An initiative of the National Academy of Public Administration, and the Woodrow Wilson School of Public and International Affairs and the Bobst Center for Peace and Justice, Princeton University Oral History Program Series: Civil Service Interview no.: G7 Interviewee: Clay Wescott Interviewer: Andrew Schalkwyk Date of Interview: 5 February 2009 Location: Princeton University Princeton, NJ USA Innovations for Successful Societies, Bobst Center for Peace and Justice Princeton University, 83 Prospect Avenue, Princeton, New Jersey, 08544, USA www.princeton.edu/successfulsocieties SCHALKWYK: Today is the fifth of February. I am with Clay Wescott, a Visiting Lecturer at Princeton University at the moment, in Corwin Hall in Princeton. Before we start the interview, can I ask if you've given your consent for this interview? WESCOTT: Yes, I have, and I should say, based on your recommendation before, that whatever I say is my own personal view and not the views of any organization that I may be affiliated with. SCHALKWYK: So before we start the interview and talking about some specific aspects of civil service reform and some of the contexts in which you've worked, I wonder if you could give me a very brief history of your work in civil service reform, perhaps touching on how you got involved and one or two of the countries you've been involved in, and what you're doing at the moment, your most recent work. WESCOTT: Okay, well, I'm not sure how you're defining civil service reform, but I think for the purpose of this interview I'll define it fairly broadly, including reform of personnel management, reform of financial management, reform in the administration of government. So I guess the first work that I started doing in it was in Kenya in East Africa. I was an advisor on a Harvard Institute for International Development [HIID] project starting in 1983. In that work there were two parts of the administrative reform area that I worked on. One was decentralization, and another was budgeting. The two of them came together in that, in doing the decentralization work, many of the local authorities were complaining that they couldn't really monitor whether they were getting the money that had been promised in the budget because the budget was not disaggregated by district. So in trying to help them on that, we then discovered that the government's budgeting system was very chaotic; it was all done manually, and it was done mostly in a series of all-nighters just before the presentation session to the Kenyan parliament. So we helped them put a computerized system in place, which helped to address both the district problem and also the overall budgeting problem. That worked pretty well and is still chugging along today. For the next ten years or so, I worked on various other projects in Africa, and one big chunk of work that related to civil service reform was starting in the mid '90s-I was with UNDP [United Nations Development Program], and at that time a number of donors were very concerned about the approach, mainly of the World Bank on civil service reform, which is focusing on downsizing. They felt that this was not going to be successful, but it was also having the negative effect of hollowing out civil services, particularly looking at Africa at that time. So as part of a donor group called the Special Program for Africa [SPA], which was actually chaired by the World Bank, UNDP led an SPA subgroup, the Working Group on Civil Service Reform. It was an effort of about ten donor agencies to find ways of looking at alternatives for civil service reform and coming up with a different approach. The outcome of that was a set of guidelines on civil service reform for Africa which were published in the-I think around 1997 or so. I have an article that was published in '99 that describes it as well. I think at the same time, the World Bank was coming also to the conclusion-the World Bank was represented in our working group-that the downsizing effort was both too limited and also not working. Many of the African countries that had downsized the countries fairly quickly turned around and either hired back the same people or different people and ended up with equally unproductive and no smaller civil services. So that was a kind of cross-cutting bit of work on civil service where we looked at many different countries. Then some more recent work that I have been doing, work in Bangladesh that I have been doing the past couple of years, has not been so much with the civil service ministry-which in the case of Bangladesh is something called the Ministry of Establishment-but it has been working with one of the departments that builds infrastructure in rural areas in Bangladesh. It is one of the largest such departments. In trying to come up with a reform program for that department, a lot of the issues are civil service-type issues: personnel management, compensation, performance management. So we've been making recommendations for them in hopes-well, some of them, we think they'll be able to carry out on their own; others, they'll need to work with the Establishment Ministry and also the Public Service Commission to carry them out. SCHALKWYK: Which departments are these? WESCOTT: It is called the Local Government Engineering Department [LGED]. They have a budget, an annual budget for projects of about 700 million dollars a year. There are, I think, 35 donors that fund projects through this agency. I think that is more donors than any other agency in Bangladesh. The donors use it because they think it is the most effective agency in the country. So they actually give it-although its core expertise is in rural infrastructure and particularly roads, the donors have given them projects in education, health, fisheries, all sorts of other things, because of bad experiences working with other agencies. But LGED, as it is called, still needs to improve a lot. So that is a recent experience. Then the other one that you know about is Vietnam. I don't know if you want me to talk about that. I guess we won't talk about it in any detail because we just had a couple of hours this afternoon about it. But that's a country where I first got involved in 1999 with the Asian Development Bank, and in 2001 was the task leader on a 45 million dollar ADB loan to support the civil service reform program in Vietnam. That's now been going for a few years, and my UNDP work that I was doing late last year in Vietnam was to essentially evaluate how that work had been going since 2001, and then to make recommendations on what should be done in the future. SCHALKWYK: Thanks. Starting on Vietnam, if I could ask you just a couple of questions. WESCOTT: Sure. SCHALKWYK: Did the Vietnamese government approach the ADB, or was there pressure from the outside to implement the reforms? WESCOTT: The Vietnam government approached the ADB. The reforms actually started in Vietnam, at least the current wave of reforms in-I believe it was 1997 or the late '90s. A new prime minister came to office. There was a lot of rising popular unrest about poor quality of government services, corruption in the civil service, and some of these were actually organized protests including war veterans, veterans from what they call the American War, which are very prestigious people. So the government took notice of this. They immediately passed something called the Village Democracy Decree, which was supposed to give more say over local affairs to local communities, but the prime minister also realized that, given the way things worked in Vietnam, he also had to look at the government structure as a whole. So that's how civil service reform came about. Civil service up to that point or just before that point had been run by a party organization, not even by a government organization. So part of the civil service reform