WORLD BANK HISTORY PROJECT Brookings Institution Transcript of interview with LYN SQUIRE Date: January 16, 1992 Washington, D.C. By: Richard Webb, Devesh Kapur 1 FOREWORD The following is a transcript of an oral interview conducted by the authors of the World Bank’s fiftieth anniversary history: John P. Lewis, Richard Webb and Devesh Kapur, The World Bank: Its First Half Century, Washington, DC: Brookings Institution Press, 1997. It is not a formal oral history, and it is not a systematic overview of the work of the person interviewed. At times the authors discussed the planned publication itself and the sources that should be consulted; at other times they talked about persons and publications extraneous to the Bank. Some interview tapes and transcripts begin and end abruptly. Nevertheless, the World Bank Group Archives believes that this transcript may be of interest to researchers and makes it available for public use. Lyn Squire January 16, 1992 – Final Edited 2 [Begin Tape 1, Side A]1 [There are large sections of this conversation in which only occasional words or phrases are audible; the recorder is apparently too far away from everyone.] WEBB: We have to hand over to our publisher in essentially two years. But it’s interesting, at least at this stage before we have to really write. KAPUR: The volume two drafts are to be done by the end of the year [inaudible] WEBB: Right now I'm sitting down to write a chapter which will be covering poverty and just trying to draw up an outline—this is the sketch of an outline, and I'll show you-- the parts A, B, and C here are really a list of ingredients, the things that should be kept in mind. One, A is what the Bank's objectives have been and how they've been defined; B are what the Bank itself directly does, operations; and C are how and to what extent the Bank has influenced countries. And then F is a sketch of an historical account, how they went about actually achieving it, the story of it. There's something here that—well, any reactions that you have to this would be, any thoughts, things that should be included, any dates, key operational persons or a certain kind of reports or projects that were done. SQUIRE: First of all, one quick reaction is—I take it the report that’s--I think that the project approach is the key here, which is—just looking at the first page, the emphasis is going to be on projects, and I think that’d be a big mistake. WEBB: I agree with you entirely. I think I understand the point. There’s a little paragraph here that I didn't include in this . . SQUIRE: Ah! The key paragraph! [Laughter] WEBB: . . which is some thoughts on the bottom line, what this might all add up to, because it's just so early yet. But my sense is that there is a whole variety, there’s been a lot of bustle here, and the Bank has done [inaudible] trying to do those, do something on poverty, it hasn’t added up to much. There’s a lot of rhetoric, a lot of operational noise, but it hasn’t added up to much. And I think that that may be the key, that the Bank really hasn't done made much of an effort to work on what the countries themselves are doing on a broader scale [inaudible] If you just look at what it does, quote unquote, for “poverty” is much smaller still. KAPUR: To add on to Richard’s point, it only seems that only recently or more recently has the Bank sort of started to looking a bit more carefully at the poverty expenditures of the countries themselves and how those issues and analyses of poverty, because in the end, if you believe that it’s the countries themselves that will have to do the majority of 1 Original transcript by Brookings Institution World Bank history project; original insertions are in [ ]. Insertions added by World Bank Group Archives are in italics in [ ]. Lyn Squire January 16, 1992 – Final Edited 3 the work, then it's what they do rather than what the Bank does per se through its own projects of the Bank and that’s the . . SQUIRE: I think that's right, but it’s—I think the point is even broader than public expenditures because the totality of the public policy, in terms of both the lives of poor people and--you know, I may choose to have a development strategy in which I sort of squeeze the agricultural sector. It may not have anything to do with public spending, but through exchange rates, marketing boards or things like whatever, and that may have a greater influence on the lives of the majority of poor people in the country than anything else that the government’s doing. WEBB: I think that it’s--we're obliged to go through what the Bank has actually done. [both speaking at once] SQUIRE: Sure. Fair enough. Okay. Fine. But then I think, if what I’ve just said is correct, you're obliged to review not just project interventions but whether in the course of ESW [economic and sector work], other advice and policy guidance, we've actually said things which are conducive to the reduction of poverty. And, you know, I'm not making a judgment on this point, but conceivably what we've done through other routes rather than lending may be more important or as important. So I think it’s, you know, it is all of the government's policies that influence what's happening to poor people, especially in many sub-Saharan African countries where the poor may be 60 or 50 percent of the population. WEBB: I am especially interested in that account. I suspect that rather than one or two big projects that are called poverty projects and look big in the Bank's portfolio [inaudible] Rather than those, most of the project work will [inaudible] anyway. It has been much smaller things that--sometimes it’s ESW or higher level conversations with country politicians on poverty can open your eyes sometimes to