an ethical map of indonesian MPs

From the SelectedWorks of Riccardo Pelizzo 2008 an ethical map of indonesian MPs riccardo pelizzo bernice ang Available at: https://works.bepress.com/riccardo_pelizzo/22/

An Ethical Map of Indonesian MPs RICCARDO PELIZZO AND BERNICE ANG Abstract Research indicates that the success of legislative ethics reforms, such as codes of ethics and codes of conduct, depends on whether the members of the legislature have homogeneous ethical standards. A survey conducted in the Indonesian legislature shows that the members standards are far from homogenous and are not applied consistently across issues. If there is a link between homogeneity of ethical standards and the success of conduct codes, a code is not likely to be adopted and successfully enforced in Indonesia. While scholars have long been concerned with whether and how far comparativepolitics concepts can travel (Lijphart 1971; Sartori 1970), constitutional experts are concerned with whether institutional solutions that work in some settings under specific conditions will also work in other settings. 1 From the beginning of the third wave of democratization onward (Huntington 1991), political scientists, comparative constitutional scholars, and practitioners have extensively debated not only which electoral system, form of government, or institutional arrangement is most likely to promote the consolidation of democracy, but also which of these solutions fits best in the history, culture, and traditions of the newly established democracies. Such questions are especially relevant in Indonesia, where the ethics councils of the two houses of the national parliament, the Dewan Perwakilan Daerah (DPD) and the Dewan Perwakilan Rakyat (DPR), are currently discussing the adoption of a code of legislative conduct to curb corruption, prevent legislative misconduct, and restore citizen trust in the legislature. 2 Assuming that the members of parliament (MPs) are able to pass a standard, will they be able to make it work? Will they be able to transplant an institutional device peculiar to parliamentarism Westminsterstyle in the Indonesian context? The purpose of this article is to investigate the factor that the literature regards as a key prerequisite for the successful implementation of a code of conduct: whether MPs have homogeneous ethical preferences (Mancuso 1993; Mancuso 1995; Snape and Skelcher 2001). The analyses are performed on data collected by surveying the members of the ethics councils of the DPD and the DPR in 2006. The work of Mancuso (1993, 1995) provides the main theoretical framework for this analysis. The analyses that she performed are replicated using the Indonesian data. As will be seen, Indonesia s MPs fall into four distinct types. Public Integrity, Summer 2008, vol. 10, no. 3, pp. 253 272. 2008 ASPA. All rights reserved. ISSN 1099-9922/2008 $9.50 + 0.00. DOI 10.2753/PIN1099-9922100304

Riccardo Pelizzo and Bernice Ang The Ethical World of Parliamentarians The literature has paid some attention to the adoption, implementation, and success of codes of conduct. It indicates that codes of conduct, originally adopted by private companies and professional associations (Fiorini 2003; Frankel 1989), have subsequently been adopted in the public sector (Bishop and Sampford 2003; Bowman 1990; Bruce 1996) and by legislatures both at the state and the national level (Brien 1999; Burgmann 1998; Jackson and Smith 1995). The literature points out that codes, whether private or public, are used to set standards for individuals, promote ethical behavior, prevent misconduct, and restore citizen confidence in the institution (Gilman 2005). That is exactly why the ethics councils of the DPD and the DPR are considering the adoption of a code of legislative conduct. 3 Researchers have investigated whether such canons are effective (Bowman and Williams 1997; Bruce 1996; Frankel 1989), as well as the conditions under which they are more likely to be successful. In this respect it has been observed that the success of a code depends on the clarity and the enforceability of its provisions (Menzel 2007), the existence of a broader framework of public accountability (Kernot 1998), the presence of sanctions to punish violations (Bruce 1996), the organizational/institutional culture (Longstaff 1994), and on whether the individuals whom the code is expected to regulate have homogeneous ethical standards (Mancuso 1995). For Mancuso (1995), as also for Skelcher and Snape (2001), the success of a code depends on several factors: The individuals to whom the code is applied must have the same attitudes and values. They must have a common understanding of the problems the code is supposed to prevent and possibly eliminate. They must have a common understanding of the appropriate solutions for those problems. If the individuals the code is expected to regulate have heterogeneous standards, they will not be able to develop a common understanding of what the problems are and what solutions may be best, or to draft and, most important, enact a coherent document. Assuming that Skelcher and Snape (2001) are correct in positing a relationship between the homogeneity of standards and successful implementation of the code, it is worthwhile investigating the ethical world of Indonesian MPs to see whether and to what extent their ethical attitudes are sufficiently homogeneous to ensure successful adoption and implementation of a code of conduct. This investigation is particularly salient because the ethics councils of the DPD and DPR are actively promoting the passage of such a document. In order to analyze the attitudes of Indonesian MPs, a survey of the members of the ethics councils of both chambers of the parliament was conducted in January 2006. The survey employed a questionnaire originally devised by Mancuso (1993; 1995) in her study of the ethical world of members of the British Parliament. Mancuso (1993; 1995) asked a sample of 100 MPs to indicate on a seven-point scale how corrupt they considered certain activities (selecting value 1 on the scale 254 PUBLIC INTEGRITY SUMMER 2008