was to move over to a government-run rather than party-run civil service. But it was-there were a number of years of deliberations by the government, various high-level committees, some of them supported by donors. But again, the idea came very much from the government itself. Then the request to the ADB came actually in March of 2001 for a loan that they wanted delivered by the end of 2001. Now that's very unusual for ADB. Normally the loans take eighteen months at least to prepare. The loan pipeline had already been agreed-the formal loan pipeline had been agreed in December of 2000 for that year. So it was very unusual that ADB would even consider a request like this. The reason that they did was the urgency with which it was brought to the attention of ADB, and also the fact that other donors were not available to do it. The World Bank was focusing on public financial management rather than civil service reform in Vietnam and did not want to expand their operations any further. SCHALKWYK: So why was there the urgency? Was that based on the public unrest? WESCOTT: I think the unrest had cooled down by that point. I think that there was urgency just because these committees had made their recommendation. They came up with something called the Public Administration Reform Master Plan. That master plan included a component on financial management which the World Bank was going to help them on. But it also included things on compensation, on downsizing, improving training, on e-government, and so forth. I think the government was coming to the realization that the implementation of these reforms was going to be expensive, and so even though UNDP had been an important partner in developing these ideas, UNDP just didn't have the financial resources to back it up. So the ADB instrument that we use is a program loan, so that gives a lot of flexibility to the government in how they use the funds. Again, UNDP doesn't have an instrument like that. So it was the need for financing and the fact that ADB seemed to be the only one available to do it. Japanese, JBIC [Japan Bank for International Cooperation] organization, which also makes loans, could theoretically do it, but in reality they favor hard infrastructure and not softer public administration reforms. So ADB was really the only possibility. SCHALKWYK: And who is taking the lead on managing the reform within the Vietnamese government? WESCOTT: The organization that had been a party organization, General Committee for Organization and Personnel, GCOP, was transformed to become the Ministry of Home Affairs. That happened in 2001, the same time that the ADB was preparing this loan. So they've taken the lead. Unfortunately they're quite a weak ministry in the Vietnamese structure. So the reforms haven't gone quite as quickly or as effectively as was originally hoped for. One of the problems is, what we call MOHA's leadership has not been as effective as it might have been. But nevertheless, they're the designated authority, so we have to work with them. SCHALKWYK: What effect was there in shifting it from a non-government organization to a ministry? WESCOTT: I think it helped to move ahead on other changes, for example one of the things which was just accomplished this year, which is sort of focusing on the passage of civil service law. The party is not regulated by laws; it is regulated by party policies. In order to regulate the civil service by law, you had to have it into a legal entity that could be regulated by laws. I think there are various other things that have happened as well which require-essentially, Vietnam has been adopting a revised version of the practices that are used in many other countries in the world. So in order to do that, they have to move to institutions as part of the government, regulated by laws, the kinds of things that other countries do. I think this transformation has allowed them to move in that direction, but they still have a long way to go. SCHALKWYK: Has there been difficulty in managing the reforms in terms of waiting for the law to be put in place, but still achieving things while the law is being drafted? WESCOTT: That may have been one reason why they haven't achieved things as quickly as they had hoped. But, you know, some of the achievements are still quite remarkable. If you look, for example, the downsizing target. They expressed it in the master program as a reduction in the civil service of 15,000 staff. That is out of a total core civil service of 250,000 or something. So it is a fairly modest target. They were not able to achieve that. However, if you look at number of civil servants in relationship to population, which is a standard indicator that is often used, there has been a sharp drop. If you go back to the mid-1990s, when these reforms started, and then you look at the situation today, there has been a sharp drop. If you look at Vietnam in comparison to comparators, Vietnam had probably more civil servants per capita than any of its comparators when it started the reforms, and now it has relatively lean civil service. Some of that was achieved by hiving off workers, particularly in the service delivery areas, so that they are now semi-autonomous commercial organizations. Anyway, that has been a big achievement. Another one has been- SCHALKWYK: Just before we go on, do you have any of the numbers, what the current per capita is? WESCOTT: Yes, I've got them, actually; you want me to look them up? SCHALKWYK: I can get them from you after the interview, if that would be easier? WESCOTT: Yes, I've got these all documented in different reports. In fact, I think a lot of that stuff is on the Blackboard site for my class, but I can point you to the places where they are if you have a hard time finding it. SCHALKWYK: So, the other successes? WESCOTT: The other one was on raising compensation. This is one of the big weaknesses that was flagged and addressed in the master program. Again, if you look at the numbers, one way that compensation is typically measured is average wages as a percentage of per-capita GDP. When the reforms first started, say in the mid-'90s, it was about one to one. The latest figures we have, I think, for 2006-it is over four to one. So again, there has been a big increase. This is striking because per capita income has gone up so much as well over that period. But still, civil service wages are now more attractive than before. There is a good reason that in a poor country you have to pay civil servants more than the per capita income, because the skills are quite rare. These people have opportunities to work in the private sector for multinational companies and so forth and get even more than that, multiples more than that. So if government wants to hold onto these people, they have to pay something which is attractive enough for them to stay. So I think that was another success. But many of the other things that were tried to achieve, I think-training is still very traditional, lecture-oriented. They haven't been able to make much headway on improving that. There is still the dual training scheme with the National Academy of Public Administration that we're going to be meeting with tonight. But then there is also something called the Ho Chi Minh Academy, which does what they call political training. I've noticed in the letter of participants that this guy who is the President of the National Academy of Public Administration is also the head of the Ho Chi MInh Academy, so that's some attempt at coordination. But the fact that they even have to do this political training shows that they haven't really moved all that far from where they were in 2001, when we started this effort. So I think that there is still a lot that needs to be done. Many of the successes, to the extent that they've taken place, have taken place outside