Ana səhifə

Review of Experience


Yüklə 0.69 Mb.
səhifə6/9
tarix25.06.2016
ölçüsü0.69 Mb.
1   2   3   4   5   6   7   8   9

Total Investigation Papers opened, by type of offense, 1998-9

Type of Offence

1998

1999

Prevention of Corruption Act 1961

7

2

Anti-Corruption Act 1997

481

406

Emergency (Essential Power) Ordinance No. 22 of 1970

5

1

Penal Code

1

4

TOTAL:

494

413

Source: ACA Annual Report 1999, Table 4 (pp. 26-27).
Comparison of Arrests by category of person, 1998-9

Category

1998

1999

Management & Professional

15

18

Officials/Support Staff

182

136

Private Individuals/Private Sector

103

127

Politicians




2

TOTAL

300

283

Source: ACA Annual Report 1998, Table 8 (p. 45); ACA Annual Report 1999, Table 5 (p. 27).
Comparison of Trial Decisions of ACA Cases for the year 1998 and 1999

Year

Convicted

Acquitted and Discharged

Discharged not Amounting to Acquittal

Charge Withdrawn

Total

1998

114 (61%)

59 (32%)

14 (7%)

-

187 (100%)

1999

89 (59%)

56 (37%)

7 (4%)

-

152 (100%)

The ACA has clearly been active in fulfilling its mandate in terms of investigations, and it has gained a reputation for professionalism. However, there are lingering concerns about its independence from political intrusion. Some view the agency as a tool of government, pointing out that its efforts focus on small rather than big “fish’ – with the agency countering that it has in fact investigated ministers, and that elite “businessmen-politicians” are in any case difficult to catch if they engage in corruption. One hopeful indication of ACA’s accountability is that complaints about its staff can be taken to a separate Public Complaints Bureau. (Langseth 2001)


Tanzania
The PCB, like other ACAs, has internal and external evaluation systems. The internal process involves an evaluation committee within the Office of the President, chaired by the Chief Secretary and including the Director General and other senior officials of the PCB as members. The committee reviews PCB’s performance, and transmits its findings and directives to all Directors and regional offices. PCB is also subject to external scrutiny by the Auditor General. The Bureau publishes an annual report, and is also subject to public evaluation against the benchmarks established by the National Anti-Corruption Strategy and Action Plan. For example, two Tanzanian NGOs have been producing an Annual Report on the State of Corruption in Tanzania.
In its annual reports, the Bureau has noted an increase in reported corruption cases, from 432 in 1998 to 1461 at the end of 2000. These include reports from government departments, parastatals, political parties, and the private sector. As of March 2001 there were 103 cases in different Courts around the country. During 2001 there were eight convictions, one acquittal, and eleven withdrawals in court cases arising from PCB investigations. The table below indicates the number of cases investigated and action taken from different sectors. Unfortunately, PCB statistics do not provide the total breakdown of “closed” cases into successful and unsuccessful ones – but judging by the number of convictions obtained in 2001, the overall (1995-2000) success rate is likely to be exceedingly small.
PCB: Reported Complaints and Steps Taken 1995 – 2000


YEAR

PUBLIC

PRIVATE

POLITICAL PARTIES

UNDER INVESTIG.

CLOSED

1995

1996


1997

1998


1999

2000


217

390


418

325


950

1311


23

45

82



104

131


88

16

6

3



3

7

62



168

291


322

352


1014

1128


37

127


131

80

74



328

TOTAL

3611

473

97

3275

777

SOURCE: PCB Department of Research, Control and Statistics, 2001
The State of Corruption report researchers have also looked at PCB’s performance to date against the targets set in the NACSAP. They found, for example, that PCB recruited 217 new employees, falling short of its planned 300 due to stringent criteria and screening. PCB met its internal training and transparency goals. PCB also aimed to secure revisions of the Prevention of Corruption Act, 1971, so that it could prosecute cases without prior approval by the DPP – it secured only a partial revision allowing it to prosecute under section 3 of the Act. On public awareness, the researchers report the following outputs: 12 public meetings, 225 seminars, 157 radio programs, and 48,800 brochures and leaflets. PCB was to have put in place a code of ethics for its investigators, but has managed only to produce a draft. Of most concern to the researchers was the rate of case disposal and prosecution at PCB. For the period 1995-2000, some 3% of cases were reportedly prosecuted and 91% were still under investigation. However, this record is not necessarily attributable to PCB – the agency’s structural limitations and the need to secure cooperation from other departments plays a major role in this outcome. On the whole, the researchers reported that PCB is not perceived as being either independent or successful in its mission.24
Conclusion
What do these stories tell us? Zooming in to this depth of description and analysis affords us a better sense of the choices facing policymakers in setting the parameters for these agencies – and of the consequences of those choices. The cases presented here also provide a helpful contrast with the most prominent examples in Part 3. In particular, exogenous conditions play a dramatically important role in agency performance – the macroeconomy in the case of Argentina, the political environment and the quality of complementary institutions in the case of Tanzania. In Malaysia, these conditions have been highly supportive in the past two decades, and this shows in the ACA’s results. This supports the notion that the successes of Hong Kong and Singapore are exceedingly hard to reproduce. Both astute internal design and conducive external conditions are needed to make success possible. We will address these points in more detail in the next part.