An Ethical Map of Indonesian MPs TABLE 1 Corruption Scenarios Name CAMPAIGN SCHOOL DRIVEWAY CONTRACT RETAINER PLANNING LORRY SECRETARY GIFT KNIGHT TRAVEL ALL PARTY ORDER PAPER PASS Text Cabinet minister promises appointive position in exchange for campaign contribution MP uses position to get friend or relative admitted to Oxford or Cambridge, or some other prestigious institution Driveway of City Council chairman s home is paved by the council s District Works Department Cabinet minister uses his influence to obtain contract for a firm in his constituency MP arranges for meetings between private corporation executives and parliamentarians Local councilor, while chairman of planning committee, authorizes planning permit for property he owns Ministry of Transport truck inspector obtains 10 from owner of a truck to pass it MP hires wife or family member as secretary MP accepts case of wine from influential constituent MP exchanges allocated 1st-class ticket for economy class, pockets difference Nonpartisan group on the aged hires full-time research assistant at the expense of the agency for the aging MP on retainer to a PR firm representing a foreign government submits several written questions for the Order paper on British industrial development in that country MP gets house pass for lobbyist whose services are paid for by an outside source to act as research assistant Source: Mancuso (1995, 34). meant that the action was regarded as very corrupt, value 7 indicated that the action was not corrupt). The British MPs were asked to express their views on fourteen scenarios (see Table 1). Mancuso found that some activities were unanimously or nearly unanimously deemed corrupt. The four scenarios unanimously viewed as corrupt (LORRY, CAMPAIGN, DRIVEWAY, PLANNING) were dropped from the rest of the analysis. In addition, Mancuso also deleted the KNIGHT scenario because it did not represent a clear case of constituency service or of conflict of interest. She used the remaining nine items to assess the MPs attitudes toward constituency service (SCHOOL and CONTRACT) and toward conflict of interest (TRAVEL, PASS, ORDER PAPER, GIFT, RETAINER, ALL PARTY, and SECRETARY). Attitudes toward the conflict of interest scenarios were often strongly and significantly correlated with one another. Mancuso employed these two sets of variables to create two additive scales. The SCHOOL and CONTRACT scenarios were used to create a scale to assess attitudes toward constituency service (SERVICE), and seven scenarios (TRAVEL, PASS, ORDER PAPER, GIFT, RETAINER, ALL PARTY, and SECRETARY) were used to assess attitudes toward conflicts of interest (CONFLICT). In performing PUBLIC INTEGRITY SUMMER 2008 255

Riccardo Pelizzo and Bernice Ang this analysis, Mancuso used the mean as a proxy for the tolerance level on each dimension. By combining the tolerance levels for service and constituency service, she was able to identify four categories of parliamentarians: Puritans: low tolerance for both constituency service and conflict of interest Entrepreneurs: high tolerance on both dimensions Servants: high tolerance for constituency service, but not for conflict of interest Muddlers: high tolerance for conflict of interest, but not for constituency service Mancuso concluded that parliamentarians have very different ethical attitudes and a multiplicity of standards, and thus that assuming homogenous standards is wrong. Last but not least, she also concluded that the multiplicity of standards may prevent ethics reforms and regulations from producing the expected results. Codes of conduct and other initiatives are established and expected to work on the assumption that the individuals they are supposed to regulate share some values and attitudes. Mancuso s work shows that this is an unwarranted assumption. The Indonesian Case To see whether this assumption is also unwarranted in Indonesia, Mancuso s analyses were replicated with data collected from seventeen of the thirty-two members of the DPD ethics council and ten of the seventeen members of the DPR ethics council. The analysis kept eight of the nine scenarios used by Mancuso and added two additional ones: INFLUENCE JOB and COMPANY DONATION. In the INFLUENCE JOB scenario, the MP uses his/her influence to get a friend/relative a job. In the CAM- PAIGN DONATION scenario, a company makes a big donation to the government party and is then awarded a major contract. The list of ten scenarios was adopted because each of them satisfied three criteria: (1) they were all clear instances of either constituency service or conflict of interest; (2) they all clearly involved individual responsibility (it is always an individual who performs the action and not the government or the city council); and (3) they all involve parliamentarians. The ten scenarios were submitted to the members of the two ethics councils, who were asked to rank on a seven-point scale how corrupt they considered each activity. Twenty-four out of the twenty-seven MPs assessed the corruption level of all eight scenarios. Answers 1 through 3 were coded as 1 (corrupt), answers 5 to 7 were coded as 0 (noncorrupt), and answer 4 was treated as system missing. When the responses were recoded in this way, it emerged that there was no unanimous agreement on any scenario (see Table 2). Mancuso s measurement of the tolerance level for specific acts of misconduct used the mean response on each item as a proxy of the tolerance for that act. When this analysis was replicated with the Indonesian data, it emerged that the tolerance level for misconduct is consistently quite low across the eight scenarios. It is well below 3 for six of the eight scenarios, and only in two instances (RETAINER, GIFT) was it higher than 4. In other words, six scenarios are solidly assigned to the corrupt camp, and only two are regarded as noncorrupt. The data in Table 3 sustain the claim that the tolerance level of Indonesian MPs is generally lower than that of their British 256 PUBLIC INTEGRITY SUMMER 2008

