, and supervise, other authorities empowered to fight corruption; to investigate and prosecute corrupt acts; to undertake preventative measures; and to monitor the state administration. After the appointment of its first five commissioners in December 2003, a process that involved community representatives, the legislature, and the president, a new organization had to be built and staffed.2 Considering that the kpk was established in response to a crisis of trust in the existing law enforcement agencies (Butt 2011), the importance of keeping this new agency free of malpractice can hardly be overstated. Only if the new agency's commissioners and staff adhered to its principles and refrained from corrupt or collusive practices could the kpk be credible and effective in its task of cleaning up the bureaucracy. The comparative literature on anti-corruption agencies stresses the vital importance of their operational independence and of having honest, well-trained, and performance-oriented staff (Meagher 2005;Klemencic, Stusek and Gaika 2007;De Sousa 2010). A study of the bureaucracies of 52 countries by Dahlström, Lapuente and Teorell (2012) found that employment terms, particularly meritocratic recruitment, reduce corruption when controlling for the impact of most standard political variables.
The above literature fails, however, to address how performance orientation and integrity of personnel can be achieved in the context of a malfunctioning civil service that is infested with corruption. An exception is De Speville (2010), who describes how the Independent Commission Against Corruption (icac) in Hong Kong induced loyalty in its personnel-both civil servants and Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 non-civil servants-by managing its human resources autonomously based on renewable contracts. This article examines how human resource management has been used to prevent the compromise of the kpk by the illicit informal institutions typical in Indonesia's civil service. It draws heavily on interviews with the first kpk commissioners, who negotiated the foundations of its human resource management (hrm) system with the Indonesian government led by Kementrian Negara Pendayagunaan Aparatur Negara (State Ministry of Administrative Reform, MenPAN). In-depth interviews were also conducted with the head of the Human Resources Bureau of the kpk and a number of other people involved in the design and application of the kpk's human resource management system. Government regulations and organizational manuals were studied and verified with data from interviews, and statistics and reports provided by the kpk.
The article begins with a brief discussion of the weaknesses of the Indonesian civil service system, the illicit practices it has nurtured, and why it was not a suitable model for the kpk. Subsequently, the main points the kpk negotiated with the government regarding the establishment of its own human resource management system are outlined. The kpk human resource management system is then contrasted with the Indonesian civil service system, and it is argued that it is mainly due to its system that the kpk has not succumbed to widespread practices of corruption. Autonomy in internal budget allocation, recruitment practices, and performance-based promotion and remuneration are discussed as indirect means, or incentives, that have fostered staff integrity. In the final section i consider the applicability of the kpk's system to the wider civil service in Indonesia, and sum up the piecemeal and hesitant efforts undertaken by the Yudhoyono government.

Weaknesses of the Indonesian Civil Service System
In late December 2003, the kpk had to be built from scratch as a new organization. Personnel had to be recruited and internal management systems had to be put in place that would support its operational autonomy and integrity. This would have been very difficult, if not impossible, under the existing Indonesian civil service system-a career system in which promotion depended on seniority rather than performance, and low base pay is 'supplemented by a maze of allowances ' (adb 2004:63). In fact, the formal incentives of the civil service, a low base salary and allowances that lack transparency, are widely considered to be the main cause of illicit informal practices, such as moon-schütte Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 lighting and various kinds of corruption. Turner noted in 2002 that Indonesia was a 'cautious diner' of new public management (npm) reforms, with decentralization and privatization initiatives evident on the reform agenda, but only slow and sporadic systemic npm-changes being made to central agencies. His assessment would likely be similar more than ten years later. The weaknesses of the Indonesian civil service system have been discussed in detail elsewhere (Palmier 1985;Rohdewohld 1995;adb 2004;McLeod 2005;Synnerstrom 2007;World Bank 2004. Hence, only a brief overview of the civil service's weaknesses most relevant for the kpk, that is, those weaknesses resulting in illicit informal practices, is presented here. This serves as a benchmark for the following discussion on the kpk's human resource and integrity management system. The Indonesian civil service is a single, nation-wide, national civil service. It originates from the Pangreh Praja (rulers of the realm), or Binnenlandsche Bestuur, as the Dutch called the lower level of their colonial administration. The Pangreh Praja consisted of chiefs and nobility (priyayi) whose traditional privileges were step by step replaced by civil-servant salaries as part of the modernization of the colonial administration in the early nineteenth century (Sutherland 1979). During the Indonesian revolution, the Pangreh Praja was restyled in the less elitist Pamong Praja (servants of the realm) and became the core of the decolonialized Indonesian civil service (Sutherland 1979).
