Being a whistleblower in Kenya is a risky business. John Githongo and David Munyakei might be exhibits 1 and 1a in supporting that assertion. More recently, blogger David Mutai was arrested and had his blogs and Twitter account shutdown after exposing corruption at a public agency and in some county governments. More generally, according to Transparency International’s 2014 East Africa Bribery Index, Kenyans reported just 6% of the bribes they were aware of, and a common reason (noted by 10% of respondents) was fear of adverse consequences.
Against this backdrop, earlier this year Kenyan Attorney General Githu Muigai formed the Task Force on Review of Legal, Policy and Institutional Framework for Fighting Corruption. It is not clear how much work the Task Force has done or is doing (you can read the opening address from the June 2015 “Technical Retreat for Development of a Draft Report” here), but it was reported over the summer that the Task Force plans to propose a whistleblower reward system similar to Dodd-Frank’s whistleblower incentive provisions (which have been discussed previously on this blog here, here, and here). Specifically, the reported Kenyan proposal would reward “a person who reports corruption [with] at least 10 per cent of the value of any property recovered after investigations and conclusion of the matter through judicial or other dispute-resolution mechanisms.”
If the Task Force is still looking for ideas (as far as I can tell it has not released any draft legislation or white papers, and the latest news story I could find that mentioned the Task Force is this one from August), I have a few for how to make sure its whistleblower reward program is effective.
There are three main areas the Task Force should focus on to increase the likelihood that the whistleblower reward program will succeed.
- First, a whistleblower reward program is unlikely to be effective without strong protections for whistleblowers. Indeed, the Dodd-Frank provision that the Kenyan proposal may be modeled on addresses whistleblower protections in the very same section. Dodd-Frank § 922(h) provides protections for whistleblowers against retaliation (which may not apply to foreign whistleblowers), although the exact scope is the subject of ongoing litigation, including this recent victory for whistleblowers in the Second Circuit. Without whistleblower protections, the reward system essentially asks citizens to weigh the odds of recovery against large economic or physical costs. In this context, it is important to remember that the risks most Dodd-Frank whistleblowers face—loss of a job, for example—might pale in comparison to the risks a Kenyan could face for reporting grand corruption. For that reason, an effective whistleblower protection program in Kenya must go beyond the protections in Dodd-Frank if the whistleblower reward program is to have any hope of having its intended effect. Kenyan law does already have some whistleblower protections (Section65 of the Kenya Anti-Corruption and Economic Crimes Act, and the Witness Protection Act are examples). But, as this Transparency International Kenya blog post describes, there is much more work to do before the protective provisions could be considered sufficient to make the potential monetary rewards of blowing the whistle outweigh the risk of catastrophic human costs. Fortunately, the Kenyan task force has a wealth of potential sources for best practices, including: Transparency International’s report, “Whistleblower Protection And The UN Convention Against Corruption”, this G-20 compendium of best practices, and this Transparency International’s International Principles for Whistleblower Legislation.
- Second, investigating and prosecuting corruption in Kenya takes time. Of course, the Goldenberg Scandal is an outlier in terms of its size, political connections, and complexity, but the fact that investigations and prosecutions proceeded in a stop and start fashion for well over a decade gives some sense of the delay a whistleblower may need to wait out before she receives any reward under the leaked Task Force proposal. Earlier this year, the Kenyan Ethics and Anti-Corruption Commission (EACC) set an ambitious deadline for investigating prosecuting nearly 200 reportedly corrupt officials, but later extended the deadline, noting the complexity of the cases in multiple jurisdictions. Delays in investigation and prosecution may create a “chicken and egg” problem for anticorruption authorities in Kenya: As the risks facing whistleblowers are unlikely to decrease over the period of an ongoing investigation, the benefits associated with a potential monetary reward lose their value; so, for the incentives to have their intended effect, Kenyan authorities must show a willingness and ability to investigate and prosecute reported corruption expeditiously. Until that happens, Kenyan citizens are unlikely to see great incentive in reporting corruption when any monetary reward they might receive will only arrive far in the future. One potential place to start would be simplifying the legal framework the EACC cited this summer as requiring an extension of its self-imposed deadline. Another way to address the concern about delay might be to provide some form of “up-front” reward for simply reporting, or for reporting that leads an investigation to a certain stage. This, however, this is a technically and ethically complicated proposition, and one I may explore in greater depth in a future post.