of the framework of the public administration reform program. They've taken place at the sub-national level. One of the most innovative of the sub-national units has been Ho Chi Minh City in the south. As I was saying, looking at the provincial competitive index, the southern jurisdictions in general have been quicker to reform their procedures and so forth than the northern ones. Ho Chi Minh City has been one of the fastest. One of the big innovations they've pioneered is something called the one-stop shop, where if you're an investor, at least in theory, you just have to go to one place to get all the licenses and whatever you need in order to make your investment reality. It's not quite so simple, but anyway apparently it's better than it used to be. So they have this for investors, and they've also tried to roll it out for other services as well. There has been some success, but they still have a long way to go. SCHALKWYK: Is there greater demand-driven reform in the south and in the districts than there is at the national level? WESCOTT: Certainly there's greater demand from the business sector, because that's where most of the business is, in the south. But I think the kind of culture in the south is also more inclined towards reform, whereas in the north, historically it is more follow the orders of the leader. Those cultural differences are starting to change, but change takes a long time. SCHALKWYK: In the discussion this afternoon, somebody mentioned that the Vietnamese government likes to use pilot programs. Is there evidence of where pilot programs have had differing results in different parts of the country? And what do you think some of the causes of that might be? WESCOTT: Oh, yes. The one-stop shops, they were done on a pilot basis and some were more successful than others. I was talking about the whole slew of reforms to improve business competitiveness, or competitiveness to businesses, and again different provinces have had different experiences with that. Some have been spurred to reform faster once they see how far down the list they are in comparison with their other provinces. Another reform that has been piloted in different places with different results is a reform to give greater flexibility over use of budgets. So under this reform, in the past, if you got a budget you had to spend it on whatever the line item said, but under the new reform you're given a lump sum, and as long as you can meet a certain set of performance targets, you can shift the money around. Typically, what successful ones say that they do, is that they find efficiencies where they can cut administrative costs, and then they have more funds left over so that they can pay extra salary supplements for people to improve the salaries. There has been some success in that, particularly in the south, but also some successes in other parts of the country. I think this year, if I remember, or last year, they agreed to-based on the results of the pilot-to roll out the program on a national basis. So there was a prime minister's decree to do that, and so that is the typical way they do things in Vietnam. They try things out and then figure out what works and what doesn't work and roll it out. SCHALKWYK: Do you have any sense of what causes the variations in outcome? WESCOTT: You're dealing with a complex environment when you look at this. There is the cultural difference that I talked about between north and south. There is the urban rural divide, where urban tends to be more successful than rural, generally, because they have better qualified people and more resources. There can also be, to a limited extent in Vietnam, pressure from various constituencies for reform. I mentioned the community-based environmental monitoring groups. They have spurred reforms in certain local areas just because they keep pressing on it. Similarly, where businesses are active, like in Ho Chi Minh City, they've been able to press for reforms that other jurisdictions without that kind of pressure haven't delivered on. SCHALKWYK: Can I shift now to some general questions about civil service reform and try to draw on some of your experiences in all the countries you've worked in? In the emails we exchanged, you mentioned you'd like to talk about windows of opportunity. I was wondering if you could just give me a short overview of some of the sorts of windows of opportunity that you have seen exploited or that you imagine could be exploited during civil service reform. WESCOTT: Okay, one of them: I mentioned this situation in Kenya that I worked on with HIID back in the '80s. The window of opportunity there was that the minister of finance was fired. The reason he was fired was that the budget was a couple of weeks late in its presentation to the parliament. That had never happened before in independent Kenya. This was just untenable for him. So something needed to be done to fix the problem. There were other, lesser opportunities. One was that the financial secretary in the Ministry of Finance was very keen to move to a computer-based budgeting system. He was also very courageous, because when we started doing this the president of the country made a very high-profile speech saying that computers were a plot by the former colonial powers to keep Kenya under their thumb. SCHALKWYK: This was [Daniel arap] Moi? WESCOTT: Yes, this was Moi. He suggested that even electric typewriters were very dangerous because they made the country dependent on these colonial products. So the financial secretary had to be careful. In fact, he was not able to authorize the Harvard project, which was USAID [United States Agency for International Development]-funded, to purchase any computers under the project because the president might find out about it and say, you're going against my policy. So what we were able to do is, we contacted some of the computer companies and got the computer companies to donate their computers to the Ministry of Finance for this work, and that was okay because there was no paperwork involved. The other opportunity that was very important for that project was that some new database software had just been rolled out which allowed us to use PCs for this kind of work and not have to go to the mainframe. The mainframe in Kenya was really very difficult and slow, and there were only a few people who really knew how to use it who were gatekeepers. So being able to go to PCs opened up the whole thing. We could use anybody, we could just train smart clerical staff in the ministry to use them. So there were a bunch of opportunities that all came together that allowed us to have a computer-based budget and get the budget in on time. The new minister was happy, and that worked out well. So that was one target of opportunity. The Vietnam one I mentioned already: new prime minister, protests from war veterans. It was a motivating force to get them to do something. Another interesting case is Cambodia. Here the government, mainly under pressure from donors in this case, since the peace agreement in 1992-various donors, first UNDP, then the World Bank, and others have been trying to get the government to take a more rational approach to managing the civil service. Cambodia, a post-conflict country; civil service was seen as part of the peace process, as a way of providing jobs and patronage opportunities to all the different groups, to get them to buy into the process. SCHALKWYK: So this was the government approach to the civil service. WESCOTT: Yes. So you had a situation where each ministry would have-I can't remember if initially they had two ministers. At least they had two of the highest officials just below minister, secretary-general or something. One from one party and one from the other party. Then there would be pyramids cascading down from them of people from their