An Ethical Map of Indonesian MPs TABLE 2 Scenario Responses and Statistics Name Scenario % corrupt Indonesia % noncorrupt % corrupt UK % noncorrupt Company donation Campaign School Pass Influence job Contract Travel Secretary Retainer Gift Company makes big donation to government party and eventually wins major contract Cabinet minister promises appointive position in exchange for campaign contribution MP uses influence to get friend/relative admitted to prestigious institution MP gets house pass for lobbyist paid by outside source to act as research assistant MP uses influence to get friend/relative a job Cabinet minister uses influence to obtain contract for firm in his constituency MP exchanges allocated 1stclass ticket economy class, pockets the difference MP hires wife or family member as secretary MP arranges for meetings between private corporation executives and parliamentarians MP accepts box of cigars from influential constituent 96.3 3.7 91.7 8.3 98 1 91.7 8.3 61 26 90 10 58 29 88.5 11.5 87.5 12.5 40 40 77.3 22.7 87 9 76 24 2 91 47.8 52.2 20 72 36.8 63.2 31 50 Note: Corrupt equals responses 1, 2,and 3; non-corrupt equals responses 5, 6, and 7, while response 4 is treated as system missing in this analysis of Indonesian data. Indonesian data were collected by the authors, British data taken from Mancuso (1995). counterparts (Mancuso 1995). Indonesian MPs may not be less tolerant of corruption, but at least they say they are. The analysis also reveals that Indonesian MPs disagree as to whether some practices should be regarded as corrupt. This finding is consistent with the notion that there is a multiplicity of standards in the Indonesian parliament. The correlation analysis supports this interpretation. In her correlation matrix, Mancuso correlated seven conflict scenarios (RE- TAINER, SECRETARY, GIFT, TRAVEL, ORDER PAPER, GIFT) with one another and found that eleven of the twenty-one correlation coefficients were statistically significant. This suggested that while the British MPs had different standards, they all used their standards consistently. Attitudes toward a specific case provide a clue as to how a member feels vis-à-vis a similar case of misconduct. In other words, standards provide consistent cognitive guidance and may offer behavior guidance, PUBLIC INTEGRITY SUMMER 2008 257

Riccardo Pelizzo and Bernice Ang TABLE 3 Scenario Responses and Statistics Indonesia Scenario M SD Mdn M SD Mdn School 2.26 1.403 2 3.3 2.3 3 Pass 2.28 1.514 2 3.4 2.2 3 Contract 2.37 1.690 2 4.1 2.3 4 Influence job 2.48 1.578 2 Secretary 2.70 2.053 2 6.6 0.98 7 Travel 2.77 2.065 2 2.0 1.6 1 Retainer 4.48 2.242 4 5.6 2.0 7 Gift 4.63 2.022 4 4.7 2.2 4.5 Source: Indonesian data collected by the authors, British data taken from Mancuso (1995). UK although neither Mancuso s data nor the Indonesian data provide any evidence in this respect. The analysis of the Indonesian data tells a fairly different story (see Table 4). When the eight scenarios were correlated with each other (RETAINER, SCHOOL, INFLUENCE JOB, CONTRACT, SECRETARY, GIFT, TRAVEL, PASS), the correlation coefficient was statistically significant only in six cases out of twenty-eight. This finding shows that the ethical world of Indonesian MPs is substantially different from that of British MPs in that standards do not provide as much guidance, cognitive or otherwise, as in the UK. Indonesian MPs have different ethical values and standards, and apply them inconsistently. This means that the application of principles varies not only from MP to MP but also for the same MPs from issue to issue. Knowing where an MP stands on one ethical issue does not provide any indication whatsoever as to what his or her position will be on a similar issue. How can this interpretation of the correlation matrix be tested? If the standards of individual MPs vary from issue to issue, then the tolerance level of a given MP for one item should not be related to the tolerance level for a different item. In order to test whether this is the case, responses were recoded in the following way: Having established that the mean response from the sample represents the tolerance level for a given scenario, individual responses were coded as 1 if an MP had low tolerance (i.e., if the MP s response on a given scenario was below the sample mean) and were coded as 0 if an MP had high tolerance (i.e., if the MP s response on a given scenario was above the sample mean). Gamma correlations were performed to assess whether or not low tolerance for a specific scenario was related to low tolerance on other scenarios. These analyses were performed on the assumption that standards are employed consistently if low tolerance for an item is associated with low tolerance for other items, but are not used consistently when no relationship can be assessed between the tolerance levels vis-à-vis different scenarios. The results of the gamma correlations show that overall ethical principles are not used consistently (see Table 5). Only two of the twenty-eight gamma correlation coefficients are statistically significant. Hence, it is safe to argue that low tolerance on a given issue has little bearing on how tolerant an MP will be toward a different issue. Statistical analyses were also performed to see whether tolerance for various sce- 258 PUBLIC INTEGRITY SUMMER 2008

An Ethical Map of Indonesian MPs TABLE 4 Pearson Correlations Matrix for CONFLICT Scenarios Scenario Retainer Influence Influence job Contract Secretary Gift Travel Pass Retainer 1 0.102 0.204 0.164 0.441* 0.448* 0.187 0.364 School 1 0.549* 0.217 0.094 0.089 0.157 0.459* Influence job 1 0.392* 0.271 0.347 0.015 0.398* Contract 1 0.144 0.015 0.226 0.560** Secretary 1 0.260 0.319 0.181 Gift 1 0.009 0.018 Travel 1 0.220 Pass 1 *Significant at the p < 0.05 level; **significant at the p < 0.01 level. PUBLIC INTEGRITY SUMMER 2008 259