Law No. 43 of 1999 on the Civil Service distinguishes between two main types of positions for civil servants (pegawai negeri sipil, pns): structural and functional. Structural positions are management positions in a hierarchy of four echelons; functional positions are non-management positions that require specific expertise, such as teachers, doctors, and technicians. Staff are allocated not according to operational needs, but by adhering to a fixed 'mechanistic' establishment plan ( formasi), which obliges a unit that is headed by an echelon i official to include at least four echelon ii, 16 echelon iii, and 64 echelon iv officials as well as additional functional and auxiliary staff (Synnerstrom 2007:165).3 In 2004, when the kpk started operations, the recruitment and appointment of civil servants were managed separately, so that entrance examinations tested only general education and not skills required for a particular position (adb 2004). Recruitment has been likened to a market place in which future income can be bought (Kristiansen and Ramli 2006). Based on a small survey of Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 civil servants who had paid senior government officials to obtain government employment in West Nusa Tenggara, Kristiansen and Ramli (2006:221) found that '[t]he amount needed to buy a position is […] approximately two-and-half years' full salary and investments are in most cases based on loans from family and close friends' . For the period 2002 to 2004, the average payment made through middlemen to senior government officials to obtain an initial position was Rp 27.4 million (us$2,900; Kristiansen and Ramli 2006). These investments have to be recovered, resulting in financial pressure on new civil servants to join in common illicit practices. Another, more sinister, reading of this practice is that those wanting to become civil servants are willing to pay substantial bribes up front in order to be recruited, and that they are knowingly making a significant investment precisely because it will give them the opportunity to earn illicit income.
Once a civil servant has become part of the system, market transactions continue. The adb reported in 2004 that promotion was predominantly an automatic process of training and examination, unrelated to performance. Especially for those positions where additional allowances, such as project honoraria, were to be expected, there is anecdotal evidence of considerable fees being paid for promotions as well (Kristiansen and Ramli 2006). This was sustained by mostly internal (non-advertised) selection for positions (adb 2004). Ross McLeod (2005 depicts a civil service that is very different to what the kpk commissioners sought to create in their own organization: When promotions are largely automatic, individuals have little incentive to perform better than their peers. Overstaffing at higher levels becomes the norm, since even poor performers are promoted rather than being either kept in their present positions or encouraged to leave. Those who do perform well are frustrated by the lack of meaningful recognition of their efforts. This may well result in their leaving the organization for private sector jobs where they can expect to be rewarded appropriately, or simply becoming dispirited and significantly less productive as a result. At the same time, there is little incentive for individuals to increase their productivity by undertaking additional education and training, since in practice there is scant payoff in terms of salary increases or promotions. The civil service's remuneration system, with its low and compressed base salaries, not only impedes performance, but also necessitates the search for additional income through both legal and, in many cases, illegal sources. In a national survey conducted by the Partnership for Governance Reform in schütte Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015)  Indonesia in 2001, the main cause of corruption was attributed to 'low civil servant salaries' , followed at some distance by poor 'law enforcement' and 'lack of morals' (Partnership for Governance Reform 2001).4 A more in-depth analysis of the problem, however, suggests that the main problem is located in the composition of civil service remuneration, rather than its level (Synnerstrom 2007). This will be discussed in more detail below.
If the kpk wanted to establish a 'clean' organization, free of the common malpractices, it had to avoid taking over the system of distorted incentives of the existing civil service. It needed to ensure that its vacancies would not become commodities in an informal market and that its employees would not be corrupted out of 'need' and peer pressure.

Negotiations for the kpk's Own Human Resource Management System
As of late December 2003, the kpk consisted of nothing more than five commissioners (henceforth also referred to as the kpk leadership). They were allocated a number of temporary, seconded staff from various state agencies and had to negotiate with the government for all other resources, including offices, a budget, and further staff.
The kpk was one of a number of new commissions to be established in Indonesia within the first five years of the reform era. Dozens of new auxiliary state bodies had been established after 1998, outnumbering ministries, departments, and non-departmental organizations (Deputi Bidang Kelembagaan 2008). The State Ministry of Administrative Reform had to approve the organizational set-up of each of these new bodies. Thus, it was no surprise that the inter-ministerial team under the auspices of the State Ministry of Administrative Reform reacted cautiously when the newly elected kpk leadership and initial staff members presented their ideas about the future human resource management system of the organization.