party and ethnic group and whatever. So this of course led to chaos and also to relatively large and very poorly paid civil service in Cambodia. But because of the political origin of this, it was very hard to fix it. So despite a number of donor projects from UNDP, World Bank and others, not much headway was made. The opportunity came in 2003-2004, when first of all there was a new election, and the ruling party got a better mandate than they'd had before. But that wasn't the main thing. The main thing was, for various reasons a number of donor agencies, led by the World Bank, got together and decided that the crucial entry point here was not civil service reform, but it was public financial management reform. So they carried out initially a diagnostic, public expenditure review and fiduciary assessment, procurement review, pulling together many donor agencies and all the key government counterparts. It was a much more participatory exercise on the government side than it had ever been before. Based on that, they came up with an agreed vision on reform, which included both financial management and civil service. The civil service part of it was that you would start with people that were working on financial management, the key people. They would be in a special merit-based pay program where their pay would be roughly comparable to what you would get working in a donor agency or NGO in Cambodia. Then, as the reform moved along, there would be more work on spreading this special pay arrangement to other parts of the civil service. Everybody got on board, both on the donor side and the government side. So far it has been proceeding along quite well. I haven't seen the very latest evaluations of it, but evaluations up to a couple of years ago are that they have, on the financial management side, something called the platform approach, where you have four levels of reform over a ten-year period. You start with fairly easy things. Once you get those, then you go up to the next things. They've done most of the level one reforms based on the latest information that we have. They've also been moving ahead on the civil service reform. So that was another kind of opportunity, a combination of a new elected government on the Cambodian side and a different approach by donors to work together and also engage the government in different ways that led to that outcome. Another kind of opportunity was in Bangladesh. In Bangladesh they had a very much contested election in 2006, I think December 2006, and the losing party refused to accept the results. There were massive civil outbreaks. So a technocrat-based government backed by the military took over. They did not hand over the power to the political party that had won the election. That government ruled for a couple of years. They had great ambitions on what they were going to do. They actually did achieve a few things. One of them was to completely redo the electoral rolls. They found, I think, about 15 million fraudulent voters out of 80 million or so. So they gave everybody a photo ID and they carried out new elections in December of last year, which were observed by international observers, and they all said that it was fair and free. The newly-elected government has taken power in January, and the losers have so far accepted the fact that they have lost. So that was a successful case of electoral reform. Another thing that was a little less successful but still promising is in the anti-corruption area where this-called the caretaker government, the technocratic government-strengthened an Anti-Corruption Commission that had been there for many years but hadn't done anything, and arrested hundreds of high-level civil servants and politicians for corruption, and managed to successfully prosecute some of them. It is not clear what is going to happen now that the elected government is in power. But anyway, it definitely raised the profile of the corruption issue and made civil servants more aware that there are consequences of this. So that I think was another target of opportunity event. I could go on about others, but I think very often when you find reforms that have worked, even partially worked, there is usually some opportune moment that helped to bring about whatever the limited success was. SCHALKWYK: How do you make sure that these reforms last past the window? WESCOTT: That's always the challenge. I mean, going back just over the things I mentioned, the Kenya one I think has lasted because this deadline for the budget presentation comes up every year, and there's really no other way that they can produce it other than using the system. In fact, what they've done is expanded it to many other areas of government management as well. It has morphed into a kind of results-based management monitoring system. But it is this practical need of the minister to have the numbers ready at a certain date that has ensured that that system has been sustainable. The Vietnam one, I think, the fact that they have the same government, the same party, is a force for stability there. I mean, there are a lot of negatives that you can say about it as well. But once the party makes up its mind to do something, it tends to do it and stick with it. It's kind of a unique situation, though China is probably the same way. It's less so at the local government level because there, I mean, even though it is all the same party, just a change of personality can make a big difference. But at the national level, changes of personalities don't seem to make as much difference as far as policy continuity is concerned. So that gives reason for hope in Vietnam. Bangladesh is a different kind of situation. There, we really don't know how much is going to be sustained- SCHALKWYK: I'd like to ask you about Bangladesh. WESCOTT: -by the new elected government. SCHALKWYK: You also mentioned in your email that your promotion and communication of vision and systems-I wanted to know what you mean by promotion and communication of vision and systems. WESCOTT: Having an idea like budget reform in Kenya or civil service reform in Vietnam is one thing, but in order to make it work you have to be able to convince people in the civil service it is worthwhile, and then ultimately convince the public that it is a good thing. I mean, to the extent that the public has to buy into it, and also you want them to think well of your government as a result of the reforms. So how you promote the idea is very important. I know in Kenya on these budget reforms, there was tremendous amount of resistance when we first started doing it from the budget officers, who were the kind of mainstream guys that work with all the ministries to prepare their budgets and then send them up to the financial secretary and the permanent secretary for approval. SCHALKWYK: Why was there resistance? WESCOTT: There was resistance because they had had pretty much a free hand in negotiating budgets prior to the computerized system. As I said, a lot of it was done in the midnight hours on the back of an envelope. I suspect that there were various unauthorized transactions going on that benefited them as a result of that. But once the whole thing was on a computer, the process of preparing the budget became much more transparent. Also, because none of these guys use computers, at least initially when we started this, there was no way they could fudge the numbers. Once the numbers were in, they were in, unless they had a good reason for putting in different numbers. So there were a number of heated discussions on that one, with the permanent secretary saying this was just a foreign advisor who was bringing this in, this was against the president's policy of using computers, and so forth. But the permanent secretary and financial secretary stuck to their guns-they had a vision. The permanent