Riccardo Pelizzo and Bernice Ang TABLE 5 Gamma Correlation Matrix for Low Tolerance on CONFLICT Scenarios (sig.) Scenario Retainer School Influence job Contract Secretary Gift Travel Pass Retainer 1 0.103 (0.829) 0.692 (0.087) 0.103 (0.829) School 1 0.857 0.660 0.100 0.077 0.250 0.412 (0.001) (0.055) (0.808) (0.881) (0.531) (0.329) Influence job 1 0.467 0.176 0.600 0.111 0.507 (0.206) (0.656) (0.189) (0.781) (0.197) Contract 1 0.100 0.077 0.067 0.688 (0.808) (0.881) (0.873) (0.062) Secretary 1 0.077 0.535 0.180 (0.881) (0.141) (0.692) Gift 1 0.569 0.366 (0.221) (0.481) Travel 1 0.419 (0.329) Pass 1 0.579 (0.187) 0.081 (0.892) 0.667 (0.106) 1.00 (0.006) 260 PUBLIC INTEGRITY SUMMER 2008

An Ethical Map of Indonesian MPs narios was related to institutional affiliation. Specifically, the analyses tested whether low tolerance for various types of misconduct was a function of the respondent s membership in one or the other of the two ethics councils, that is, either DPD or DPR. Institutional membership was coded as a dummy variable, taking value 1 if the respondent was a member of the DPD ethics council and taking value 0 otherwise. There was no detectable statistical relationship between institutional membership and tolerance toward corruption and misconduct (see Table 6). This suggests that there was more intra-institutional disagreement (i.e., between members of the same body) than inter-institutional (i.e., between members of one body and members of the other). A Typology of Indonesian MPs The evidence presented so far illustrates that the ethical standards, preferences, and values of members of the Indonesian parliament are not homogenous. 4 While this situation is similar to what Mancuso found, the UK case differs from the Indonesian one in two major respects. First, members of the ethics councils of the DPD and the DPR have lower tolerance for misconduct than the British MPs. Second, ethical standards provide, in general, much less cognitive guidance in the Indonesian parliament than in Westminster. Data analyses also revealed that ethical attitudes cannot be reduced to institutional affiliation that is, to whether an MP is a member of the DPD or the DPR. One of Mancuso s interesting findings was that the differences in ethical standards made it possible to categorize distinct types of parliamentarians. Each type was identified depending on the level of tolerance an MP had for constituency service and for conflict of interests. It is worth investigating whether Mancuso s typology can be employed in the Indonesian case. In order to do so, it is necessary to proceed in the following way: After constructing two additive scales for SERVICE and CONFLICT (of interest), the tolerance level on each dimension can be estimated by calculating the mean response. The sample is then split into two subsamples (tolerant and intolerant) on each dimension depending on whether responses are below or above average. A fourfold classification is finally obtained by combining these two divisions. Misconduct scenarios belonged either to the general category of constituency service activities (a member is trying to deliver some goods to his/her district or to people in the district) or the general category of conflict of interests activities (the member is using his/her power to do something from which the member or a relative may receive a benefit). In order to capture both facets of misconduct, two distinct additive scales are created, SERVICE and CONFLICT. The SERVICE scale is constructed by combining SCHOOL, INFLUENCE JOB, and CONTRACT. The SERVICE variable so constructed has a mean of 2.37 and a standard deviation of 1.19. The CONFLICT scale is created by combining TRAVEL, PASS, GIFT, RETAINER, and SECRETARY. This variable has a mean of 3.25 and a standard deviation of 1.18. Two considerations are in order at this point. First, the tolerance level on these two dimensions is significantly lower than the tolerance level recorded by Mancuso (1995, 48). Second, Indonesian MPs, just like the British MPs, display higher levels of tolerance for conflict of interest scenarios than for constituency service scenarios. This PUBLIC INTEGRITY SUMMER 2008 261

Riccardo Pelizzo and Bernice Ang TABLE 6 Gamma Correlation Matrix for Tolerance and Institutional Affiliation Scenario Retainer School Influence job Contract Secretary Gift Travel Pass Membership in DPD ethics council 0.103 (0.829) 0.412 (0.289) 0.176 (0.656) 0.241 (0.552) 0.561 (0.127) 0.525 (0.271) 0.535 (0.141) 0.294 (0.489) 262 PUBLIC INTEGRITY SUMMER 2008

An Ethical Map of Indonesian MPs Figure 1. Ethical Locations of Sample means that MPs are more willing to tolerate misconduct activities from which they can benefit than misconduct activities that would be beneficial to their constituents. Having estimated the mean on both dimensions, a proxy variable for tolerance can be generated. All MPs who score lower than 2.37 in the SERVICE dimension have low tolerance for service, and conversely those who score higher than 2.37 in the SERVICE dimension have high tolerance for service. Similarly, MPs who score lower than 3.25 on the CONFLICT dimension have low tolerance, while those scoring higher than 3.25 have high tolerance. The combination of these two divisions generates the fourfold typology presented in Figure 1. Some words of explanation are in order. The sample included twenty-seven respondents, twenty-four of whom assessed the level of corruption of all the service and conflict scenarios. The responses were used to create the scales, measure the tolerance levels, and create the ethical locations of MPs, while the other three MPs were treated as system missing. Of the twenty-four valid cases, eight fall in the Puritan category (low levels of tolerance for both service and conflict). Six cases fall in the Muddlers category (low tolerance for service but high tolerance for conflict of interest). Five cases fall in the Servant category (high tolerance for service and low tolerance for conflict). Five cases fall in the Entrepreneur category (high tolerance on both dimensions). PUBLIC INTEGRITY SUMMER 2008 263