5. Assessment and Conclusion


What are we to make of the experiences of anti-corruption agencies that we have just reviewed? What can we say overall about the value of these agencies? About whether and how to set them up? What do we know, what don’t we know, and what must we do to know more – and make better choices in this area?
Here, we bring together the threads of the various cases discussed in Parts 3 and 4 for a final assessment. First, we examine the results to see which agencies have performed best, and we review their experiences to determine which of the proposed success factors have the most explanatory power. Next, we suggest some broader lessons to be learned from these cases, and last, we make some recommendations concerning policy and further research.
Explaining Success and Failure in the Mission
How well did the agencies reviewed in this paper perform in terms of their mission and goals – and why? Enabling statutes or the ACA’s own public information will usually describe the mission as one of preventing and deterring corruption through several types of activities. As mentioned in part 2 above, the activities fall into the following categories:


  • Receive and respond to complaints

  • Intelligence, monitoring, and investigation

  • Prosecutions and administrative orders

  • Preventive research, analysis, and technical assistance

  • Ethics policy guidance, compliance review, and scrutiny of asset declarations

  • Public information, education, and outreach.

Most of the ACAs reviewed here report some data on their performance in these areas. A few of them (e.g. the Hong Kong ICAC and the U.S. OGE) publish benchmarks against which their performance can be measured.