- Third, the whistleblower reward provisions must discretely establish what type of information or assistance qualifies a whistleblower for a reward. Failure to clarify what constitutes qualifying behavior is one criticism of the Dodd-Frank § 922 whistleblower reward program. If citizens are unsure whether the information they report will qualify them for a whistleblower reward, it decreases the likelihood they will report. It might not be possible to craft a reward provision that is crystal clear on what information qualifies for rewards and what does not. Indeed, entirely clear qualification provisions might create perverse incentives for officials to receive information/assistance only to the extent it does not qualify for a reward, even when enough information/assistance to qualify for a reward might be available. But given the risks associated with blowing the whistle in Kenya, the proposed incentive structure is unlikely to encourage citizen reporting unless any ambiguity in the eligibility language is read in favor of the reporter. In other words, Kenyan authorities should adopt a presumption in favor of interpreting its whistleblower reward provisions akin to construing ambiguous contract provisions against the drafter—the government in this context.
There is reason to be hopeful that the Task Force is or will consider these issues when crafting its recommendations. In addition, the Task Force may want to consider a list of anticorruption priorities Kenyan civil society groups released for the Task Force this summer. Hopefully, the Task Force members will take the suggestions in that list and given in this post (and in your comments!) seriously, and remember that although monetary rewards for whistleblowers can play a role in an effective anticorruption program, they are most effective within a well-designed anticorruption framework.
GAB | The Global Anticorruption Blog 
This is fascinating and a quite exciting potential step. I wholeheartedly agree with all of your suggestions about what could make it work better. Your second point about the potentially long wait for promised rewards seems to strengthen the need for your the whistleblower protections suggested in your first point. As a corollary to the problem of long waiting times, are there instances in which corruption investigations and prosecutions don’t happen at all, either for problems with parties under investigation influencing the investigation or due to lack of resources and the complexity of the cases? Also, even if cases (and headlines) are happening, are there issues with lack of trust in institutions that would discourage whistleblowers from reporting? It seems like the attractiveness of whistleblower rewards would decrease as the likelihood or perceived likelihood of investigation/prosecution decreases.
Another interesting problem, related to your third point, might be designating who can act as a whistleblower. In addition to knowing what kind of information qualifies for the rewards, it seems like there should be some restrictions on the parties able to collect rewards. For one, should public officials be allowed to claim rewards? It seems intuitive that those investigating corruption should not then be paid for the information found in their official capacity. At the same time, it might be possible that rewards are needed to encourage disclosure from certain government officials with unique knowledge of corruption. Second, what happens in the case of multiple whistleblowers disclosing the same information or when whistleblowers reveal more information on those already under investigation? Would it make sense to allow splitting recovery, to pay out double the amount of reward, or to disallow later whistleblowers from rewards based on the same basic information? The third option seems to be the most consistent with U.S. law, but I’m not sure it would be the most fair.
Thank you for all these very interesting points! As to your first concern, with what effect whistleblower rewards would have if there was no confidence that the agency receiving the report would investigate at all, I wholeheartedly agree. This concern is another factor that makes “up-front” rewards more attractive. If an agency does not typically investigate reports of corruption—perhaps because its investigative arm is captured by corrupt actors—”up-front” rewards would constrain the agency budget in a way that might increase pressure on investigators to actually get some value out of reports of corruption, rather than just “paying off” tipsters. Of course, the actual results of “up-front” rewards would depend a great deal on the specific budget structure of the entity, the involvement of investigators in corrupt acts, and a number of other factors. Still, it is something to think about to overcome the problem of “non-investigation.”