secretary was a very erudite guy; he had been a Mason Fellow at Harvard. He had traveled to some extent around the international conference circuit and had this idea that he wanted to be running a professional administration in Kenya. So he somehow was able to communicate this vision to these budget officers and tell them that both persuasion and through this kind of "You better go along with this or else you're out" authority, that this was the way things were going to be. These reforms are always going to run into difficulties like that. Each one, there is some different reason why people want to oppose it. So it takes that kind of drive and communication skill to work with the people that are opposing them and somehow persuade them to get on board, and that is-many of these reforms that don't work out, the reason they don't is because they don't have that. SCHALKWYK: What are some of the ways that you've tried to do that? Is that something that can be-? WESCOTT: Well, one thing that I find is very effective is, by trying to put local people in as many positions as possible surrounding the reform, and to really keep the foreign input as much back-channel as possible. I find that when people can see locals, they're doing all the-I mean, ideally, the ideas should come from locals, so you have some local scholar at the university who has done some research and comes out with recommendations that say you should do this, which is exactly what you want to do in the reform-or, if you can't find that, maybe somebody from a neighboring country, so if it's Kenya then somebody from Ghana, or from South Africa, or a country that is kind of higher up the scale, to come and say, we've done this and it has worked very well. Rather than having somebody who looks like me. They may still need people like that, but they should be as far in the background as possible for the reforms to be successful. There's a really interesting article in The New Yorker a couple of weeks ago about this Iranian economist who teaches-who used to teach, he was just fired, apparently, for speaking out too much. But he is trained at Duke and taught at Stanford. Then he went back to Iran. He was apparently a leading economic advisor when-not this current government, but the one before it, [Akbar Hashemi] Rafsanjani came to power and experimented with some much more market-oriented reforms than they had had before. The fact that it was him rather than-I mean if it were the World Bank or anything foreign to Iran, they never would have listened to it. But getting the same kind of ideas that you'd get from the World Bank, but getting them from this Iranian guy, made a huge difference. Apparently, he is still very influential, and they're expecting that there may be a change in government and his ideas may come back again. So that's absolutely crucial, putting a local face on the reforms to making them work. SCHALKWYK: That touches somewhat on the importance of context in planning reforms. WESCOTT: Yes. SCHALKWYK: I'm wondering if you could describe some specific examples of times when donors-or you have been giving advice that has been particularly successful in being sensitive to the national context, and sometimes when you've been unsuccessful in realizing what the context demands. WESCOTT: Yes, OK. There was some work on civil service reform in Ghana that I looked at as part of the independent evaluation group at the World Bank. Ghana had a long history of attempts, both by the government and various projects, supported by international donors, particularly by DFID [Department for International Development] to improve the civil service. There had been a period of downsizing for a while, like many countries did, and that didn't work out. That kind of thing went on hold for a while. Starting in the late '90s, the government came to the World Bank and was very interested in coming up with more state-of-the-art civil service reform programs. So the World Bank prepared a civil service reform, a major project, and it was supported by the vice-president at the time, who was expected to win the election in 2000 and become president. So the World Bank felt that they had their political bases covered. Therefore, this was fully in line with the context. Well, what actually happened was that he lost the election, but this was after the project had been approved. So the project went absolutely nowhere. I looked at the evaluation; it was evaluated as an abject failure. Then more recently, there's been some possible positive signs. As happens in many countries, the government has started to move to some extent anyway on public financial management. There they had had another problem, which was, the support from both DFID and the World Bank was very much on a project basis. So there were lots of disagreements between the donors and the government on which consultant should be hired, whether it should be a Ghanaian consultant or an international consultant, even though the Ghanaian consultant maybe had no experience and so forth. Lots of these kinds of useless administrative arguments. So very little progress had been made. Well, starting around 2000, 2001, the World Bank shifted to a different approach to supporting public financial management through program loans, something they called PRSCs, Poverty Reduction Strategy Credits. So now the money was going into the government budget, and in return the government had to agree to a certain reform program. But it was up to the government to figure out how they were going to achieve it. If they needed a consultant, fine, they get whoever they think they need in order to achieve the result. Well, that turned out to be a much more successful approach in Ghana. They have made some headway on public financial management. I think partly as a result, they are now moving faster on the civil service reform agenda. In 2005 they set up a Ministry of Civil Service to be the focal point for the reforms, and just last month this guy who had been the vice-president who was not elected in 2000, was elected President. Presumably he still has this interest in civil service reform. So maybe finally the World Bank is on the same wavelength as the government of the day in terms of moving ahead on the civil service reform agenda. That showed a couple of things. First of all, the importance of understanding the context-it shows how difficult it is. I mean, how can you predict how an election is going to turn out. Almost nobody has that kind of soothsaying. But it also shows that you need to take a pretty long-term approach to these things, and projects that fail in one decade may be revived in the next decade. The fact that you did a lot of the groundwork ten years before means that some of that stuff-you can be up and running more quickly today, presuming that you now have a more amenable regime in power. One of the ways that the donors have been trying to improve their understanding of context is by doing what they call political economy assessments. This seems like a good thing in theory. I haven't seen a lot of evidence in practice that it has delivered a good result. It may be that it is too new. SCHALKWYK: I was going to ask you, what do you do when you go into a country to work out what are the relevant contextual features that will affect a civil service reform program? WESCOTT: Well, in the case of, I mean this project in Bangladesh; this is one I've recently been involved in. In that case, what we did is, we had a team of people, and one of them was looking at civil service issues, another is looking at procurement, another is looking at engineering design, because this is basically an engineering department. Somebody else is looking at IT. So as far as the guy that is looking at personnel management, organizational structural issues of this department, he