Performance Measures:
How do the ACAs compare on performance of their missions? Comparative performance data are summarized in Table 10 (Annex 1). It is difficult to base any analysis on these data alone, due to their incompleteness and lack of uniformity. The contrast between the two premier agencies, the Hong Kong ICAC and Singapore’s CPIB, is instructive. ICAC reports comprehensively on its outputs, outcomes, and efficiency – while CPIB only provides its efficiency benchmarks, without reporting any performance numbers. Comparing the Hong Kong ICAC with its counterpart in New South Wales illustrates the problem posed by the lack of uniformity. Even though the agencies share almost a common blueprint, they take vastly different approaches to case selection. The Hong Kong ICAC pursues every case that comes to its attention, while the NSW ICAC refers most cases to other agencies, selecting only a few strategic ones to pursue. Thus, the numbers on investigations completed, prosecutions, etc. are not comparable. Also, the NSW ICAC reports on the percentage of its studies and recommendations that have been considered or adopted, while its Hong Kong counterpart reports only the total number of studies for the year. Worse still, agencies such as the U.S. OGE, the French SCPC, and India’s CVC have dramatically different mandates from the ICAC-model agencies, and so cannot meaningfully be compared.
In light of the above, what can we say by way of comparing the performance of the anti-corruption agencies reviewed above? We can, first, compare the agencies in terms of performance of the six core tasks listed above. Any conclusions to be drawn from this will need to take into account the different approaches, contexts, and data reporting across countries. Secondly, we can make a broader qualitative assessment of the agencies’ performance, looking at their reputations, the quality of their outputs, and the overall professionalism and cohesion of their activities.
Here is how the agencies compare on performance in the six task areas (numbers on an annual basis, various years):
Receive and respond to complaints: The Philippines (9739), followed by India (total cases received: 5762), had the largest intake in terms of raw numbers. However, on a per capita basis – and given the reported levels of corruption in those countries – these numbers are in effect quite low as compared to Hong Kong’s (3777 total new cases). New South Wales (1509) and Botswana (1475) had lower totals, but were still competitive on a per capita basis (and in light of reported corruption). Botswana’s total is especially remarkable, since it reports 74% of complainants as having voluntarily identified themselves. Hong Kong is the only country reporting on its efficiency in handling complaints.
Intelligence, monitoring, and investigation: In this area, the data are not uniform. Some countries report the numbers of investigations started, i.e. Argentina (1784), Botswana (390), and Tanzania (1128). Others report completed investigations in various ways, i.e. NSW (6 investigative reports), Ecuador (322), India (3568 investigation reports received), Malaysia (413), and Uganda (98). It is hard to make any sense of these data. Numbers of completed investigations seem more meaningful than numbers of investigations started – but neither gives any indication of the quality and focus of the effort, nor the outcomes.
Prosecutions and administrative orders: Here again, the reporting is disparate. Some agencies report overall numbers, while others report actions taken by prosecutors and administrative supervisors, and a few report both – viz. Argentina (317 referrals, 44 prosecutions), NSW (10 prosecutions, 10 disciplinary proceedings), Ecuador (79 referrals), Hong Kong (504 prosecutions), India (59 prosecutions), Malaysia (152 prosecutions, Tanzania (94 prosecutions over 6 years), the Philippines (2,209 prosecutions by the Ombudsman itself). The Philippines far exceeds the other countries in prosecutions, but there is a problem of comparability, since its ACA is the only one with authority to prosecute on its own. New South Wales reported only 10 prosecutions, but this is surely a result of the narrow focus of its investigations – the number of referrals without investigation is not reported. Numbers and percentages of convictions are perhaps more telling, for the few that report them: Argentina (does not report convictions, but did report about half of prosecutions being dismissed), Hong Kong (302 convictions, for a 60% success rate), Malaysia (89 convictions, for a success rate of 25%), the Philippines (514 cases in which penalties were imposed, or 23%). By this measure, Hong Kong is by far the successful, while Malaysia and the Philippines report more modest success. Again, the numbers do not convey the quality of the proceedings, nor the harshness or deterrent effect of the penalties imposed.
Preventive research, analysis, and technical assistance: Only a few ACAs report results in this area, i.e. NSW (17 reports, 265 recommendations of which 148 fully implemented), Hong Kong (106 government studies, 260 private sector requests for preventive services), and the Philippines (10,583 preventive assistance requests addressed). These numbers are not really comparable, since they do not indicate the scope of the assistance activities counted. The most telling number is perhaps NSW’s 56% success rate in having its recommendations fully implemented.
Ethics policy guidance, compliance review, and scrutiny of asset declarations: Most of the ACAs covered here do not process asset declarations. Of these only, Argentina reported on this (a near-perfect compliance rate).
Public information, education, and outreach: Here too, there was little reporting. Numbers are published by NSW (3 conferences, 5 training events), Botswana (145 presentations), and Tanzania (12 public meetings, 225 seminars, 157 radio programs). The volume of activity seems uninformative, especially since one would probably expect the quality of the NSW ICAC’s few outputs to be quite high, and since Tanzania’s impressive level of activity is not matched by either public credibility or success in its overall mission. The Hong Kong ICAC does not include such figures in its reports.