I had not thought much about who should qualify as a whistleblower, but I am glad you raised it because it is quite important. I agree that those whose job it is to investigate corruption should not also be able to qualify as whistleblowers. But there is a concern with how to draw the line on which public officials should qualify and which should not. You make a good point that public officials will often be best suited to report on improper acts within a government agency. Perhaps the provisions on who should qualify as a whistleblower should be scaled over time. For example, perhaps all public officials could qualify in the first two years that the law was in effect, then for years three and four public officials who are not working within an anticorruption agency or inspector-general style role in another agency would not qualify, finally from year five on, only non-leadership public officials would qualify.
Finally, on your point of potential duplicate whistleblower reports, I am conflicted. The U.S. approach which only gives the reward to the first reporter seems easy to administer on paper, but we both no that it is often difficult in practice. I also worry that limiting whistleblower reward receipt based on who reports first might chill reporting in the first place, particularly for individuals who might be concerned they are not the first to report. Perhaps this is another area where the provisions could be broader in first few years after enactment, before narrowing application to the first reporter in later years.
I think your ideas about scaling provisions over time are fantastic. Especially when this is the first implementation of this kind of provision in a country, it’s smart to think about easing the transition by adjusting the scale of the legislation itself. Just as a side note, the scaling of awards to levels of government officials might increase the visibility of the reward system in its first few years. I would imagine those in the limited-time classes who already knew about actionable issues would be motivated to jump on the opportunity while it was available, and any resulting suits could be news to let the wider community know about the availability of the rewards to encourage reporting from a wider base.
Thank you for a great post! I think the first point is especially fascinating, because to me, the problems I envision with implementing adequate whistleblower protections implicate general anti-corruption issues of public trust that exist in the first place. The point that the types of risks facing Dodd-Frank whistleblowers may pale in comparison to those a Kenyan could face underscore this point — in such a context, it seems that the government will need to not only implement technically adequate protections (such as confidentiality, burdens of proof, waivers of criminal/civil liabilities where appropriate) but also must overcome deficiencies in public trust that these protections will be properly implemented and enforced. I’d imagine that task to be particularly difficult in an environment acknowledging widespread corruption, where public trust in government may be lacking. Moreover, even with best practices in place, potential whistleblowers may be wary of uncertainty, such as those rising from determinations of good faith requirements, where the failure to meet such requirements might result in a lack of protection.
I also think that Kait, in the comment above, raised interesting questions about who will qualify — when she raised the issue of public officials, my automatic reaction was to assume that public officials investigating corruption must qualify, because they might be particularly susceptible to influence by targets who may want to evade liability. Thus, these public officials might particularly benefit from an additional incentive to report corruption (similar to the ways that, as has been suggested on this blog/comments before, raising salaries of those in a position of power to stop corruption, such as judges, may be a useful mechanism to combat the issue). However, as Kait raises, it seems perverse if people will be incentivized to not complete their job responsibilities through official channels because they will be rewarded for it otherwise. In that case, a limitation allowing public officials in an investigatory capacity only to report on other colleagues/peers who are not within the scope of allowable investigations or mandate might be one way to think about the issue. Finally, with respect to Kait’s point about multiple whistleblowers — I think that point is fascinating, because it seems that governments would benefit best from achieving a balance of accuracy and speed. Rewarding only the first whistleblower would incentivize first reporting – which might be desirable in an environment where people are initially reluctant to report – but this interest should not be pursued at the expense of directing resources at more reliable reports. Perhaps the lack of protection for non-good faith reports would counter the incentive to race to report, regardless of quality.
Your point about trust in government with respect to the effectiveness of whistleblower protections is an important one I wish I had addressed in the post. It presents a “chicken or egg” problem that you described: how will people be incentivized to report if they are not confident that the government will actually provide protections, and how can the government gain the trust of potential whistleblowers unless citizens actually report? Perhaps there is a role for civil society to help solve this problem. Specifically, maybe the government could provide money to a trusted entity to implement the protections for whistleblowers. I would need to do more research to determine what entity that might be in Kenya, and even if there is a trusted institution, there is no reason it would be well-equipped to implement whistleblower protections, even if the government was footing the bill.