The kpk Law is clear about the right of the commission to recruit and employ qualified non-civil servants based on their expertise as 'kpk officers' (pegawai kpk) (Arts. 24, 1, and 25, kpk Law). Nothing was stipulated, however, as to the remuneration of the commissioners and the employees. Dis-Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 cussions began, but it took two years to produce a government regulation on the human resource management system of the kpk, and two and a half years to produce a government regulation on the remuneration of its leadership.5 The kpk demanded its own remuneration system in line with the performance-oriented aspects of the corporate sector, not the seniority system of the civil service (Hardjapamekas 2006). The kpk wished for as much autonomy as possible, including the right to set its own budget without approval of the Ministry of Finance and salary levels competitive with the corporate sector. The outline for the human resource management system that the kpk prepared was based on the assumption that the kpk would have only its own employees, not civil servants seconded from other organizations, such as the police and the Attorney General's Office. It was feared that seconded personnel might infect the new organization with the old practices, and would also be the subject of a conflict of interest when the kpk had to investigate or take action against their original organizations.6 In December 2005 a compromise was stipulated in Government Regulation 63 of 2005 on the Human Resource Management System of the kpk. Instead of spelling out the details of a human resource management and remuneration system, the government regulation allows the kpk autonomy in managing its annual budget, including most significantly the budget for human resources and salary levels. The overall budget, however, needs approval of the Ministry of Finance.7 Unlike other state agencies, the kpk has no possibility to spend beyond the budget ceiling. The regulation also specifies the kpk's right to manage its own employees (pegawai kpk), as well as employing and managing seconded civil servants (pegawai negeri sipil) from other state agencies, such as the police and public prosecutors. Prosecutorial powers are tied to an oath exclusive to public prosecutors.8 To date, the kpk cannot administer such an oath and has to employ seconded civil servants to exercise its prosecutorial function.
Government Regulation 63 of 2005 does not give the kpk the complete autonomy over human resources that its commissioners had sought. Within   2005, 2006, 2007a, 2008a, 2009, 2010, 2011, 2012, 2013, 2014 the budget approved by the Ministry of Finance, and under the condition that it employs public prosecutors for prosecutorial tasks, the kpk is, however, entitled to develop and manage its human resources autonomously.

The kpk's Human Resource Management System
Unlike other new auxiliary state bodies, the kpk has not been hampered by insufficient budget. There have been no complaints of scarce resources, with the exception of the matter of the construction of a new office building, which for several years was rejected by the legislature as it reasoned that the kpk could continue to rent office space. After a public campaign to 'collect coins for the kpk' , the budget allocation for a new building next to its current rental premises was eventually endorsed (Schütte 2013). Table 1 shows a steady increase in the kpk's total budget and its human resources budget over the ten years since its inception. The percentage of human resources expenditure in relation to the overall budget has fluctuated because of other major expenditures, such as the restoration of an office building in 2006 and 2007. The expenditure per staff member, however, steadily increased Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 from 2005 to 2012. The slight decline in 2013 can be explained by the recruitment of a large number of staff towards the end of the year, thereby raising the denominator while not affecting the numerator as much. The overall rise is partly to be explained by the introduction of the kpk's own remuneration system after Government Regulation 63 of 2005 was enacted, but also by the introduction of bonuses in 2007. It may also be explained by a declining ratio of civil servants to kpk employees. kpk does not pay the civil servants' base salary, thus seconded personnel are slightly cheaper. By way of comparison, in 2009 the Attorney General's Office (ago) had an overall budget of Rp 1.98 trillion, human resource expenditures of Rp 1.1 trillion, and 19,459 staff.9 That is about Rp 56.5 million per staff member, hence only about a quarter of the Rp 218 million the kpk spent per staff member in 2009. Thus, in absolute terms, the kpk invests considerably more in its human resources than the ago. The budget ceiling stipulated in Government Regulation 63 of 2005 explains the cautious expenditure of only about 70-96% of its annual human resource budget.10

Recruitment
Under Government Regulation 63 of 2005, recruitment must be conducted openly. It must therefore allow external applications. Since 2004 individual vacancies have been publicly advertised and several general recruitment rounds called 'Indonesia calls …' have taken place. The slogan 'Indonesia calls …' is meant to evoke a sense of vocation to serve the country.11 Recruitment rounds and vacancies are advertised on kpk's website www.kpk.go.id and in major newspapers. Applications, except for secondments, must be submitted online. Any direct contact with kpk personnel leads to the disqualification of the applicant. To avoid (accusations of) nepotism, kpk does not allow the employment of immediate relatives or spouses, even when couples meet and marry at the kpk. In the latter case, one of the spouses must leave the agency.