Overall, these numbers indicate volume of activity more than they do quality or success. Hong Kong and New South Wales seem to report the most impressive data on outcomes, although other agencies report higher levels of raw output. Malaysia and the Philippines appear to be next in rank. The outcome numbers for India and Tanzania seem especially disappointing. Since Singapore does not report results, we can only make inferences from perceptions of its quality and the long-term results, which are impressive.
Explanatory Factors:
In order to explain differences in performance among ACAs, we return to the factors presented in part 2 above (Box 1). We find that some of these factors do appear to have a certain predictive value, while others do not. We review each factor in turn:
Establishment: The success of an ACA depends on its being carefully situated from the start within a set of well-defined supports. These would include a comprehensive anti-corruption strategy, careful planning and performance measurement, realistic expectations, and strong enough political backing (across class/party) to make it effective regardless of (political and personal) consequences. The agencies that seem to score highest on these measures are those in Hong Kong, Singapore, Malaysia, Uganda, and Australia/NSW – i.e. largely those that are also the most successful. The agencies in Botswana and Tanzania benefited from strategy development in the aftermath of scandal, but fell short in the areas of political backing, planning, and expectation-setting. In the Philippines, Korea, and Ecuador (perhaps Thailand as well), there does not appear to have been either sufficient political backing or wide consensus on strategy. In Argentina, the agency got off to a strong start in these areas, but has faced severe challenges due to the economic crisis.
Focus: The ACA itself needs to be strategic in defining its focus in a way that will maximize its effectiveness. For example, an agency could focus on prevention and monitoring government implementation of anti-corruption policy (foregoing a comprehensive mandate, as in Korea); its jurisdiction could be mainly prospective (only limited concern with past cases, as in Hong Kong); it could choose cases selectively, based on clear standards (as in Argentina and NSW); or it could deal only with the probity and reputation of the public service (as in the U.S. and India). The Hong Kong ICAC model discourages focus and selectivity – the only decision taken to limit ICAC’s jurisdiction in Hong Kong was the 1977 amnesty, which enabled the agency to clear away a huge potential backlog. Similarly, Singapore, Malaysia, and Botswana have eschewed selectivity in favor of comprehensiveness – but Singapore still managed to sequence its proactive agenda in a way that suited its needs. In Botswana, by contrast, comprehensiveness appears to have created an overwhelming workload for DCEC. The agencies in Uganda, Thailand, and the Philippines also have large and varied responsibilities, with no clear mechanism for paring these down. By contrast, Argentina and NSW have explicitly focused their investigations on cases deemed strategic. It appears that clarity of focus is indeed consistently associated with success – except where massive resources are available (Hong Kong).
Accountability: This comprises such things as the application of legal standards, the availability of judicial review, systems for public complaints and oversight, a requirement that the agency answer to all branches of government and the public, and precise and comprehensive expenditure accountability. Some commentators also suggest keeping the agency’s size, as well as the “free” support given by aid donors, to a minimum. Of all the agencies, Australia’s NSW ICAC is the one that seems to meet all these criteria unequivocally. The Hong Kong ICAC seems to meet all but one – the size limitation – and this, interestingly, is the one that Singapore’s CPIB meets most clearly. Apart from internal monitoring of CPIB within the President’s Office, it is the agency’s small size – together with its professionalism – that keeps it within some bounds. Only the Hong Kong and NSW agencies have citizen oversight committees, and these agencies, along with Malaysia’s ACA and the agencies in the Philippines, Thailand, and Ecuador, report to parliament. Like CPIB, the agencies in Argentina, Korea, Botswana, and Tanzania are subject to audit and oversight within government, not outside it. However, the Argentine ACO and Botswana’s DCEC also publish their annual reports. All of these agencies are subject to judicial decisions when their cases are brought to trial, but only Hong Kong, the Philippines, and Australia/NSW appear to exercise effective judicial restraint on investigative methods and detentions. Overall, accountability is not uniformly associated with success in terms of ACA goals – indeed it is frequently a hindrance, stopping or delaying agencies from taking desired actions. Still, many of the successful ACAs are strongly accountable, but this is probably an outgrowth of the rule of law, which seems to be more consistently associated with success (see below).
Independence: This in some cases arises simply from outside accountability, sometimes from the agency’s placement and line of responsibility, the appointment and removal procedures for top officials, or some form of fiscal autonomy. The most important sign of independence is the absence of political intrusion into the agency’s operations. As noted above, many of the agencies have some (non-trivial) form of outside accountability – the exceptions being Singapore, Korea, and Tanzania. Interestingly, not many agencies have formal independence or fiscal autonomy. Those in Australia/NSW, Uganda, the Philippines, Ecuador, and Thailand are structurally independent. These same agencies, along with Hong Kong’s ICAC, have fiscal autonomy in the form of their own budget line provided by parliament. In some cases, a combination of outside accountability and strong political support from the press and the public can overcome the absence of formal guarantees of independence. This is most obvious in the case of Hong Kong, and also true of Argentina and Malaysia. Another compensating factor seems to be the professional and non-partisan nature of senior staff, notably in Singapore and Argentina. In sum, formal independence is no more a cause of success than accountability. De facto autonomy, on the other hand, enables ACAs to operate on a consistent and professional basis with relatively little partisan intrusion – and in most environments, this mode of operation is important for success.