I agree with you that “good-faith” requirements can play an important role in preventing abuse of whistleblower protections and rewards. I do have some concern that drawing the line between good faith and bad (or non-good) faith is difficult. In a country such as Kenya where corruption is pervasive perhaps we should hesitate to draw lines that would reduce the likelihood of reporting on the margin, even if it means some citizens who might not be acting in good faith get the benefit of protections they might not deserve in a perfect system. That is, to move the ball forward on reducing corruption where it is so common, it may be necessary to be more inclusive with whistleblower protections than in a country where corruption is less common. Phase out provisions such as those I proposed in my response to Kait’s comment might work in this context, too.
Thanks for a very interesting post! Regarding the second point, I would be interested to know if there could be targeted efforts to allocate more resources to speed up the process in particularly sensitive cases, where the risks to the whistleblower may be higher.
As you note, in a country with a limited judicial capacity like Kenya, prosecution of these matters takes time. But not all reports of corruption are alike, and surely in some types of cases the threat to the whistleblower is more acute. I wonder if certain cases–perhaps complaints implicating the police, or the military, or certain higher ranking officials, or even in regions of Kenya with more historic ethnic tension–might be more susceptible to retaliation against the whistleblower. In such instances, it might make sense to allocate more judicial resources to expedite those cases. I don’t know if it is possible to make these determinations ex ante, but I would be interested to hear your thoughts as to whether some sort of triaging along these lines would help to overcome the challenge of slow prosecutions.
Daniel, I really like your proposal to “triage” whistleblower reports in a way that can speed up investigation (and perhaps prosecution) where the danger to a whistleblower is particularly acute. It would be important for citizens to be aware of how judicial resources are being allocated to certain types of reported corruption. Your concerns about whether such determinations could be made ex ante are certainly valid, but I would expect that even a system that sometimes misallocates resources might be better than a system that treats all reported corruption equally with respect to whistleblower protection concerns.
I am intrigued by Daniel’s “triage” point as well. However, I might advocate doing the opposite of what he suggests. Instead of focusing on corruption at the highest levels — which is likely to be more complex, carry the highest risk of politicization, and bear the greatest risk to whistleblowers — why not focus on corruption at lower levels? Often times the corruption that effects people the most day-to-day is the small kickback they need to pay a local public official to secure a permit or access to public assistance. The less complex facts surrounding these cases might lead to more efficient resolution and thus generate early success stories for your compensation proposal. Combating corruption at this level could be a good way of building public consensus and useful judicial precedent before taking on the more powerful actors. Furthermore, much of the public corruption rhetoric talks about a “culture of compliance.” Pressuring public officials into behaving appropriately and discouraging those who want to hold public office to line their pockets from running could create this culture of compliance at lower levels. Then, when some of those officials seek higher office, they may be more inclined to reject the assumption that corruption is an inevitable, if perhaps undesirable, product of the system.
Really enjoyed this post! As with the posters above, your second point (and Daniel’s inquiry regarding governmental ability to expedite the process) also caused me to wonder about the execution of a whistleblower program in Kenya.
Specifically, I wonder if you foresee that for certain cases, as a result of the drawn out investigation and prosecution process, the financial costs associated with such processes may tilt the cost benefit analysis in favor of forgoing the process altogether (from a purely financial perspective). As some contributors of this blog maybe already be aware, even developed countries (e.g. Canada) can incur monumental (and perhaps disproportionate?) costs in the investigation of corruption scandals. Do you think in the Kenyan context, disproportionate spending of tax dollars resulting from inefficiency, unestablished procedures, etc on certain “less costly” corrupt activities can be considered corruption in and of itself? If so, did you come across research that addressed this problem, and propose ways to prevent the government from dragging its feet, especially in the form of financial incentives (in the spirit of this post)? Would love to hear your thoughts!