The selection of candidates is conducted in a multi-tiered process, the first steps of which are managed entirely by a consultancy firm. The process consists of administrative screening, potential and competencies tests, reference checks, interviews (by the hiring unit where the candidate will be employed), health examinations, and background checks. Other than the interview by the recruiting unit, all steps are conducted by the independent consultant (Kemala 2009a). The reasons given as to why somebody wants to join kpk are taken very seriously.12 Table 2 shows that the kpk has been flooded by applications, the majority coming from the Jakarta area.13 Vacancies are also filled through in-house applications, based on the applicants' performance. As mentioned before, all vacancies, except for those for police work and public prosecutors, are announced on the kpk's website and recruitment is open. kpk staff members are eligible to apply for these vacancies. No statistics were available on promotions. Interviews indicate that the kpk supports in-house careers, while at The kpk also submits requests for qualified, seconded personnel to other state agencies. Although the kpk may sometimes have a certain candidate in mind, the selection of candidates from the civil service is at the discretion of the originating civil-service organization. The short-listed civil servants must undergo the same selection process before they are admitted. Secondments are for a period of four years and can be extended once for another four years, and then for another two years (Government Regulation 103/2012 Amending Government Regulation 63/2005). The civil servant may also choose to leave the civil service and become a kpk officer, and 54 staff had done so by mid 2014. schütte Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 The return of civil servants to their former organization has been generally conducted without much public attention. In 2009, the head of the Badan Pengawasan Keuangan dan Pembangunan (Finance and Development Supervisory Board, bpkp) and a former police general, Didi Widayadi, questioned the extension of seconded bpkp personnel after four years and announced their dishonourable dismissal from the bpkp should they choose to become kpk employees.15 This threat was considered by many to be part of the conflict between the police and the kpk in 2009. The threat never materialized. In late 2012, however, the police withdrew 36 police investigators after their four-year term was up, in what many see as a direct retaliation for the kpk's investigation of several senior police officials involved in a corruption case in the procurement of a Rp 1 trillion contract for a driving simulator. Several other police officers working as kpk investigators also resigned of their own accord. The kpk admitted that this impeded the speed of its investigations. Nevertheless, it used this as a strategic opportunity to argue for the training of its own investigators, the first of whom took on cases as of January 2013.16 In May 2015, the Pengadilan Tindak Pidana Korupsi (Court for Corruption Crimes) ruled in a corruption case at first instance that the investigation of the defendant by kpk investigators (who were not seconded from one of the other law enforcement agencies) was illegal. The kpk announced that it would appeal.17 Table 3 shows that in the first year there was a high number of temporary seconded staff (see 'Other' column). Since 2005 the majority of seconded civil servants originated from the police, the Attorney General's Office, and the Finance and Development Supervisory Board. The majority of the police officers and all prosecutors work under the law enforcement department of the kpk, which leaves the other departments (prevention, data and information, internal oversight and public complaints, and the administration) predominantly staffed by kpk permanent employees (pegawai kpk).
The growth of kpk's personnel brought a change in the ratio of kpk officers and seconded civil servants. In 2008, kpk permanent staff for the first time outnumbered seconded civil servants. Following national policy, the number of contract staff has been considerably reduced since 2008.18 In 2013, seconded 15 VIVAnews, 25-6-2009. 16 Kompas, civil-servants constituted 19% and kpk officers 66 % of the total of 955 personnel ( Figure 2). Based on a workload analysis conducted in 2008 together with the Bandung Institute of Technology, the kpk planned to increase its personnel to almost 1,300 by 2011.19 Figure 3 shows that by the end of 2012, the number of staff had actually temporarily decreased due to the early departure of several police investigators. The kpk continues to conduct annual workload analyses, but the actual number of recruits has been constrained by the number of qualified applicants and the size of the current office building, the construction of a new purpose-made building being underway at the time of writing.

Performance Management
In 2006, the kpk introduced the balanced scorecard and key performance indicators (kpi) to the organization as mechanisms to manage and assess staff performance. The balanced scorecard is a performance management tool that allows managers to monitor their organization from several interrelated per- spectives and identify those measures most critical to success (Niven 2003). kpk has developed strategic maps for all its areas of intervention and its internal management. These strategic maps represent different perspectives: internal, financial, learning and growth, as well as stakeholder expectations (kpk 2008b). Key performance indicators have been developed for the whole organization and cascade to the departments, directorates, teams, and individuals. Each staff member has five to eight key performance indicators to fulfil. These kpi steer individual performance towards the fulfilment of strategic organizational goals (Kemala 2009a).
For a senior officer working in the Human Resource Unit, internal procedures would be given most weight on the annual scorecard, while the externalstakeholder perspective would be less important for such a position. Key performance indicators would include the recruitment of a certain number of new Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 staff, the training of staff, and workload analysis, feeding into the annual performance indicators for the whole unit.
The kpk has also compiled a list of minimum core competencies (that is, organizational commitment and self-control), primary competencies (that is, planning and organizational skills and self-confidence), and secondary competencies (that is, team leadership and analytical thinking) (kpk 2008a). These criteria are assessed in performance evaluations and used for individually tailored training programmes and the setting of remuneration incentives.
Performance evaluations (scorecards) are conducted annually and are compiled via an intranet application. The performance evaluation is based on selfassessment, assessment by immediate superiors, and assessment by a more senior superior, and is followed up with a feedback session between the superior and the subordinate during which areas for development are agreed.
Individual scorecards are 50% result-oriented and 50 % process-oriented. Result orientation here means the achievement of previously agreed performance targets (kpi), while process orientation means the development of new competencies and adherence to the code of conduct (Kemala 2009a;kpk 2009). Performance evaluations directly affect staff remuneration: the grading of the salary as well as potential bonus payments. Hence, there is not only a strong incentive to perform well in line with previously agreed targets, but also to develop competencies and participate in training programmes.