Powers: Observers have suggested that a successful ACA will have strong research and prevention capabilities, along with the authority to do the following: access documents and witnesses, freeze assets and seize passports, protect informants, monitor income and assets, propose administrative and legislative reforms, and exercise jurisdiction over the chief of state. Here again, the NSW ICAC meets all these criteria – and its credibility is underlined by its successful prosecution of the provincial executive not long after its creation. Hong Kong’s ICAC can do the same in principle, and it has nabbed some “big fish.” However, it has not brought down a governor, and its ability to do so has come increasingly into doubt since 1997. Leaving aside jurisdiction over the chief executive, the agencies in Singapore and Malaysia appear to exercise the other powers effectively – perhaps all too effectively, as a result of limited judicial scrutiny. Other agencies have most or all of these powers on paper, but frequently cannot put them into effect due to lack of coordination, weak capacity in cooperating institutions, and political factors. In short, it is an agency’s ability to exercise these powers in practice that contributes to its success.
Staff: Agencies in this field, as in others, depend on well-trained personnel – including sufficient numbers with highly specialized skills. Staff should also be well-compensated, subject to integrity reviews and quick removal, and endowed with a strong ethic of professionalism, integrity, and high morale. Hong Kong, Singapore, Australia/NSW, and Malaysia lead the way here – along with the U.S. and other leading industrial countries. Argentina attempted to follow this pattern, but its economic crisis has severely reduced compensation levels, thus diminishing the ACO’s ability to recruit and retain highly professional staff. The other countries considered – notably Uganda, Tanzania, Ecuador, India, and the Philippines – appear to have considerable difficulty recruiting, paying, and training personnel of sufficiently high quality. Botswana, Thailand, and Korea are in a better fiscal position, but the integrity of recruitment processes and agency procedure are subject to some doubts. While one cannot specify a benchmark number of staff, due to the variations in agency design and context, it is quite clear that a sufficiency of highly professional, well-compensated, and motivated employees is strongly associated with success.
Other resources: These include sufficient funds, adequate facilities and assets, and high-level information sharing and coordination with other government bodies. The picture here is essentially the same as for personnel. The overall budgets of these agencies (apart from Hong Kong’s) do not correspond closely with strong capability and success. Singapore’s CPIB had a budget of $3.23 million in 1991-2, considerably less than most of the other agencies, and yet has highly compensated and qualified staff, as well as a strong reputation for effectiveness. Singapore’s relatively small size and population, and CPIB’s ability to compel cooperation without great effort, would account for some of the difference. On the other hand, Singapore has a much larger economy than most of the countries reviewed here – as does Hong Kong, which spent about 20 times as much on ICAC. As of 2000-1, the budget for Tanzania’s CPB was at the same level as CPIB’s budget in 1991-2 – a much larger amount as compared to the size of the economy and of fiscal resources, but a much smaller amount in terms of the size, population, and range of governance difficulties experienced. In short, the size of the budget doesn’t tell us much. A country that is serious about restraining corruption will allocate sufficient funds, and ensure that they are well-used.
Complementary institutions: The stated criteria here include adequate laws and procedures, basic features of the rule of law including functioning courts, free and active media, an active community of NGOs and public interest groups, and other capable institutions such as supreme audit and central bank. This list of criteria seems to cut in two directions. On the one hand, the most successful ACAs do seem to operate in environments characterized by effective laws, procedures, courts, and financial system governance – and ACAs are not successful in the absence of these factors. On the other hand, civic factors such as free media and capable non-governmental watchdogs are not as clearly associated with ACA success. Their relatively low profile in Singapore and Malaysia -- and their relatively high profile in such countries as Argentina, Uganda, Korea, the Philippines, and India -- suggest that these conditions are neither necessary nor sufficient for success. On the other hand, media and civic organizations have helped make a fundamentally open and cooperative approach to corruption control successful in Hong Kong, Australia, and the U.S., and have likely been important to modest successes achieved in the other countries cited above.
Other exogenous conditions: Another set of necessary background conditions would include macroeconomic stability and the absence of crippling distortions, and an environment where corruption – though it may be deep in a few sectors – is not entrenched across the whole system. One could argue that Argentina’s ACO had very good prospects for success, but that the economic crisis has made this impossible. Similarly, poverty and economic shocks in many African countries, including Tanzania and Uganda at various times over the last decade, have undermined even promising initiatives. As important, corruption that touches virtually all of government and the private sector makes it impossible for an anti-corruption program to gain traction. There are too many opposed interests with the power to undo it. This appears to be the case in India, the Philippines, Tanzania, perhaps Argentina, and some other countries and regions not discussed here – e.g. Indonesia, several Latin countries, and most of the former Soviet Union (apart from the Baltics), Africa, and the Middle East. The apparent outliers – Hong Kong and Singapore – actually fit the above profile. They were able to attack widespread entrenched corruption in a situation of accelerating economic prosperity, when the top officials in government undertook an honest and credible program of suppressing corruption through dramatic measures. It was the dramatic measures themselves that were unusual, and were made possible by the peculiar city-state contexts. Moreover, these measures arguably were not necessary to defeat corruption over the long-term, but they were critical to the goal of a rapid reversal.
