Innovations for Successful Societies
		


AN INITIATIVE OF
THE WOODROW WILSON SCHOOL OF PUBLIC AND INTERNATIONAL AFFAIRS 
AND THE BOBST CENTER FOR PEACE AND JUSTICE


	
Interview no.:		A 1

Interviewee: 		Richard Messick			

Interviewer:		Gabriel Kuris

Date of Interview:	3 February 2012	

Location:		World Bank
			Washington, DC


Innovations for Successful Societies, Bobst Center for Peace and Justice
Princeton University, 83 Prospect Avenue, Princeton, New Jersey, 08544, USA www.princeton.edu/successfulsocieties



KURIS: 	I'm here on February 3, 2012 with Richard Messick of the World Bank and we're going to talk about anticorruption authorities. Why don't you talk a little bit about the state of the field? What are the current viewpoints on anticorruption authorities? You mentioned there are very diverse organizations. 

MESSICK:	There has been a series of articles in the last few years saying "Gee, these agencies aren't performing, was it a mistake to create them?" The reply from one of those most involved in creating them, a former staff member at the Hong Kong independent anticorruption agency, ICAC (Independent Commission Against Corruption), is, "Oh you World Bank and development types don't understand" and so forth. My own view was always: it wasn't such a good idea to create them unless you could answer two questions. What would a new agency do that existing agencies aren't doing? Why would it perform the same functions better than the agencies that are failing? If it is a question of investigation and enforcement, prosecution, why aren't the police investigating? Why aren't the prosecutors investigating? What makes you think a new agency would do better? 

	There are practical answers to these questions: We can hire new people who aren't tainted by the corruption in the old agencies. We can set the bar higher in terms of education and training. Given the civil service rules and regulations, we can't get rid of the corrupt people in the police and prosecutors' office so they'll taint any new people who come in. Those would seem to me second-best answers but they are answers. 

	Of course the real problem is the political independence, and the neutrality, the even-handed enforcement that bedevils these agencies. Example of the one in Nigeria when it was run by Nuhu Ribadu, he said in so many words what I think is just obvious. That is, if the day after the President appoints you to head an anticorruption agency, you open an investigation against the President you're not likely to get very far. So you need to build some support before you go after those in power. Of course typically the way you do it is you get lower-level people or big fish from the opposition. When you get big fish from the opposition the cry is that you're just a tool of the people in power. 

	It looks like we're saddled with these anticorruption agencies, whether creating any particular one in any particular country was a good idea or not. So then the question becomes how do we help them do better.  The questions of the neutrality and independence are structural issues involving the existence of adequate political competition and a strong middle class;  matters  beyond the power of any outsider, or even the World Bank, to do a great deal about. But outside of those, we can try to get them to realize that they have to cooperate with other agencies and what are the best ways to do that. 

	In Iraq, for example, I convinced the government to put somebody in the prime minister's office as a special assistant to the prime minister to see that the various agencies that had anticorruption enforcement and investigation powers worked together. In Iraq you had the Board of Supreme Audit, you had an Inspector-General for each department, an American transplant. Then you had an anticorruption agency. Those three were operating on different planets and they needed to work together. It looked like the only way you were going to get them to work together was to get the prime minister to pressure them to work together. 

	My suggestion was a special assistant in the prime minister's office with that brief. I don't know if he is still there. For a while it seemed that it was working. The coordination problem is a big problem because anticorruption covers so many different areas. 

KURIS: 	Do you think-the three main areas we talked about were repression, investigation and prosecution, then prevention functions and then research functions.

MESSICK:	I'd say research and education. 

KURIS: 	Do those areas conflict with each other? 

MESSICK:	I think organizationally part of the problem can be: is this agency full of cops or teachers? How do we make sure the one function doesn't squeeze out the other? Now Hong Kong does it because they essentially have different cones. You come in as a preventer or you come in as an enforcer and you go all the way up to the number two-there are several assistant commissioners, one of them is through enforcement, one of them is prevention and one of them is for community outreach. So you can get to the second position in the agency through your specialty. That is the kind of thing you have to pay attention to when you start mixing teachers and preventers and enforcers.

	You have problems of course with the prevention types. You'd like people to be more candid with them about problems they face so the preventers can get some inside scoop and maybe offer some solutions. But people may be reluctant to come to an agency and disclose sensitive information if the agency also has an enforcement role. It is one of the advantages for example in the US of the Ethics in Government office, or the designated ethics officers in each agency. They can counsel people and the people who come to them for counseling and advice know that they're not going to have an enforcement person show up on their doorstep the next day. 