Riccardo Pelizzo and Bernice Ang Figure 2. Ethical Locations of Members of DPR Ethics Council A Comparison of the Two Chambers The analysis can be taken a step further by investigating whether the ethical locations of members of the DPD and DPR councils are substantially similar by computing the tolerance levels of each chamber. For MPs in the DPR, the mean score on the SERVICE dimension was 2.33, while the mean score on the CONFLICT dimension was 3.28. As for the MPs in the DPD, the mean score on the SERVICE dimension was 2.39, while the mean score on the CONFLICT dimension was 3.22. This means that the tolerance for SERVICE is higher in the DPD than in the DPR, while the tolerance for CONFLICT is higher in the DPR than in the DPD. Having measured the tolerance levels, the ethical locations of the members of the two councils can be estimated. Figures 2 and 3 highlight several distinctive features of each chamber. First, the attitudes of DPR respondents are not too different from those of the DPD respondents. The Puritans are the single largest group in the DPR, while the Muddlers are the largest group in the DPD. Moreover, the ethical map of Indonesian MPs reveals that in the ethics council of the DPR, in addition to four Puritans, there is a balanced mix of three Servants, two Entrepreneurs, and one Muddler. This is not the case in the DPD. In fact, in the DPD there are four Puritans, two Servants, and three Entrepreneurs but only if the three system missing respondents (who rightfully belong to the Entrepreneur typology) are ignored. Should these three candidates be considered, the Entrepreneurs would represent the largest group in the DPD. Regardless of whether these three missing observations are considered, there is virtually no 264 PUBLIC INTEGRITY SUMMER 2008

An Ethical Map of Indonesian MPs Figure 3. Ethical Locations of Members of DPD Ethics Council difference in the fractionalization of the two ethics councils. Given the distribution of preferences among the members of the DPR ethics council, the fractionalization equals.77. If the three system-missing cases are discounted, the DPD ethics council has a fractionalization of 0.78, whereas if the three system-missing cases are counted in, the fractionalization takes a value of 0.765. 5 This evidence sustains the claim that there is no major difference between the two chambers in regard to distribution of opinion, and therefore that the ethical attitudes are not a function of institutional affiliation. A comparison between the means on the CONFLICT dimension shows that DPD members have lower average tolerance levels (3.22) than DPR members (3.28). When the three missing respondents who are assumed to be Entrepreneurs are included in the analysis, there are ten DPD members with high tolerance levels for CONFLICT situations, whereas there are only three in the DPR. This result is worrisome, because tolerance for conflicts of interest may prevent ethics committees from performing their tasks effectively and enacting the reforms necessary to improve the standards of parliamentarians and the quality of governance. Heterogeneity in Ethics Standards and Prospects for Reform The existence of these four types of MPs is of some importance not only because it shows that not all MPs think alike, but also, and more important, because these differences in ethical standards may present a major problem in crafting and imple- PUBLIC INTEGRITY SUMMER 2008 265

Riccardo Pelizzo and Bernice Ang TABLE 7 Statements on Corruption by Puritans and Entrepreneurs Puritans Entrepreneurs Political corruption is a widespread problem People don t trust politicians because they do not understand politics Corruption reflects the standard of society 2.0 3.6 4.37 2.4 3.12 2.6 Note: The data are the mean responses provided by the Puritans and the Entrepreneurs on these questions. Responses were provided on a 7-point scale, where 1 indicated strong agreement, 7 indicated strong disagreement, and 4 indicated a neutral position. menting reforms. This point can be illustrated by comparing the general statements on corruption of the Puritans with those of the Entrepreneurs (see Table 7). While most of the Puritans agree with the statement that corruption represents a major problem, most Entrepreneurs took a neutral stance. However when respondents were asked to say what they meant by political corruption, they gave very different answers depending on which group they belonged to. For Puritans and Servants, corruption amounts to using power to further either personal or party interests. For Entrepreneurs, it concerns failing to adequately perform legislative functions and lying to the public (a point also emphasized by the Muddlers). With one exception, all the respondents reported that they have at some point or another suspected that corruption occurs in parliament. 6 Second, while Puritans and Entrepreneurs from both chambers agree that corruption reflects the moral standards of society, Entrepreneurs are more supportive of this view than Puritans. Third, and this is what represents the most significant difference, Entrepreneurs agree with the idea that citizens do not trust politicians because they do not understand what politics is all about, while Puritans disagree. What the Entrepreneurs are saying is that while it is true that there is some corruption, it is not as widespread as Puritans believe, and it is not the reason why citizens have little trust in political institutions. The lack of trust, according to the Entrepreneurs, has nothing to do with corruption but is instead due to the fact that people do not understand politics. By making this claim, the Entrepreneurs are saying that corruption (like other forms of misconduct) is not responsible for the lack of trust, and therefore the enactment of reforms and the adoption of a code of conduct will not be very useful in restoring public confidence and trust. If the Entrepreneurs were right in saying that the sources and causes of corruption are exogenous, then the Entrepreneurs should also conclude that the adoption of a code is not terribly useful because it would not be able to perform either its external or its internal function. Since there is no relationship between corruption and misconduct on the one hand, and trust on the other, the adoption of a code may not be an adequate step to convince the citizenry of the parliament s commitment to higher moral standards, honesty in government, and good governance. The Entrepreneurs should conclude that under these circumstances the code would fail to perform its external function of restoring citizen confidence. 266 PUBLIC INTEGRITY SUMMER 2008