Remuneration
Government Regulation 63 of 2005 gives the kpk the authority to establish its own remuneration system, thus determining the salary of its staff within the limits of the overall annual budget allocated from the national budget. The only type of remuneration not regulated by the kpk itself is the salaries of the leadership. As state officials (pejabat negara), their remuneration is determined by government regulation.  kpk Staff Remuneration Each month kpk employees are reminded that their salaries are paid for by the Indonesian people. 'My salary comes from the people' is written on the cover of the salary slips and is there to remind kpk employees of their role as agents for the Indonesian people. This stands in marked contrast to what has been found to be the opinion of Indonesian government employees: that they 20 Detiknews.com, 26-10-2009. This includes allowances, but it is unclear whether the residences provided by the state and their maintenance are also included in these sums. 21 Detiknews.com, 26-10-2009.
Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 are 'the "owners" of the nation and the general public their servants' (Schwarz 1999:135). The 'my salary comes from the people' slogan is therefore a direct attempt to influence employees' mind sets in line with the incentive system provided by the remuneration system. Every kpk employee receives a monthly, single salary payment as take-home pay. Employees are further entitled to transport allowances that are based on actual office attendance and the submission of a timesheet. Life and health insurance as well as pension payments are made directly by the commission to the service providers. Unlike civil servants outside the kpk, progressive income tax is also deducted directly from kpk employees' salaries. Seconded civil servants only pay tax on the top-up provided by the kpk.
The remuneration system developed since 2005 takes into account position, personal qualifications (competency), and performance to determine salary grades (the so-called 3p system). All positions are graded by a points system on a scale ranging from 5 to 22. Grade 1 to 4 positions, such as cleaning and building security personnel, are outsourced to third parties. Each grade has five subdivisions, depending on individual competencies. This creates a great incentive to participate in training and take on new job assignments to increase individual competencies and hence salary. In terms of line management, however, hierarchies in kpk are relatively flat, comprising only four levels (commissioner, deputy, director, staff) sometimes adding a fifth level (project leader) for better team coordination. Table 5 shows the grading of administrative, functional, and structural (management) positions. It is remarkable that senior functional staff can actually earn just as much as a kpk director (Grade 19-20), a structural position corresponding to echelon ii in the civil service. Specialist expertise is highly valued at the kpk, whereas in the general civil service functional staff earn less than structural staff, especially when allowances are considered.
Annual and monthly bonus payments for extraordinary performance can be added to the base salary (gaji pokok) of kpk staff. The annual performance evaluation determines whether an employee is entitled to a single annual bonus payment of up to 200% of his or her monthly salary, or a monthly incentive payment of 9%, 7%, or 5% of the monthly salary (or nothing at all) during the next twelve months. Unlike the civil service, the kpk leadership and employees are not allowed to receive additional income such as honoraria for attending workshops or speeches, or project fees. Salary levels are determined by benchmarking against salary levels for employees in the corporate sector and are regularly adjusted. Surveys such as the Indonesia salary handbook (Kelly Services 2008) have been used as a reference by the kpk. Benchmarking against the corporate sector is not a practice used by the Indonesian civil service. In the civil service, base salaries are based entirely on length of employment and level of education, not performance. The base salary as displayed in Table 6 is well below salaries in the corporate sector. Beside the absolute salary levels, the ratio between the highest and the lowest salary, or the level of salary compression, is also indicative of an egalitarian or a more competitive system with strong monetary incentives to rise through the ranks. Salary decompression means increasing the ratio between the highest and the lowest salary, ideally giving employees an incentive to perform well and get promoted rather than move to the private sector (undesa 2005). At the kpk, the ratio between the gross salary of the lowest and the highest grade is 1:13.6.22 Not only do the kpk salaries start at a higher entry level, they are also 22 Low-paid cleaning and clerk jobs (ranks 1-4) are outsourced and not included in this calculation. If these salary grades were included, the ratio would be even bigger.
Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 source: appendix of government regulation 8 of 2009 to the 11th amendment of government regulation 7 of 1977 on the salary of civil servants much more decompressed than in the civil service and therefore provide higher monetary incentives to achieve a higher rank. The ratio of the lowest-rank to the highest-rank civil-servant salaries is only 1:3.3 (Table 6).
But the base salary is only part of the overall income. The ratio of allowances for the lowest rank compared to those for the highest rank also needs to be considered.
In contrast to the kpk's decompressed base salaries system, it is the allowances that are decompressed in the civil service system. In 2009 a World Bank mission established a ratio of allowances of 1:15.3 (echelon ia to ive) for structural positions (World Bank 2009:24). The ratio of the lowest-grade to the highest-grade transport allowance at kpk the same year was 1:4.7. The overall ratio of the lowest to the highest gross salary (including transport allowance) at the kpk was 1:10.9 (not considering possible bonuses). In fact, because bonuses are calculated as a percentage of the base salary, they will further decompress the overall income structure.