Lessons
What are we to conclude from the varied experiences just discussed? As we have suggested, neither the information base nor our understanding of how anti-corruption agencies affect overall governance quality is sufficient for us to draw hard and fast conclusions. However, the record does suggest some lessons and some recommendations to guide future action in this field. To these we now turn.
Overall, the record of experience with ACAs seems by and large to support both the applicability of the success factors discussed in Box 1 above, and the hypotheses presented in Box 2. Again, spotty and inconsistent information makes rigorous documentation of these findings difficult. In more specific terms, the following lessons emerge from the record:
Establishment: The “constitutional moment” of establishment seems to be critically important, at least for agencies that operate in contexts that are effectively pluralistic. This means capturing the momentum created by scandal and crisis, gaining consensus on a reasonably clear and realistic strategy, and mobilizing the resources to implement it. Totalitarian regimes, and even “lite” authoritarian regimes such as those of Singapore and Malaysia, can to some extent do without this, relying instead on hierarchical command and social norms of rule-obedience.
Environment: Exogenous conditions, such as macroeconomic stability, public order, and the existence of effective complementary institutions (e.g. courts, audit agency, media), are equally fundamental. It is possible (though proof is unavailable) that these conditions themselves are sufficient to bring about the desired anti-corruption results without the need for a single powerful ACA – or that an ACA has impact by helping bring about such conditions. What does seem clear is that agencies absolutely need these prior conditions to succeed. The Argentine case shows the necessity of macroeconomic stability, while Uganda and Tanzania most clearly illustrate the need for effective and cooperative partner institutions.
Powers and Resources: The capabilities of the agency – in terms of powers, well-trained staff, budget, and capital resources – also play a central role in success. What we cannot say with any confidence is how much of a given capability or resource an agency will need in the abstract. There is no minimum or maximum size, no standard budget or number of investigators. Clearly, the staff cannot be so few as to be overwhelmed by their caseload, and it is also clear that the professional staff need to be highly trained and motivated – and such personnel are extremely difficult to find in significant numbers in most poor countries. A successful ACA cannot be run on the cheap. A small, comparatively inexpensive agency such as Singapore’s CPIB invests a very large amount in compensation, training, technology, and support for each professional. Besides this, Singapore facilitates the agency’s operation by giving it strong coercive powers and relieving it of any duty to mobilize or account to the public (one might think of these last as off-budget costs).
Safeguards: Both accountability and formal independence seem to be somewhat overrated in the literature – at least as they affect an agency’s performance strictly in anti-corruption terms. Formal independence, like formal dependence, can be overridden by political factors. It does appear, however, that an agency’s de facto autonomy to operate in a professional and non-partisan manner increases its prestige, hence its ability to mobilize political support and cooperation for its aims. Accountability has much the same effect. It moreover affords observers within and outside government the opportunity to monitor the agency’s performance and to propose corrections. Obviously, from a broader perspective considering civil rights and democracy, accountability and independence are desirable in their own right.
Focus: The element of focus turns out to be more fundamentally important than expected, and is underemphasized in the literature. No agency can cope with an unlimited mandate. Choices must be made. One approach is that of the Hong Kong ICAC, which has broad anti-corruption jurisdiction over the public and private sector – and interprets this as requiring it to pursue every allegation, without imposing selection criteria. One could consider this a “social contract” between ICAC and the people of Hong Kong, in which the citizens offer support and trust, in return for which ICAC handles all cases big and small – “without fear or favor.” Certainly, this is one way to avoid any hint of flawed or self-interested choices. It also seems to conform to the “broken window theory,” whereby every infraction, no matter how trivial, must be addressed in order to deter more serious offenses and to arrest a potential slide into a high-corruption equilibrium. This approach appears to enable social forces opposing corruption to gain the upper hand and sustain a low-corruption equilibrium. However, in Hong Kong this required a huge, and hugely expensive, agency.
At the other extreme, the New South Wales ICAC in Australia seems to take a “test case” approach, choosing to pursue only those allegations (and only those institutional reform studies) that will result in high-impact action. Argentina does something similar, using its social-political-economic weighting, but is less selective. A selective approach seems to require both a strong ability to justify such choices, and capable alternative institutions to pursue cases on referral. NSW ICAC thus sticks to a relatively narrow focus, in the interest of using scarce resources strategically. This works in an environment where such an institution is less likely to be suspected of partiality. It is also important, of course, that the NSW ICAC statute permits it to be selective. In such a case, selectivity seems to serve the broader objective, which has less to do with the “retail” work of investigating every complaint, and more to do with ensuring leadership, coordination, and attention to the anti-corruption effort.