An Ethical Map of Indonesian MPs Given their assumption about the causes of corruption, Entrepreneurs should think, as well, that a code of conduct would also fail to perform its internal function of curbing/preventing corruption in parliament because corruption has little to do with politicians, but is instead the product of customs, traditions, and values. Hence, if the Entrepreneurs developed their thinking consistently and coherently, they should conclude that given the societal standards, as well as the nature of corruption, a legislative code of conduct is bound to fail and should probably not even be adopted. 7 However, this is not what the Entrepreneurs said in the course of the interviews. All the respondents, including Entrepreneurs, argued that a code should be adopted. It is also interesting to note that six of the nine members of the DPR ethics council said that they never face ethical dilemmas while performing their legislative functions. If codes are good in preventing dilemmas, but members of the council never personally face dilemmas, one wonders whether the code of conduct should be adopted to guide Six of the nine members of the DPR ethics council said that they never face ethical dilemmas while performing their legislative functions. the actions of their fellow parliamentarians but not their own. The members of the DPR council unanimously agreed that a standard should be adopted not only for parliamentarians but also for staff members. In addition, some of members of the councils in both chambers made it clear that they entirely opposed the notion of adopting certain reforms, such as, for example, disclosure rules or the adoption of a register of interests. One member of the DPR ethics council challenged the allegation that the political system is corrupt and suggested that there is no need to make institutional changes to fix a problem that actually does not exist. 8 The same MP blamed the media for their negative attitude and suggested that press coverage is the only reason the DPD and DPR are considered to be corrupt by citizens and international observers. The opinion of this MP, who was an Entrepreneur, was shared by others, including the Puritans. In fact, one Puritan suggested that if an inquiry committee were created to investigate corruption, the investigation should begin with the presidency, which, according to the members of the DPR ethics council, is more corrupt than the legislature. 9 Several members of the council were quite perplexed when informed by Transparency International that the legislature is perceived as the single most corrupt institution in the country. The reason they were surprised is quite simple. The DPD has very little power, and therefore its members have few opportunities to be corrupted. Why would anybody, any citizen, want to corrupt a powerless elected official? The DPR is a more powerful institution than the DPD, but it is not nearly as powerful as the presidency. 10 In fact, several members of the DPR ethics council suggested that a properly conducted investigation would find that the government has a remarkably higher level of corruption than the legislature. To some extent the bad reputation of the legislature, and of the DPR in particular, is due not so much to wrongdoing by the DPR but to the behavior of the political parties. Indonesian parties seem unable to understand their role in a democratic society and in competitive elections. In Indonesia s predemocratic era, decisions were endorsed unanimously through inter-party agreement. 11 Since the democratic transition in 1998, the three main PUBLIC INTEGRITY SUMMER 2008 267

Riccardo Pelizzo and Bernice Ang political parties have retained the unanimity principle. As a result, ordinary citizens do not see any difference between the politics of the newly established democratic regime and what took place in the years of authoritarian rule. 12 In addition, they seem to view patterns of inter-party competition, or rather the lack thereof, as a sign of inter-party collusion. Hence, the political parties have successfully reached what Katz and Mair (1995, 5 28) defined as cartel party democracy a democracy that is, or is perceived to be, imperfectly representative, responsive, accountable, and democratic. What the members of two ethics councils seemed unable to grasp was that Transparency International s Corruption Perception Index and Global Corruption Barometer measure the level of perceived The legislature enjoys little public corruption, not objective corruption. And, of course, an institution the legislature in confidence, is regarded as extremely this case does not have to be more corrupt corrupt, and needs to take steps, such than any other in the country to be perceived as the adoption of a code of conduct, as the most corrupt. 13 to curb corruption and regain citizen But these caveats are somewhat beside the confidence. point. The Indonesian legislature enjoys little public confidence, is regarded as extremely corrupt, and needs to take steps, such as the adoption of a code of conduct, to curb corruption and regain citizen confidence. And insofar as ethical homogeneity is a prerequisite for the successful adoption and implementation of a code, the heterogeneity in the standards of its members is not terribly auspicious. The fact that ethical standards are applied inconsistently does not bode well for the adoption or the enforcement of a code of conduct. Conclusion Indonesian MPs have lower levels of tolerance for corruption and other forms of misconduct than their British counterparts. Like them, Indonesian parliamentarians have very different standards. Actions or scenarios regarded as very corrupt by some parliamentarians are seen as perfectly legitimate by others. Depending on their level of tolerance toward service and conflict activities, Indonesian MPs, like their British counterparts, can be classified as either Puritans (low tolerance for service and conflict scenarios), Entrepreneurs (high tolerance for conflict and service scenarios), Muddlers (high tolerance for conflict scenarios but low tolerance for service), or Servants (high tolerance for service and low for conflict). Indonesian MPs have heterogeneous standards and, in contrast to their British counterparts, use their standards inconsistently. In light of this, the inconsistent application of heterogeneous standards is the major obstacle for the adoption and effective enforcement of a code of legislative conduct in Indonesia. This conclusion has important implications both theoretical and practical. Scholars have argued that the success of a code depends on the ethical conditions (Burgmann 1998; Mancuso 1993 and 1995; Pasquino and Pelizzo 2006; Skelcher and Snape 2001). Previous studies pointed out that a code has to be applied to individuals with homogenous standards to be successfully implemented. This article suggests that the success of a code depends not only on whether standards are homogeneous but also on whether they are used consistently. This conclusion is of theoretical interest, not 268 PUBLIC INTEGRITY SUMMER 2008