reviews all the prior studies that have been done. There actually had been some management reviews done of this agency going back to the early 1980s. The recommendations don't change that much. I mean, they implement some of the things, but a lot of the things continue over the years. So he looked at all that stuff. Then you talk with the people to get an idea of-try out some ideas, see what would be feasible, what things can you do yourself, what things do you need to get clearance from the Public Service Commission, the Ministry of Establishment. Then what are some things that are sort of environmental issues that you have no control over at all. But the fact is that Bangladesh is one of the most corrupt countries in the world, and the politicians are heavily involved in trying to use roads and other rural infrastructure for patronage purposes. There's not much that this agency can do about that, or other government agencies; that's just the environment that you're working in. So you need to understand that. The other thing that we did in that study is, we had two consultants who did what we call a political economy study. That work is confidential in the sense that we don't share that with the client, the local government engineering department. It's just for the benefit of the World Bank and the team. In that we estimate for example what people have to pay to get certain positions in the department, what the rates of bribes are for different types of things, what the unauthorized income is for people at different levels of the structure. That's another way that we inform ourselves about what is really going on as opposed to what these guys in the department are telling us. Then, using all that information, we come up with recommendations. The recommendations are partly what our clients and the agency tell us, partly what we pick up in previous reports, things that still seem like good ideas and haven't been implemented. Then it is constrained to some extent by the political economy analysis which tells us, well, this type of thing is not going to work because of this and that. So that's how we've approached it in that project. SCHALKWYK: So let me continue asking some questions about Bangladesh. Why do you think the local government engineering department is more effective than other parts of the government? WESCOTT: I think there are a number of reasons for it. One is that they had a very dynamic and charismatic leader for a period of about ten years in the 1990s. He was a very kind of western-oriented technocratic kind of guy, but he was also a war hero from their civil war, 1971, so he had credibility. Yet he had these western values. So he is the ideal kind of person to have behind a reform program. People have a lot of respect for him. He was intolerant of flagrant corruption, particularly the kinds of things that donors would notice. So as a result, donors perceived that his agency was the cleanest one around, and the money flooded in. He used part of that money to build these two fifteen-story office buildings that have a big sign on them, LGED, which is a very different profile from the typical government organization of Bangladesh, as you'll see when you go there. Then I think it is also the nature of the work that LGED does, which is engineering-based. There have been studies that show that organizations where the work has what some people call a high degree of specificity, in other words you can actually define the result, the expected results, then you can measure at the end of the day whether the results have been achieved. The kind of work that LGED does is high-specificity work, and therefore it is relatively easy to manage, say, in comparison to a ministry working on primary education, where that is a low-specificity exercise, and therefore it is very difficult to manage for results in that sort of a business. So it a combination of leadership, nature of the work, and then the fact that donors have come on board. Partly it is because in the kingdom of the blind, the one-eyed is king. Other alternatives are so bleak that despite the problems of LGED they're still the best ones around. SCHALKWYK: Has that been maintained since he left? WESCOTT: People say there has been some decline in the standard, and that's one of the reasons why this risk assessment that I've been leading was commissioned. LGED often points to a section of a report published by the World Bank in 1996, saying LGED was the best department in Bangladesh, and here are the ten reasons-bang, bang, bang-why it is. They say, how can the World Bank now come to us and talk about operational risks when you said we were the best agency around. Well, we try to say, you may still be the best agency around, but in order to keep that position we think it is important to always be improving. So to some extent they've seen that. One of the challenges for our study is, when we first started out, LGED wanted to have a counterpart on their side that would do the kind of day-to-day work with us. So they appointed this chief engineer, who is the one who had been their dynamo during the 1990s, who had retired, but we hired him as a consultant to be this interface. Everything worked great for our project under his leadership. Unfortunately in early September, he was visiting his cousins in New Jersey and died of a heart attack, so we lost him. He was only like my age. We have a new guy who is in there, another chief engineer, and he's been doing a good job, but he doesn't have the same stature as this guy that we lost, so that's a bit of a concern. Again, if LGED is even going to get back to where it was in the '90s, much less do better, they've got to do some of the things that we're recommending; at least that's our best judgment. But whether there will be the leadership to push that stuff through is hard to say. SCHALKWYK: So why in general does Bangladesh struggle to get reforms implemented-reform recommendations implemented? Do you have any sense as to why this is true? WESCOTT: I've written stuff on that. There's a combination of things. I think, first of all you have to be careful in how you phrase a sentence saying that Bangladesh has struggled, because it begs the question: struggled in comparison to what? If you look at Bangladesh in comparison to comparator countries who are, say, about the same level of development on the human development index or whatever indicator you use, you find that Bangladesh is about the same as its comparators in terms of, say, the World Bank governance indicators. I can show you this; I have this in one of my reports. The comparator countries that we selected for Bangladesh were-at least some of them are Ghana, Nepal, Pakistan, Sudan, maybe a couple of others. Of those countries, when you look at the kind of margin of error of each of the World Bank governance indicators, Bangladesh is well within the range of all its comparators, except for Ghana. Ghana is the one that is ahead. The question in many ways-Ghana is quite a different kind of country from Bangladesh, but in terms of human developments about the same. The only area where Bangladesh excels from its other comparators, besides Ghana, is in the World Bank governance indicator on public participation. That is because of their strong NGOs, BRAC [Bangladesh Rural Advancement Committee], Grameen Bank, and so forth among other things. Strong NGOs is an indicator of this kind of strong participatory climate. So to the extent that Bangladesh has been able to excel its competitors on some indicators, for example on economic growth in the last few years, Bangladesh has been doing pretty well, I mean a lot better than Sudan and Nepal. Probably better than Pakistan. Ghana has also been doing well in growth. But the success of Bangladesh is in part because the strong NGOs are able to