As in other areas, it is difficult to prescribe an approach in this area in the abstract. Given the limited budgetary resources of developing countries, there is reason to favor a more limited focus to maximize impact (as in Argentina). However, in these same environments, the need to both tip the equilibrium and constrain discretion, and the frequent lack of effective complementary institutions, suggest that the Hong Kong approach might be preferable. Unfortunately, the resource constraint, as in Botswana, can easily render this approach ineffective.
The other aspect of focus has to do with breadth of jurisdiction. The importance of the private sector in corruption suggests a need for some flexibility about the limits of an ACA’s scope. Again, however, we come to the resource constraint, with Botswana illustrating the problem of overstretch. There is certainly some logic to avoiding artificial barriers, as demonstrated by Hong Kong and Singapore, and indeed there may be some cost-saving synergies. The point is perhaps simply that the mission, jurisdiction, the selectivity question, and the resource base all need to be considered together and integrated into the design of the agency.
Limits and Risks: Last, it is important to highlight the risks as well as the limitations of ACAs in terms of the problems they can address and the impact they can have. As discussed in part 2 above, observers have commented on such risks as politicization, predation, diversion of attention and resources from other necessary areas, and bureaucratic duplication. All of these dangers have been realized in one or more of the various existing and predecessor agencies discussed in parts 3 and 4 – although the gravity of the risks can vary greatly across countries. Singapore and Malaysia have ACAs embedded in the executive and the civil service system, which itself suggests that they operate under some, perhaps mild, political imperatives. However, it would be wrong to conclude that a situation like this is always benign. In most of the world, in fact, there is a serious risk of political manipulation and predation, as happened in Tanzania at least prior to 1995. This risk grows with the tendency to concentrate powers in these agencies. The diversion and duplication risks are also especially important in countries with fewer resources, less mature political systems, and more powerful patronage networks. Here, the ultimate danger is simply that the effort and resources will be wasted, and that this in turn will reinforce public cynicism.
The limitations of ACAs are usually much more severe than people realize when they set out to establish them. Where the ACA is not structurally independent, then it can be no more powerful than its bureaucratic and political patrons. This may be very powerful or very weak, depending on the environment. An ACA’s success depends to a great extent on cooperative relationships with other elements of government. In a sense, this is a strength, since it forces anti-corruption champions to achieve strategic consensus and to commit to concrete forms of cooperation, before moving forward. Unfortunately, this is rarely the case, and it probably breaks down often even in cases where it has been achieved. As a result, ACAs are regularly frustrated by their inability to secure information, cooperation, prosecutions, etc.
ACAs are also largely incapable of addressing the larger forces driving systemic corruption. For example, the fact that most ACAs describe their preventive function in terms of accountability within agencies or sectors suggests that bigger issues such as government-business networks, inducements for rent-seeking, and campaign finance might be getting short shrift. Most obviously, there is no way that ACAs can be effective in a situation where essentially every important institution is compromised. Even if this is not the case, an ACA is essentially a response to symptoms. Fortunately, the Hong Kong ICAC model has highlighted the need for focus on preventive measures, but these appear to be mainly agency-specific, or at least to take the governmental system broadly as given. Anti-corruption agencies cannot address macroeconomic distortions, the lack of credible courts and watchdog agencies, regulatory incentives toward bribery and rent-seeking, and other large-scale “drivers” of corruption. Promoters of ACAs need to be aware of their limitations, and adjust expectations accordingly.
Moreover, we need to keep the limited applicability of these success stories carefully in mind. As experiments with the ICAC model in other countries have shown, success in the four particular environments just cited does not mean that the same blueprint will produce such positive results elsewhere. Our analysis of contributing factors to success goes some way towards explaining why this is the case. Also, whether these agencies provide sufficient net benefits in broader terms – beyond corruption control alone -- is a question we cannot answer. Observers of Singapore in particular wonder whether the anti-corruption benefits of CPIB truly counterbalance the risks posed by its draconian powers and lack of transparency.
In this connection, we should stress again that it is unwise to draw very specific conclusions about the impact of the various agencies reviewed in this paper. We have suggested, above, that the ACAs in Hong Kong, Australia/New South Wales, and Malaysia have been significantly more successful than the others, based on agency performance data, for some of the reasons cited in the literature. Our qualitative analysis, though brief and far from comprehensive, gives us some confidence that those agencies – along with Singapore’s CPIB – actually are adding value in anti-corruption terms, and this probably contributes to the relatively strong governance ratings of those countries. Still, we do not have hard evidence that these results could not have been achieved through multi-agency cooperation in the absence of an ACA. Rather, the qualitative information broadly indicates that the ACAs in these countries have overcome coordination, information, and leadership constraints that a multiple-agency approach might not have.