An Ethical Map of Indonesian MPs only because it suggests new ways in which ethical conditions may affect the fortunes of a code of conduct, but also because it opens up new questions for research: Are the ethical conditions really important for the success of a code? Can a properly drafted document help to homogenize standards and thus create the conditions for its own success? How can the consistent application of ethical principles in the administration of a code be promoted? Under what conditions does a code work best? These questions, however important at the theoretical level, are even more significant at the practical level. For some developing countries with a history of corruption and democratic instability, the effective administration of a code may represent a way to improve governance, preserve citizen trust, and make democracy survive an issue not just relevant for Indonesia. Political commentators in Fiji suggested, before the military takeover, that the adoption of a code of conduct was the only solution to curb corruption and save the democratic regime (Bolatiki 2006). This advice was neglected and little to no effort was made to fight corruption. As a result, the democratic regime lost its legitimacy and was eventually overthrown by a military coup in 2006. In Thailand, where democracy also collapsed in 2006, political reformers have recognized the importance of establishing codes of conduct and other institutional mechanisms that fight corruption as a way to make democracy work better and survive. As these examples illustrate, there is some awareness in Thailand and Fiji that a code of legislative conduct can play a key role in consolidating democracy which is why it is likely that both countries will adopt such a code. 14 If they do, they may want to know under what conditions a code works best. This analysis does not provide an answer to the question, but it offers new research venues that point to where the answer may be found. NOTES 1. The debate generated by Linz (1994) on the failure of presidential systems is a perfect case in point. Contra Linz, who argued that presidential systems are bound to fail because of their constitutionally designed problems (rigidity and dual legitimacy), other scholars noted that presidential democracies fail only when the presidential form of government is coupled with hyper-fragmentation at the party-system level (Mainwaring 1993). For Mainwaring, this means that institutional success or failure is conditional. It depends on whether the country has the proper conditions for the institution to work. 2. The DPR, the lower house, adopted a code of ethics on September 29, 2004, that became part of its standing orders. The DPR and DPD are both now trying to adopt a code of conduct to regulate the behavior of their members, but neither house has yet done so. 3. The ethics councils are standing committees in charge of monitoring and possibly sanctioning the behavior of MPs. The DPR, the lower chamber of the Indonesian parliament, was institutionalized by the 1945 constitution. The DPD, a chamber of regional representation, was created in 2001. 4. This is consistent with what Mancuso (1993; 1995) found in the British case and what Jackson and Smith (1995) found in New South Wales (Australia). 5. The fractionalization of the ethics councils in the DPD and the DPR is measured on the basis of Rae s index of fractionalization (Rae 1967), which is defined as:. PUBLIC INTEGRITY SUMMER 2008 269

Riccardo Pelizzo and Bernice Ang 1 F = 1 n( n 1) N i= 1 f i f ( 1) i 6. These data were collected only from the members of the ethics council of the DPR. 7. As discussed later, an MP said that it was a good idea to adopt a code of conduct even though there were no ethical problems. 8. This member was nonetheless in favor of adopting a code of conduct to prevent ethical dilemmas. 9. The members of the DPR ethics council were asked the following question: If a committee were created to examine corruption in government, where should its investigation begin? Some Puritans and some Entrepreneurs mentioned the presidency. Other Puritans, along with some Servants, named the police and the judiciary. Two Servants stated that the investigation of corruption should start in the monetary authority. 10. Schneier (2004, 31) noted that the DPR had a historic and continuing disinclination to legislate. 11. Although in the early 1950s there was an initial momentum toward democratic governance, authoritarian dynamics gradually prevailed over the fragile young Indonesian democracy, so that by the late 1950s Indonesia had become an authoritarian regime under its first president, Sukarno. The nondemocratic New Order regime was further consolidated under President Suharto in the mid-1960s. Having eliminated former Sukarno and the Indonesian Communist Party (PKI), Suharto engineered the creation of Golkar, a progovernment political party based on bureaucratic and military interests. During his thirty-one years in power (1967 1998), Suharto created a patrimonial political system that rewarded his supporters, including family and friends, at the expense of his opponents and the public at large. 12. There are, of course, some differences between the newly established Indonesian democracy and the former authoritarian regime. As Edward Schneier (2004) points out, constitutional amendments and rules changes adopted since 1999 have resulted in a significant shift of authority from the executive to the legislature. 13. The literature has long acknowledged, however, that corruption is a pervasive problem in Indonesia (Schneier 2006, 59). 14. Elections in Fiji are scheduled for the first quarter of 2009 (Bolatiki 2007). In Thailand the new constitution was approved by a referendum in 2007 and new general elections were expected to be held in December 2007.<author, please update; also update text if appropriate> REFERENCES Bishop, Patrick, and Charles Sampford. 2003. Management Organisation and Ethics in the Public Sector, London: Ashgate. Bolatiki, Maika. 2006. Code of Conduct for MPs. Fiji Sun, June 30.. 2007. Positive Reaction Changes Tune. Fiji Sun, June 22. Bowman, James S. 1990. Ethics in Government: A National Survey of Public Administrators. Public Administration Review 50, no. 3:345 353. Bowman, James S. and Russell L. Williams. 1997. Ethics in Government: From a Winter of Despair to a Spring of Hope. Public Administration Review 57, no. 6:517 526. Brien, Andrew 1999. A Code of Conduct for Parliamentarians? Research Paper 2. Canberra: Parliamentary Library of Australia. Bruce, Willa. 1996. Codes of Ethics and Codes of Conduct: Perceived Contribution to the Practice of Ethics in Local Government. In Public Integrity Annual, pp. 23 29. Lexington, Ky.: Council of State Governments. Burgmann, Meredith. 1998. Constructing Codes. Pitfalls and Challenges. In Ethics 270 PUBLIC INTEGRITY SUMMER 2008