find ways of working with government authorities to strengthen the weak service delivery on the government side and deliver basic education services, basic health services. You have this relatively strong engineering department that is able to deliver roads and jetties, because a lot of the transport is by water transport, building rural market towns to get the farm-to-market thing working. So a combination of whatever work the government can put in, plus support from these NGOs, has led to what some people call the development paradox in Bangladesh: that you have weak governance on the one hand, but pretty strong growth numbers on the other in recent years. So Bangladesh's government is weak-for one reason, because it's a poor country. If you're a poor country you can't pay civil servants very much; levels of education are low, health standards are low, so this affects capacity. But there are other reasons as well. The political system that has been set up is kind of a winner-takes-all political massacre that happens each time there is an election. So the winner of the election becomes kind of authoritarian ruler for their time in office. Then people become disgusted, vote them out of office, the new one comes in, and they become their own authoritarian ruler. This of course is not good for continuity, policy continuity; it's not good for coordination among different groups, because one major group is left out. The elections in the past have been quite close; this last election was not very close. So there's a lot more of the country on the side of the ruling party now than there had been in the past, and that's a good thing, looking ahead. Another factor in Bangladesh is that there is a lot of migration, labor migration of various sorts from rural to city, from city to overseas, and then moving around rural areas. One of the unusual features of Bangladesh is that land is created and disappears all the time because of the river system. So your farmer on a certain plot of land-that land may all of a sudden disappear, and then you've got to move somewhere else. Well, all this constant moving around, studies have shown, makes it very difficult to run elections, for one thing. It's hard to keep track of migrant workers, and has other implications for governing the country as well. It makes it more difficult than it would be to govern other countries. So those are some of the reasons why government is weak. SCHALKWYK: Perhaps, more specifically then, there have been a number of reform recommendations, the Public Administration Reform Commission in the late '90s, and these haven't-it seems as though very few of these have been implemented. WESCOTT: Right. SCHALKWYK: Do you have any sense as to why there is a particular resistance to civil service or public administrative reform? Because certainly countries like Tanzania may be in a similar state of development as Bangladesh, but have been more successful in doing public administrative reform. WESCOTT: Yes. I think there are a number of reasons, but you should also talk with people like Tanveer [Hussain] that have been right in the thick of it. My own view is that, first of all, there hasn't been a strong enough political impetus to push through the reforms. The politicians have many things on their plate. Of course, all these commissions are called for by the leader of the day. So they are not things that come out of thin air. There is some kind of political support for them. But when it comes to actually implementing the regulations, there is only so much that the politicians can do. One thing that they could do would be to make a lot of speeches and try to hold civil servants accountable-and hold particularly ministers accountable for these things-and I don't think that they've done enough of that. But in the Bangladesh context, many of the decisions and work have to be done by officials in the Ministry of Establishment and Public Service Commission, and these officials have just been unwilling to do that work. They see their position threatened for various reasons. They're very conservative. So simply, they say they're going to do things and then they don't, or they say it is more difficult, we need more time. So I think those ministries have been dragging their feet. They haven't had enough fire under them from politicians to force them to move forward. It has been different in the public finance sector. I mean, many of the donors also complain the financial management reforms have been going too slowly as well. But there has been more push to get those reforms through, at least the reforms that are of a fairly technical nature. There has been more success in that than there has been on the civil service side. SCHALKWYK: What has your involvement in Nepal been? WESCOTT: When I was with ADB, I went there a couple of times and was following ADB's effort in Nepal. SCHALKWYK: When was that? WESCOTT: That was from '99 onwards. Most of the time following Nepal was up until maybe 2002, when I switched to another job in ADB. But the approach that ADB was pushing-and I think there were some other donors involved as well. Nepal was, of course-at that time there was still civil conflict going on, so conditions for civil service reform in those situations are always pretty difficult. What we agreed on with the government was a very low-key approach to start with. We support a civil service census because they had no personnel information system. They had no idea how many people were getting paid, or what they were doing or anything. So that was one thing that we did. It was quite difficult to get them to agree to do it, because it might reveal ghost workers and other kinds of arrangements that people didn't want revealed. But they did agree to it in the end. The second thing we got them to agree to was to freeze vacancies. At least in certain areas, the civil service seemed to be grossly overstaffed, so freezing vacancies, you didn't have to worry about downsizing, but at least gradually through attrition you could get the numbers down a bit. The third thing they agreed on was, to some extent, outsourcing of services that had been government functions could now be switched over to outsourcing. So that was a way of taking the pressure off the service because you were freezing vacancies but you could still deliver the services through outsourcing. Those things went along fairly well. The idea was always, by doing these fairly low profile things, you might build up a critical mass of people to work on more comprehensive civil service reforms, and also you could hopefully get some improvement in government performance that people would notice and therefore start to demand more reform to get more of an improvement. I haven't been following Nepal closely over the last few years, so I'm not sure how things are going. I know a good friend of mine, Robert Piper, is now head of the UN office there. SCHALKWYK: I'll be meeting him. WESCOTT: He'd be a good person to know how things are coming now, because UNDP is usually involved in these things. But I think they-during the time that I was watching it from ADB, I think they moved along reasonably well despite some mistakes that ADB made. ADB had a project leader who was this fiery woman from Singapore, who went in and said, this is the way we do it in Singapore; this is the way you have to do it here. Of course, that didn't set too well with the people in Nepal. SCHALKWYK: It wasn't context- WESCOTT: No, not context-specific. But anyway, despite that lack of sensitivity, I think we made some good headway there, and hopefully it is still moving ahead today. SCHALKWYK: You mentioned listening to local ideas. Are there any local ideas that seem particularly innovative or successful that you've come across? WESCOTT: The local ideas usually aren't that different