Similarly, the higher echelons in the civil service are generally able to earn considerable additional income from speaking fees and other honoraria that lower-ranking officers do not have access to, therefore significantly decompressing real earnings in the civil service.23 The lack of transparency regarding schütte Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 these additional sources of income not only makes comparisons difficult, but is also frequently cited as an opportunity for abuse.
The salaries at the kpk are much higher than the base salary of civil servants and, together with other elements of the working environment at the kpk, have been eyed with envy by some civil servants outside the kpk.24 Nevertheless, including their various allowances and honoraria and by not being liable for income tax,25 civil servants in the upper echelons may take home as much as their counterparts in the kpk. The difference is that the kpk applies clear and transparent, performance-based rules on who is entitled to what kind of payment. Thus, the main distinction between the kpk system and the civil service system lies in the performance orientation and transparency of the kpk system, not necessarily in the actual amount that employees take home. In general, it is very difficult to compare take-home pay, even across agencies that have already undergone considerable reform, as different grading schemes are applied (Haypoint and Factor Evaluation) and even the same grading schemes are applied differently.26 kpk's seconded civil servants are supposed to be treated as equal to kpk employees, but incompatibilities remain. If the income of seconded civil servants from their organization of origin is below kpk level, the kpk pays for the gap. If civil servants get promoted upon their return to their organization of origin, as has happened in a number of cases, the drop in salary may not be too significant.27 However, there has been at least one case where a highly qualified, former functionary of the civil service was seconded to a structural position in the kpk but, upon return to his functional position in the organization of origin, was not able to keep the rank and the salary that had come with the secondment. The lack of transferability of ranks between agencies is one of the disadvantages of having a stand-alone system. Another example of incompatibility is the current inequality in retirement age, which is 60 years for civil servants, but 56 for kpk officers. The kpk acknowledges there is room 24 Kompas, 9-9-2008, and interview with kpk's head of the hr Bureau, Jakarta, May 2009. 25 The income tax on the base salary of civil servants is charged to the national or regional budgets (Art. for improvement in the current system.28 Anecdotal evidence, however, also indicates that those who have left the kpk, both civil servants and kpk officers, have generally had no difficulty in finding a new assignment and have been successful in their new jobs.29 This is not least due to the outstanding training and career development schemes that the kpk offers to all its employees.

Comparison of the kpk and the Civil Service System
The kpk's human resource management system has avoided the weaknesses of the Indonesian civil service system by providing more transparent and performance-oriented remuneration, training, and promotion. Table 7 summarizes the key differences between the kpk's human resource management system and that of the civil service. The majority of my interviewees among former and current kpk staff drew attention to the quality of the kpk's personnel and their commitment as a crucial factor in the kpk's performance. Recruiting and retaining qualified and motivated staff is important. The kpk has access to the same pool of applicants as other organizations in Indonesia. Its mission to clean Indonesia from corruption, and its reputation as a clean organization, may have attracted applications from persons who usually would not apply for a 'government' job. It may also have attracted a higher proportion of applicants who would like to change how things are done in government. Even if the proportion of motivated people in the kpk's pool of applicants is bigger, and even if its rigorous recruitment process detects the most suited candidates for the job, it is unlikely that these recruits would perform equally well under the misaligned incentives of the normal civil service scheme. Anecdotal evidence from interviews suggests that some kpk employees applied to the kpk because they were tired of the conflict between their consciences and the pressure exerted on them to participate in corrupt practices in their old workplaces.30 The salaries at the kpk are comparable with the Indonesian private sector, particularly if an employee performs well and gets an annual bonus. kpk staff 28 Interview with kpk's head of the hr Bureau, Jakarta, June 2009. 29 Interview with a former kpk commissioner, Jakarta,  formance indicators lead employees to work in line with organizational goals. There are strong incentives to perform well-not just bonuses, but also opportunities for specialist training, both in-country and overseas, which particularly appeal because of the relatively young age of the staff. The opportunity cost of misconduct and getting dismissed is high: it would mean missing out on a salary higher than in the civil service (earned honestly) and training opportunities. The kpk also introduced 'at cost' reimbursement for operational expenses, including travel costs. Inflexible and unrealistic travel rates had distorted travel planning in the civil service, with civil servants often forced to report more travel than they actually undertook to make ends meet. The 'atcost' system for travelling has since been introduced in all ministries (Kuris 2012:10). The incentives laid out in the human resource management system and in operational standards may discourage staff misconduct. Nevertheless, the kpk has also put in place more direct measures to prevent and penalize misconduct. Internal rules are codified in a Human Resource Manual and two codes of ethics. These are based on a broad consensus and set boundaries of permitted behaviour that are controlled by peers, superiors, and technology. The potential costs of misconduct are increased by the additional severity of penalties for corrupt conduct by kpk staff. A commissioner or staff member of the kpk who is found to have committed corruption will have his or her punishment increased by a third of the sentence requested by the public prosecutor (Art. 67, kpk Law).