I wasn’t able to find any research on what the (for lack of a better term) return on investment for corruption investigations is. If another reader knows of such research, please share it here in the comments! Even though I do not have empirical support for this contention, I could foresee instances where the costs of investigation exceed the loss from the corruption, particularly in countries where the anticorruption framework is new or underfunded. Still, it may be worth “losing” money on corruption investigations at first, in order for investigators to get necessary experience that will allow them operate more efficiently in the long-term. As with many other aspects of standing up a strong anticorruption regime, there are likely to be growing pains. But if the return on investment remains negative after a few years, that would be a significant concern that I am not sure how to best address. Perhaps financial incentives could play a role there, too, even if they would be flowing to investigators rather than reporters/witnesses.
Your post gives substantial meat for the Kenyan lawmakers to chew on. Strong protection, expeditious prosecution and specific information are basic requirements of any whistle blower protection regime and your post throws light on the complexities inherent in each of them. Your suggestion to have a more inclusive system with greater flexibility at margins is especially important for developing countries like Kenya where officials may se a very narrow interpretation of rules to exclude most whistleblowers. A system built to accommodate greater margin of error in initial years and self-correct itself over time will be more effective than a rigid one. Drawing upon the TI best practices, I felt that two other crucial elements of a whistle blower protection regime should include a) creation of a range of communication channels that allow the whistleblower to reach out without fear of being exposed and b) built in mechanism to use whistle blower disclosures to initiate reforms that would preclude future wrong doings. I also wonder if Dodd Frank is a useful analogy for the Kenyan system as the DF primarily focus on SEC violations. In other words, the focus is on corporate wrong doings and the chief protector is the Government. However, the situation becomes murkier when the whistleblowing pertaining to Government-Corporate collusions or exclusive government corporations. The implications for the whistle blower in such situations are graver and incentives further distant in the horizon.
You are absolutely right to note that the analogy to Dodd-Frank is not perfect (it might not even be that close a fit!) The dangers associated with reporting government corruption are certainly greater, and I really like the TI proposals you pointed out in your comment, including mechanisms for protecting anonymity. I think this gets back again to a concern that my post doesn’t truly address: how can the anticorruption agency convince prospective whistleblowers that they will be protected before some whistleblowers come forward and give the agency an opportunity to prove itself. Until government actors prove that they are not susceptible to capture, it is unlikely that any citizen will simply take the government’s word for it.
In addition to Dodd Frank, I wonder if there could be lessons to be drawn from rewards programs like the one run by the NYPD that pays $10,000 for evidence about the shooting of police officers and lesser amounts for lesser crimes. (Here’s a list of all the NYPD awards) http://www.nyc.gov/html/nypd/html/home/rewards.shtml .
Looking at the victories and pitfalls of such programs could provide two things: 1. people may reasonably fear physical retribution for reporting, so figuring out how to protect them is key and 2. When you get further down the rewards list to $500 for reporting those committing graffiti, there could be lessons to learn on effective ways to use this rewards for whistleblowers fight petty corruption in addition to grand corruption.
Particularly interesting is the idea that you can report completely anonymously and yet somehow still collect the reward; such mechanisms could potentially lessen the risk for whistleblowers if no one knows that they were in fact the whistleblowers.
Thanks for drawing attention to the NYPD program, Sarah. I agree that it might be a better model for what Kenyan authorities are considering. I do wonder how anonymity works with a reward scheme, because at some point, presumably, someone will have to glean the identity to deliver the reward. This might not be of such concern in cases where there is not a government actor engaged in the crime, but I could see difficulties otherwise. Perhaps there could be a third-party facilitating the exchange of the reward to alleviate the concerns associated with government involvement. Maybe a civil society organization could coordinate delivery of the reward, so that the whistleblower remains anonymous to the government.
Pingback: Revisiting Kenya’s Anticorruption Task Force and Assessing its Legislative Proposals | Anti Corruption Digest