Recommendations
At the beginning, we referred to the context of this report, namely that World Bank officials desired an analysis of experience with anti-corruption agencies in order to provide sound advice to certain member-states. In response to this, we would make two modest sets of recommendations.
The first group of suggestions concerns advice to member countries. They should be encouraged to avoid the “tackling-the-symptom bias” referred to in part 2 above. Where a country does not have an anti-corruption agency, or where it does not have a single-agency strategy, this should not be the first recourse. The relevant question here is: “What are existing agencies doing, how are they falling short, and what will most cost-effectively address this problem?” If an ACA is deemed to be the answer, the design and establishment of such an agency must come only after a sufficient political consensus is achieved concerning an anti-corruption strategy, and about what exactly the agency will do to ensure that the strategy is effectively implemented. At the design stage, careful, integrated analysis will be needed of the core issues of: mission, jurisdiction, powers, selectivity, relationships, and resources. As a separate matter of high priority, accountability will need to be addressed. If all of these issues are not fully and adequately addressed, the member country should be encouraged to consider alternatives. Those could range from court-strengthening or administrative reform projects to contracting revenue and audit services to an internationally reputable company, to establishing a less complicated agency such as an ombudsman.
The second set of recommendations concerns the information and research base concerning anti-corruption agencies. In a word, it is inadequate. The steps we recommend below are quite costly, but these costs will need to be weighed against the continuing costs imposed by our flawed understanding of corruption dynamics and the effectiveness of anti-corruption agencies.
If there is a serious interest in documenting the experiences and impacts of these agencies, then they will need to be encouraged, perhaps supported as well, in collecting and reporting the full range of performance data following a common standard. These agencies will also need to report more regularly and uniformly on the full range of resource inputs, mandates, strategies, and activities. There will also need to be an effort to collect and systematize data on intermediate outcomes for sister agencies operating in the same fields as the ACAs, in order to make some estimate of the value added by ACAs to overall performance of anti-corruption functions.
Much more complete qualitative data on agency performance will also be needed for a meaningful assessment. For example, in the area of investigations, only in one of the above cases do we have any data on the type or seniority of the officials prosecuted (arrest figures in Malaysia). To have a real idea of how serious anti-corruption efforts are, we would need complete and consistent cross-country data on this point. This should also include information on the affiliations of those investigated, in order to determine if the investigations and prosecutions are non-partisan, and on whether convictions are secured by means that are consistent with human rights and due process protections. In the field of public outreach, how are these initiatives really being received? In many societies, it is hard to imagine that governmental anti-corruption propaganda would receive a favorable hearing. The source and style of the message will be at least as important as its content.
Once the above steps have been taken, a serious research effort can then be launched to address the issue of impact. The first step here should perhaps be improving our understanding of causal links between the intermediate outcomes of anti-corruption efforts and overall indicators of corruption and governance quality. One approach would be to develop an analytical narrative based on a few in-depth cases studies such as the ones presented in part 4 of this paper – but combining the case material with a game-theoretic analysis of what ACAs (for example, of the ICAC type) do, and how they succeed. This research would be part of a broader effort, which is underway but incomplete, to understand the dynamics and effects of corruption and of anti-corruption measures. A later step would be to pursue further improvements in cross-country corruption and governance indicators. Again, efforts on this front have been underway for some time – but there is still some distance to travel.
Bibliography
ACO (2001), “Argentina ACO Annual Report”, ACO.
Anechiarico, Frank and James B. Jacobs (1996), The Pursuit of Absolute Integrity. Chicago: The University of Chicago Press.
“Appendices”, ICAC Annual Report 2000-2001, Australia: New South Wales.
Asian Development Bank (1999), “Governance in Thailand: Challenges, Issues and Prospects”, .
Asibuo, Sam K, “Role of an Anti-Corruption Agency in the Struggle Against Corruption: The Case of the Serious Office in Ghana.”
Baqwa, Selby (2001), “Anti-Corruption Efforts in South Africa”, The Journal of Public Inquiry, Fall/Winter.
Bardhan, Pranab (1997), “Corruption and Development: A Review of Issues”, Journal of

Economic Literature, Vol. XXXV.
“Botswana’s Approach to Fighting Corruption and Economic Crime”, International Anti-Corruption Newsletter, Jan. 2000.
Bunnag, Marut, “Unusual Wealth”, Marut Bunnag International Law Office, Bangkok,

Thailand, < http://www.marut.th.com/law_eng.html>.


Camerer, Lala (1999), “Tackling the Multi-Headed Dragon: Evaluating prospects for a

single anti-corruption agency in South Africa”, Paper presented at the 9th International Anti-Corruption Conference.


CCCC (2000), “Informe de Actividades II”, CCCC.
Chulalongkorn University Political Economy Center (1998), “Thailand Counter-

Corruption Programs”, Paper presented to The Asia Foundation, Strategies for

Counter Corruption Workshop, Bangkok, Thailand, May.
Damrongchai, Dr. P, “Good Governance & Counter Corruption in Thailand”,

1   2   3   4   5   6   7   8   9


Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©atelim.com 2016
rəhbərliyinə müraciət