The kpk System: A Role Model for Civil Service Reform in Indonesia?
In 2009 the World Bank (2009:27), in a report on civil service reform, recommended the kpk's salary system as a 'template for general pay and grading reform in the Indonesian civil service' . Elements of its system could be applied in the general civil service. Nevertheless, Government Regulation 63 of 2005 on the kpk's human resource management allowed the kpk to bypass the existing regulatory framework for civil servants. Unless a new agency obtained a similar exemption from the general civil service system, an accurate replication of the kpk system would be impossible. Reforming an existing agency along the lines of the kpk system is a much more complicated endeavour, considering that the personnel would already be in place and not necessarily possess the qualifications needed for a more performance-oriented system. At the national level, an increasing number of exemptions would also be problematic; after all, schütte Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 the Indonesian civil service is a single, national service and different human resource management systems in different organizations may result in confusion and inter-agency jealousy. In recent years, the reform efforts by the government have meandered between 'reform for all' and 'pilot projects' in only a select number of agencies. The kpk's strategy to foster bureaucratic reform has been part of this meandering.
The kpk was quick to identify bureaucratic reform in general and civil service reform more specifically as an important factor in preventing corruption in Indonesia (Sunaryadi 2006). Officially, there were already other agencies in charge of the bureaucracy: the State Ministry of Administrative Reform, the Badan Kepegawaian Negara (National Civil Service Agency, bkn) and the Lembaga Administrasi Negara (Institute of National Administration, lan). This triad of agencies with partly overlapping mandates and the absence of a clear, single authority were considered by many as the main obstacle of reform. Synnerstrom (2007:162) noted that 'this set-up has so far proven incapable of delivering reforms. A body with the capacity to push for, coordinate and follow up on reforms needs to be established, with the ultimate aim of restructuring the current set-up' .

From Gentlemen's Agreement to Grand Design for Bureaucratic Reform
In the kpk's first years, its commissioners regularly lobbied and consulted with those in the state apparatus keen to reform and those critical of reform. Its strategy shifted between the creation of a central authority to manage bureaucratic reform, directly reporting to the president, and a more decentralized approach of pilot agencies, such as the Supreme Court.
Starting in late 2006, the kpk, the Supreme Court, the Department of Finance, and the Supreme Audit Board met bi-weekly and consulted with each other about the progress of reform in their organizations.31 Reform in the Department of Finance had already begun under the Megawati government and gained momentum under the first sby government, when Sri Mulyani was finance minister (2005)(2006)(2007)(2008)(2009)(2010). Tied by the regulations of the civil service, and with an existing staff of 63,000 spread across Indonesia, reform in the Department of Finance had very different dimensions compared to the institutionbuilding of the kpk. The reform at the Department of Finance lends itself as a more direct role model for other existing agencies equally tied by the 31 Correspondence with a senior adviser to the Minister of Finance, Jakarta, July 2009; interview with a former kpk commissioner, June 2009.
Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 civil service law. In 2008, McLeod noted in an assessment of developments in Indonesia that the Finance Department 'is serving as a pilot project to lead reform of the bureaucracy, rather than waiting for centrally driven wholeof-government reforms that seem unlikely to materialize in the near future' (McLeod 2008:197). When Sri Mulyani took over as Finance Minister in 2005, 'she decided to push forward vigorously with reform in her own ministry, clearly signalling her intentions by replacing the heads of the "notoriously corrupt" tax and customs and excise directorates general (dgs) within four months of taking office' (McLeod 2008:197). She replaced 1,200 personnel in the customs office in Tanjung Priok. Sri Mulyani established a system of both rewards for good performance and increased monitoring, and even requested the kpk in 2008 to raid the customs offices for another clean-up (McLeod 2008:201). Apart from these symbolically strong expressions of commitment to reform, under Sri Mulyani the Department of Finance's organizational structure was changed and the human resource management system underwent a comprehensive overhaul, including: [The provision of] detailed job descriptions for each position; grading of each position on the basis of its scope, the competencies required and the risks that need to be managed by the incumbents; determining a structure of remuneration that reflects this grading; identifying training needs by comparing individuals' skill sets with the skills required in the positions they occupy or are likely to occupy in the future; quantifying surpluses and deficits of skills within the existing workforce; and developing a system for monitoring performance and rewarding or penalizing individuals accordingly. mcleod 2008:198 In principle, this is not very different from what the kpk established, with the difference that the kpk did not have to deal with the inertia of incumbent staff. Unlike the kpk, the Department of Finance could not raise the base salaries; instead, it introduced a special allowance. This special allowance for Finance Ministry officials (tunjangan khusus pembina keuangan negara, also frequently called 'performance allowance') supposedly replaced all other allowances. Higher-graded positions received relatively higher allowances, hence decompressing overall remuneration and making