	The US Office of Government Ethics and its designated ethics officers in each agency are effective because of that counseling function and pre-clearance. People don't get caught up in a conflict of interest; people don't get promoted into a job to handle contracting involving companies that they have stock in until they divest themselves of the stock. So it's not that there are lots of enforcement personnel. The prophylactic work is where the Office of Government Ethics has been really successful. 

	Of course with the Senate confirmation people don't get to be confirmed to be Secretary of Defense if they have a large portfolio of defense stocks. Another example of where the preventive function has been very helpful.

KURIS: 	Many of these single agencies like in Indonesia also have a role in coordinating government-wide anticorruption policy. How do you think that works in practice?

MESSICK:	Where I've seen-where agencies have had that power I haven't seen it exercised effectively for reasons of bureaucratic jealousy, conflicting mandates. That's why I j question whether coordination can be done by anything but the head of government. 

KURIS: 	Do you think it matters where the anticorruption agency is seated, in the legislature, the judiciary or executive branch or total independence? 

MESSICK:	That will tend to be a function of both the formal constitutional structure and the way power is actually exercised.  In Armenia, the government proposed the creation of an anticorruption agency accountable to the President. The Armenian constitution is set up along the French lines where the President is a bit removed from the government of the day, more of a sort of semi-monarchical figure. But nonetheless as a practical matter in Armenia today the President and the party in power are the same party and they work very closely. So when the government proposed an anticorruption agency that would report to the President and would have powers over parliament, the parliamentarians strongly opposed it because they saw this as a way of the government in power controlling them: "If you don't vote our way we'll investigate you." 

	Perhaps if there had been a more-the way the French Fifth Republic was at least conceived with the President divorced from day-to-day politics this structure might have been accepted. But the fact of the matter is, the realities were the close connection between the President and the prime minister and the ruling party, it was taken in a different way. It was interpreted, probably quite correctly, as a way for the ruling party to control the parliament. 

KURIS: 	I want to ask you a little bit about the pressures leading to the creation of all these anticorruption agencies. First of all on the international side, is it mainly because of UNTAC (United Nations Transitional Authority in Cambodia) or is it kind of-? 

MESSICK:	 On one level it has been the Hong Kong people saying, "Look, it worked here." So you don't know what to do when somebody comes and says, "We cured our problem with this so you should try it." It is so-when a political leader wants to do something about a problem, particularly one that doesn't admit of any easy, short-term solutions, creating an agency to address the problem is very appealing politically, very appealing politically. 

	In the US you can cite the Department of Energy, the Department of Education and the Department of Homeland Security. 

KURIS: 	The investigation into the mortgage department.

MESSICK:	Yes, even when they raised the Veteran Affairs Administration to a cabinet level. These are-you've got an energy problem, we'll create a Department of Energy to solve it. We have a problem with protecting the homeland, we'll create a Department of Homeland Security to protect it. Got a corruption problem? We'll create a corruption agency to solve it. Certainly in the short-term it is something to campaign on and boast about. Of course measuring the effectiveness of these agencies in the long-term is itself still a contested art. 

	Now, what the World Bank and others are trying to do is develop effective performance measures, something quite difficult because of the attribution issue. Did corruption decline or increase because of something the anticorruption agency did?  Or is it because huge deposits of oil were suddenly discovered that fueled a sharp increase in corruption? Or, alternatively, was there a massive crisis that led to a change in government and different outlook. So getting the performance measures right for these agencies-. 

KURIS: 	It's hard enough to measure corruption right?

MESSICK:	It's not at all easy to measure at all except in isolated cases like in sectors where you can use quite expensive techniques, like Olken did with roads in Indonesia.  Then even if you have a reliable measure the attribution issues looms.  How much is does the presence of one agency affect the level? That's tough.  

KURIS: 	Within the World Bank, is corruption usually measured through the cost-of-doing-business index?

MESSICK:	No, it is usually measured-[interruption end of file 1]

	There is no formal agreed-up, institution-wide way to measure corruption. For a while there were the World Bank Institute Governance Indicators. They became very controversial. In fact the board directed that they not be used. One way we measure it now is with the judgmental factors in the country policy and institutional assessment, the CPIA. And that you'll see is a twenty-factor judgmental criteria that spans everything from status of health and education through a couple of questions on governance and anticorruption related. Those are just judgments of the staff in the country that are reviewed by people who have less of a stake in the particular country than the country staff. 

KURIS: 	Do you find that these anticorruption agencies, now that they've been around for a few years, are they starting to share lessons learned or best practices?