from ideas that you get around the world, because there's not a lot of new ideas out there. The important thing is not that they're innovative in my mind; it is that they're local. If you have people that have good solid training in public management theory and practice that are there somewhere in the system to advise people, then you'll get reasonable ideas put forward. There's no one-size-fits-all for civil service anyway. As long as people have some plausible good practices that they're talking about that come from other countries that are more or less like them, that's the kind of idea that you need. Then you need somebody to pick up on it and keep the fire under whoever the recalcitrant ministries are that need to actually carry things out. The actual innovations-this thing we did in Kenya on budgeting, it was very innovative at the time, using PCs to prepare a national budget, but you know, the idea, that particular idea actually came from the foreign advisors. But the Kenyans quickly internalized it and made it their idea. That's fine when that kind of thing happens. As long as they end up thinking of it as their idea, then it doesn't really matter where it comes from. SCHALKWYK: Lastly, you've done some work in Timor Leste, Laos, and Eritrea, and these are very different areas from places which have long-standing, if somewhat problematic civil services. WESCOTT: Yes. SCHALKWYK: How is it different working in places like that? What sort of issues arise that aren't similar to more stable countries? WESCOTT: The capacity issue is a big one that comes up in those countries. I know in Eritrea, the senior government was mainly made up of people who had been freedom fighters in the trenches for twenty or thirty years. A lot of them had no formal education at all. SCHALKWYK: When were you in Eritrea? WESCOTT: I worked there off and on from '92 to '96. At that time, Eritrea was considered a reasonable performer among the low-end fragile-state-type places. I had real opportunity working there. This is when I was with UNDP. We somehow got in touch with this guy named Sam Harrigot, who had been a very close confidant of Haile Selassie's but he was originally from Eritrea, which was then part of Ethiopia. He was then locked up by the Derg government that got rid of Haile Selassie and was in the dungeon of the palace for ten years or so. Eventually he got out. He had been a good friend of Ralph Nader's at Harvard Law School. So Ralph Nader helped him out and helped to kind of rehabilitate him. Then when Eritrea got its independence, he went back as a senior advisor to the government. He brought in UNDP to help on the civil service side. One of the things that we helped him to organize is dealing with the capacity problem there. They set up a program with the Open University in the UK to do kind of a quick course in public management. Taking these people, some of whom had barely been to primary school, and getting to the point where they could understand some of the basic management theories and go through some case studies and that kind of thing. They had courses in Asmara in the evening and on weekends, and then some time that they spend at the Open University campus, Milton Keynes in the UK. That program seemed to be going quite well. I never saw an evaluation of it, but it seemed to be an interesting concept anyway for dealing with a situation where you had to rapidly bring people up to speed. But they also had to serve as minister of agriculture, or whatever they were doing. The work that I did in Timor Leste was also focusing on severe capacity constraints. In that case, you didn't have quite as bleak a situation as in Eritrea. What had happened there was, in 1999, when they had the civil war, most of the people that were in university were in university in Indonesia. Since they were fighting Indonesia, all these people were sent home. So their university education was interrupted. A lot of the civil service is currently made up of these people who were studying-. Civil services in developing countries tend to get people that are five or ten years out of university. Those are typically the most productive people in a lot of these civil services, because after that the best ones go off to the private sector or whatever. Well, in Timor Leste these people had their education interrupted. So what I did was-Timor Leste has a little bit of money now because they have oil and gas reserves, and they've worked out an agreement with Australia-although Australia has been fighting tooth and nail against them-that Timor Leste can get a couple of billion dollars a year out of this, which for a small country is significant. So I worked out a program-this is, actually, in the Ministry of Finance, because that's the one that the World Bank was working with, but it really has implications for the civil service as a whole-to take people in key positions and, first of all, work with them to get their English-language skills up to speed, because most of the potential universities are in English. Or, alternatively, if they can get their Bahasa skills up, then they can go to Indonesia. Then to place them in appropriate universities for either undergraduate or graduate training in relevant areas, mentor them while they're there to make sure that they maintain their contact with Timor Leste, and then work with them when they come back to integrate them into the system. I designed that project in May to July, and they tell me that they hired somebody to manage it now who is somebody from the Timor Leste diaspora who came back from Australia, and that it is going along well. I hadn't mentioned the diaspora, but that is another theme that has come across in a lot of my work, that countries, not just in civil service reforms but in reforms across the board, find-. If there are ways that you can integrate people from the diaspora back into the home country, that can be a way of localizing the reform message, because these people tend to be relatively sophisticated because they've been living outside. Yet they've got a local face, they speak the local language. Sometimes there's resistance. In Afghanistan there has been a lot of resistance to the diaspora people who have come back, for various reasons. There's some resistance in every country to these people. In fact, in Vietnam, the problem is that most of them come from the south. For government work that is a difficulty, so most of them, when they come back, they work in the private sector. People from the diaspora can be a huge resource if you can find a way of using them productively. For Timor Leste they'd be nowhere if it wasn't for the diaspora. This minister of finance had lived in Melbourne for ten, fifteen years. A lot of them have dual passports. They're really terrifically valuable to these countries if you can persuade them to come back. SCHALKWYK: Do you have anything else to add? WESCOTT: No, I think I'm very interested in seeing how this comes out. How many people are you interviewing like this? SCHALKWYK: We're hoping to interview as many as we can. We currently have about 300 people interviewed in our policing program, and it will be great to get this program up to that level in as many different countries as we can that we can afford to go to. WESCOTT: It will be interesting to see the results. When do you expect the final outputs of this to be available? SCHALKWYK: Well, I'll stop the interview, thank you very much. Innovations for Successful Societies Series: Civil Service Oral History Program Interview number: G-7 ______________________________________________________________________ 18 Use of this transcript is governed by ISS Terms of Use, available at www.princeton.edu/successfulsocieties Use of this transcript is governed by ISS Terms of Use, available at www.princeton.edu/successfulsocieties