An Ethical Map of Indonesian MPs and Political Practice: Perspective on Legislative Ethics, edited by Noel Preston and Charles Sampford, with C-A Bois, pp. 118 126. London, Routledge. Fiorini, Ann. 2003. Business and Global Governance: The Growing Role of Corporate Codes of Conduct. Brookings Review 2:4 8. Frankel, Mark. 1989. Professional Codes: Why, How and with What Impact? Journal of Business Ethics 8, nos. 2 3:109 115. Gilman, S. C. <1 st name?>2005. Ethics Codes and Codes of Conduct as Tools for Promoting an Ethical and Professional Public Service: Comparative Successes and Lessons. Research paper. Washington, D.C.: World Bank. Huntington, Samuel P. 1991. The Third Wave of Democratization. London: University of Oklahoma Press. Jackson, Michael, and Rodney Smith. 1995. Everyone s Doing It! Codes of Ethics and New South Wales Parliamentarians Perceptions of Corruption. Australian Journal of Public Administration 54, no. 4:483 494. Katz, Richard S., and Peter Mair. 1995. Changing Models of Party Organization: The Emergence of the Cartel Party. Party Politics 1, no. 1:5 28. Kernot, Cheryl. 1998. Codes and Their Enforcement: Necessary But Not Sufficient for Ethical Conduct. In Ethics and Political Practice. Perspective on Legislative Ethics, edited by Noel Preston and Charles Sampford, with C-A Bois, pp. 134 142. London, Routledge. Lijphart, Arend. 1971. Comparative Politics and Comparative Method. American Political Science Review 65, no. 3:682 693. Linz, Juan J. 1994. Presidential or Parliamentary: Does It Make a Difference? In The Failure of Presidential Democracy, edited by Juan J. Linz and Arturo Valenzuela, pp. 3 91. Baltimore: Johns Hopkins University Press. Longstaff, Simon. 1994. Why Codes Fail: And Some Thoughts About How to Make Them Work. In Ethics for the Public Sector, edited by Noel Preston, pp. 237 246. Annandale, New South Wales: Federation Press. Mainwaring, Scott. 1993. Presidentialism, Multipartism and Democracy: The Difficult Combination. Comparative Political Studies 26, no. 2:198 228. Mancuso, Maureen. 1993. Ethical Attitudes of British MPs. Parliamentary Affairs 46, no. 2:179 191.. 1995. The Ethical World of British MPs. Quebec city: McGill-Queen s University Press. Menzel, Donald C. 2007. Ethics Management for Public Administrators: Building Organizations of Integrity. Armonk, N.Y.: Sharpe. Pasquino, Gianfranco, and Riccardo Pelizzo. 2006. Parlamenti Democratici. Bologna: Mulino. Rae, Douglas. 1967. The Political Consequences of Electoral Laws. New Haven: Yale University Press. Sartori, Giovanni. 1970. Concept Misformation and Comparative Politics. American Political Science Review 64, no. 4:1033 1053. Schneier, Edward. 2004. Emerging Patterns of Legislative Oversight in Indonesia. In Trends in Legislative Oversight, edited by Riccardo Pelizzo, Rick Stapenhurst, and David Olson, pp. 22 32. Washington D.C.: World Bank.. 2006. Crafting Constitutional Democracies: The Politics of Institutional Design. New York: Rowman & Littlefield. Skelcher, Chris, and Stephanie Snape. 2001. Ethics and Local Councillors: Modernising Standards of Conduct. Parliamentary Affairs 54, no.1:72 78. ABOUT THE AUTHORS Riccardo Pelizzo is a research fellow in the Griffith Asia Institute and the Centre for Governance and Public Policy at Griffith University. He has a laurea in philosophy from the University of Bologna (1995), a graduate diploma in international affairs from the PUBLIC INTEGRITY SUMMER 2008 271

Riccardo Pelizzo and Bernice Ang Bologna Center of the School of Advanced International Studies (1997), and an M.A. (2002) and Ph.D. (2004) in political science from Johns Hopkins University. His books include The Role of Parliament in Curbing Corruption (2006); his articles on parties and legislative affairs have appeared in Finance and Development, Legislative Studies Quarterly, Party Politics, Policy and Politics, and West European Politics. Bernice Ang has a B.A. from Singapore Management University (2008). 272 PUBLIC INTEGRITY SUMMER 2008