MESSICK:	Probably the one area where there has been a good deal of sharing has been  OECD (Organisation for Economic Co-operation and Development) / ADB (Asian Development Bank) anticorruption initiative in Asia. The World Bank has tried to do some of that and there will be a second meeting of the International Alliance of Corruption Hunters June 5, 6 and 7 of this year in Washington where it will bring together people from all the agencies. Not only anticorruption agencies but those involved in enforcing the anticorruption laws. So it could be not only an agency or a prosecutor or even a fraud squad of a police department. 

	As much as the lessons learned that can be shared across borders, it is important, particularly with grand corruption that these people cooperate across boundaries because e the bribes paid by someone in country A to someone in country B who then stuffs the money in country C or directs that it is to be paid into a bank account in country C. So you want to get everybody involved. You have to have the enforcement people of countries A, B and C working together so that the obstacles to gathering evidence and so forth when you cross national boundaries don't just bog down the investigation. There is still a lot of work that has to be done on that. 

	The only way that you can do it is to develop personal ties so that people in country A know there are trustworthy people in country B they can call and say, "It looks like an official in your country was paid a bribe by a company based in our country, can you follow up?" They won't make that call unless they know who is on the other end of the line and they have some sense that the information won't be immediately disclosed to the wrong people. 

KURIS: 	Right. Do you have any opinion on whether it is better for these agencies to focus on the petty corruption, or grand corruption, or try to do everything at once?

MESSICK:	Often the mandate of the agency will dictate that. With many of them the mandate will be grand corruption. This reflects the view that it is going to be hard to deal with petty corruption if all those potential petty corruptors know there are large amounts being stolen on a grand level. That's just the conventional wisdom; the evidence is all anecdotal.

KURIS: 	 What are some of the most important powers for anti-corruption agencies to have?  

MESSICK:	It depends on what kind of agency you want. In Morocco the agency has no investigative or enforcement power, it is purely education and prevention. Now, is it important that they have enforcement powers? Well, if the police and the prosecutors are working well, no. So that-because the agencies are so heterogeneous it is hard to say. 

	I think once you give it investigation or prosecution powers than you've got to think about these questions of political neutrality, evenhandedness and then go to all these things about who has power over the budget, who has power over the personnel selection, promotion and termination.  

KURIS: 	Are there specific personnel policies that you think kind of reinforce independence here?

MESSICK:	 Merit hiring is going to be your best sort of policy with a small p. Policy with a capital P, of course, is no interference on political grounds. Enforcing that is less a question of law per se and more a question of the institutional landscape of the country. One view is the reason courts are independent in some countries is because there is significant political competition and the competitors in the political arena all find it in their interest to have a nonpartisan, neutral referee governing those disputes. So they all have an interest in an independent agency and they watch each other to make sure that the other side doesn't compromise that. 

KURIS: 	Turning towards Indonesia, one issue I've been struggling with before I go there is is there anything unique about corruption in Indonesia or the region? If you'd like to talk about this more generally, is corruption the same everywhere? 

MESSICK:	We don't know enough to say. People will-one of the things people will say is that corruption was predictable in Indonesia. If you wanted to build a power plant there was a price list. You paid that price and you got your power plant. So it wasn't arbitrary and it didn't vary from person-to-person. And so therefore it was just like a tax, a cost of doing business. I have no idea whether that is true or not, and I've never seen anybody try to establish that empirically. That is just the lore of Indonesian corruption. So I don't know then, even if it were true, does that hold in Lao, does it hold in Cambodia, does it hold in New Jersey? I have no idea. That's the problem when you can't taste, touch or feel it and you only see occasional manifestations of it and you don't know that you're seeing 90% of it or only 1% of it. Are you seeing a representative sample of it or just a biased sample because that's what you, the enforcement agencies, can pull out and detect.   

KURIS: 	This would also be true in Latvia, Lithuania.

MESSICK:	It's hard.

KURIS: 	The same perspective.

MESSICK:	Particularly of course economies that were all state run, the former communist countries, well wasn't there a lot of corruption in one sense when people didn't work full time but they got paid full time? What does it mean to have corruption in a state-run economy? People lied about meeting their quotas or falsified statistics in order to look as if they met the five-year plan. Well, is that corruption? We certainly saw a lot of it in the transition, but was this a new phenomenon or just social pathology that manifested itself in a different way because it was now a market economy as opposed to a state-run economy?   

KURIS: 	What do you think anticorruption agencies are best at? What are the kinds of good things that they're doing? I don't know if you can generalize it but if you look at the better ones-?

MESSICK:	Well if you tell me what you think a better one is and tell me-I mean, I think the Office of Government Ethics in the US does an excellent job on preventing conflict of interests using the income and asset disclosure statements. The Federal Election Commission does a good job in making sure people disclose the political committees, candidates disclose contributions and expenditures. Do they do a good job of catching election law violations? That's a more controversial question. 