high-level positions more competitive with the private sector (McLeod 2008:199-200). In theory, such a single allowance would make the system more transparent. In practice, McLeod reports in 2008 that the old allowances continued to exist and schütte Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 that overall pay may have therefore even risen. Unlike the kpk, the Department of Finance-tied by the civil service system-has not been able to recruit for senior positions from outside the department. Despite these caveats, the Department of Finance was widely considered to have undertaken the most advanced and wide-ranging reform efforts by (McLeod, 2008.32 For the kpk, the Department of Finance and its committed minister, Sri Mulyani, were natural partners: the Department of Finance was a laboratory of reform efforts under the existing constraints of the civil service. It also had the authority to decide on the budgets required for reform in other agencies, such as the Supreme Court and the Supreme Audit Board. There was no formal agreement among these organizations, but rather what has been dubbed by one of its initiators an informal 'gentlemen's agreement' .33 When the initiative of the four organizations-and the introduction of performance allowances at the Department of Finance in particular-became public, it was met with sharp criticism by renowned Indonesian economist Faisal Basri and others.34 Faisal Basri was quoted by Kompas as saying: 'If the Department of Finance undertakes reform according to its own model, with the increase of salaries determined on its own and differently from other departments, then this is morally difficult to account for.'35 The non-involvement of the State Ministry of Administrative Reform was another concern. Subsequently, the State Ministry of Administrative Reform became the official coordinator of this loose cooperation and the reform initiative became more formalized. In August 2007, in a speech on the government's budget plans for 2008, President Yudhoyono (2007) announced to the legislature that reform of the bureaucracy had started, with pilot programmes taking place in the Department of Finance, the State Ministry of Administrative Reform, the Supreme Court, and the Supreme Audit Board. A working group was formally established, and a blueprint developed and endorsed by the State Ministry of Administrative Reform in July 2008.36 This blueprint (Pedoman reformasi birokrasi, Guidelines for bureaucratic reform) outlined the components and the sequencing of bureaucratic reform. The list of organizations that are to be prioritized in the reform efforts is as follows: 32 Interview with a kpk director, Jakarta, June 2009. 33 Interview with a former kpk commissioner, Jakarta, June 2009. 34 Kompas, 10-7-2007. 35 Kompas , 10-7-2007. 36 Minister of Administrative Reform Decree 7 of 2008 on the General Guidelines of Bureaucratic Reform.
Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 -First priority is ministries/agencies/local governments tasked with state finances, law enforcement, financial control and oversight, and organization of the state's apparatus. -Second priority is ministries/agencies/local governments tasked with economic activities, production systems, sources of state revenue, and organizational units that serve the people directly (including local governments). -Third priority is ministries/agencies/local governments not listed under (1) and (2)  Instead of changing the overall framework, particularly the regulatory framework for human resource management, an organization-focused approach of pilots was rolled out. For instance, the introduction of performance allowances in addition to the base salary under existing regulatory constraints may be seen as a pragmatic solution from the perspective of one single agency, but it seems hardly adequate for integrated, national reform. In fact, what had been pragmatic became increasingly piecemeal: not all of the 15 government entities that had requested, and were granted, a budget for reform were part of the 'priority list' ,37 nor have they followed the sequence of reform steps recommended. Before performance indicators had been firmly established and an independent evaluation mechanism had been put in place, the introduction of performance allowances by several departments had been problematic. This way it was just another allowance that was handed out, and an important incentive to change behaviour was lost.38

A Grand Design-Without the kpk
Disillusioned with the emerging 'first come, first served' approach and the increase of allowances before performance indicators had been established, the kpk issued a letter to the president in March 2009.39 It pointed out that the Guidelines for bureaucratic reform were not being followed, and that in its implementation the change of the remuneration system had been prioritized, leaving the other elements of reform behind. It also criticized the lack of 37 Interview with a hr management consultant who was closely involved in the drafting of the Guidelines for bureaucratic reform, Jakarta, June 2009. 38 Kompas, 7-3-2009; interview with a hr management consultant who was closely involved in the drafting of the Guidelines for bureaucratic reform, Jakarta, June 2009. 39 Interview with a kpk director, Jakarta, July 2009.
Bijdragen tot de Taal-, Land-en Volkenkunde 171 (2015) 423-454 position auctions for civil servants (lelang jabatan).44 This is very much in line with McLeod's (2005) suggestion that reform of human resource management has to start at the top and work its way down, and that it should entail incumbents having to reapply for their own positions if they want to benefit from the new, much more attractive salary scale. It is now up to Joko Widodo to show the same resourcefulness and persistence in taking on the much larger challenge of national reforms. The recent open call for applications for the position of head of the National Civil Service Agency (MenPAN 2015) raises hopes that reforms are finally reaching the very heart of the civil service.