KURIS: 	So what is it about disclosure rule that makes that kind of work better?

MESSICK:	It is  if the rule is to disclose period it is easier to enforce than a rule that says don't spend more than a given amount. It is very hard to enforce this type of rule. How ambitious your law is, what you mean by corruption to begin with is caught up with your question.  

KURIS: 	It seems like going off of what you were saying about how often times governments create these anticorruption agencies for political reasons, how can these agencies help to manage public expectations of their performance?

MESSICK:	That's really tough because usually by the time they're created they're expected to dance across the Potomac and if they just walk across the Potomac, expectations are dashed. I'll tell you what we do know from studying the federal agencies here, the person who is the first one to run it is really important. If you look at how J. Edgar Hoover ran the FBI, he did-he made it an elite agency, which it still is today. You have to have at least a college degree and typically a more advanced degree in law or accounting. He played to the newspapers with consistently increasing arrests but at the same time he kept them away from organized crime and drugs because those are areas where the agents can be corrupted and lead to bad press. 

	So the FBI had this aura and this reputation that it still enjoys today that in some sense was just a product of their most popularly-the offence that they were most successful at dealing with for most of the period that Hoover was director was the Dyer Act, that is taking a stolen car across a state line.  If you steal a car in Virginia and drive it into Maryland it is a federal offense. The FBI had relations with local police agencies that would allow them to make the arrest when the car crossed the state line. Of course the offenders were almost always juveniles, who typically didn't corrupt FBI agents, did not shoot them. Particularly as automobile traffic increased in the '50s and '60s, the bureau's arrest record and case solved data went through the roof and so it looked really good. 

KURIS: 	The anticorruption agencies, in Indonesia specifically, they often go off for low-hanging fruit.

MESSICK:	 That's certainly the advice. To build political support you've got to start somewhere. If you can get a couple of high-profile cases and those, as I say are typically going to be people who aren't powerful in the government of the day. Then you can get a lot of low-level people so you can say, "we're doing a lot." It's managing expectations. 

	Allen Doig and his colleagues have a piece in the U4 series about four African anticorruption agencies.  All cane into being with hugely inflated expectations, partly because of the donors. The political economy issue is just getting more complex when there is outside funding because of course the outside funders are persuaded to give the money because it is going to be super successful in six months so it makes it all the more challenging. 

KURIS: 	Going forward how do you think the scale looks on anticorruption agencies is going to evolve? What are the directions things are moving in right now?

MESSICK:	They'll all make adjustments -it's like if you put on a shoe that doesn't quite fit over time your foot will fit into it and the shoe will kind of fit around your foot. The question is, will the ultimate adaptation of the agency to its environment and the environment to the agency result in a useful contribution to the combating of corruption or just a dead weight loss. Of course a lot of that is a function of broader political issues. 

KURIS: 	If these agencies start with kind of bad founders-you say that the founders are very important-. 

MESSICK:	Crucial.

KURIS: 	Do you think that it is possible for agencies that are on to abolish be more effective or is it-the first impression really the most important? 

MESSICK:	First impressions are always really important. Get off to the wrong foot and it is really tough to recover. 

KURIS: 	Have bad anticorruption agencies, is it still ongoing or do you think the wave of creation of anticorruption agencies-? 

MESSICK:	We've about run out of countries to create them in. There's a fixed number. There are two hundred and some odd countries in the world; there's a limit to how many we can create. Although some of the federal states they're now creating state-level anticorruption commissions.  

KURIS: 	Do you think that the multi-agency countries are moving more toward a single agency or are they sticking with the path they started on?

MESSICK:	See all the multi-agency countries are established OECD countries. 

KURIS: 	They seem to be more popular in Europe.

MESSICK:	It's unlikely you're going to see much change there. 

KURIS: 	Just, I guess for ISS' sort of study of this-what do you think are some of the most important kind of questions we can look into?

MESSICK:	How they developed political support. There is clearly a constituency to combat corruption; how can they tap into that constituency to support them when they go after people in power. 

KURIS: 	That does seem to be pretty critical in Indonesia and I think in Argentina, as well.

MESSICK:	Yes.

KURIS: 	Are there any other kinds of questions that you think are important? That's a really big one.

MESSICK:	Good luck.

KURIS: 	Great and thank you. 

MESSICK:	Oh, absolutely. 

KURIS: 	Thank you very much.

Innovations for Successful Societies				Series:	Anti-corruption
Oral History Program						Interview number: A1
______________________________________________________________________




8



Use of this transcript is governed by ISS Terms of Use, available at www.princeton.edu/successfulsocieties