phrases that come in. “Absolute poverty” didn’t exist as a phrase even when we were working [inaudible] rural was extremely [inaudible] a phrase. As a phrase it does make a difference. Before it was all “distribution,” “inequities,” and so those phrases were convenient [inaudible] bashing the rich. But “absolute poverty” obligates, especially [inaudible] Those things can also be more concrete, things that have to do with the design of approach, that sometimes a small bit of thought [inaudible] appropriate. I’d really be interested in finding examples. SQUIRE: Of interventions? WEBB: Of countries where we can say, “Look, here the Bank's conversations, the Bank's rhetoric, a little bit of arm-twisting maybe, has had some effect. This country really is now more [inaudible]” Maybe no country exists, but if there were it would be very interesting. And an example with the projects [inaudible] SQUIRE: Take a country which has done really well, at least as far as we can see, in terms of helping the poor. Let’s take Indonesia, okay? I think that--first of all, it's always difficult to say what the Bank’s influence was in anything, whether it’s poverty or general Lyn Squire January 16, 1992 – Final Edited 4 development. Put that aside. Let's say the Bank's input into Indonesia is suggesting guidance on rural policies, maybe public expenditures and so on and so forth. My guess is that that was not immediately motivated by a concern with poverty. It's a more general concern with development across the board. But the fact of the matter is that Indonesia's development has been such that it’s provided substantial benefits to the poor. And in many ways what I think we were trying to do in the WDR [World Development Report] on poverty was to say, “Look, you know you really do have to worry about development strategy in a country.” I mean that’s really the key to solving the poverty problem. You don't start with and focus in on the poverty problem and somehow the country's development strategy is separate. It is the country’s development strategy, so that a lot of advice and assistance from the Bank—you know, whether you think they’ve had influence is a separate point--but a lot of advice and assistance from the Bank I think is directed towards overall development which, in my view, is done the right approach. If done right, it’s going to be largely beneficial to everyone [inaudible] So, you know, if you’re going to try to look for a situation where we went to a country and said, “Okay, now we're going to concentrate on the poor people in this country and figure out how we should design policies to help them,” I don't think you'll ever find that. KAPUR: In a way hasn’t that been one of the tensions within the Bank, this whole thing between direct and indirect [inaudible] SQUIRE: Absolutely. KAPUR: I mean, even ostensibly the bringing of the [inaudible] in the ‘80s doesn't, that had ostensibly an underlying philosophy of general change, change in the general direction which would translate into [inaudible] poverty. How is this tension—and it’s still perhaps a tension or has that been [inaudible] between direct and indirect? SQUIRE: I think—and this is partly the result of the experience of the ‘80s, there is now general recognition that project interventions by themselves are never going to solve any particular problem unless there is a sensible overall economic framework. I think that point is more or less generally accepted in the Bank. Now, what does that actually mean in practice? Well, in lots of situations where with a little more distance we might say the economic framework's not very good, we're still going to be in there doing projects. So in a sense, you know, the bread-and-butter business of the Bank will always continue. I think the general point is accepted; there’s no question about it. How all these different interest groups in the Bank resolve the issues that arise and what will actually be done about this . . . WEBB: I always thought there was—that [inaudible] is a different attitude. It's not that you’re--the basic business is really not that different but rather that you're going about your basic [inaudible] on poverty. Before it was [inaudible] with the idea that there are bound to be ways in which you design that would make sense [inaudible]. Lyn Squire January 16, 1992 – Final Edited 5 SQUIRE: I think that’s right. WEBB: If you don't bother about that, you're going to miss a lot of the poverty. SQUIRE: You’ve seen this poverty policy paper? KAPUR: In ‘91, the one that you wrote? SQUIRE: Yeah. KAPUR: After the WDR? SQUIRE: Right. Yeah. WEBB: But that's really assistance strategies? No? SQUIRE: No. [seem to be looking at documents] [inaudible] It’s probably the same; it looks a little nicer. WEBB: Oh, I see. Okay. SQUIRE: The idea there was that, yeah, we do need to worry about poverty but not simply in the context of our interventions, but rather, the argument was, first of all, we need to understand what is happening in the country as far as the poor people are concerned and how are policies within the country directed towards reducing poverty, and that's what we call the multi-assessments. And from that we then go and ask, “What is it that the Bank should be doing?” It might be that the answer is, “Well, we should be doing a telecommunications project,” which on the face of it has absolutely nothing whatsoever to do with poor people but it should be seen in the context of what the country itself is doing. Maybe they're doing everything else in the social sector that we think is important, but telecommunications may need help, and that's where we should focus our effort. And there would be no particular effort to make that relevant to poor people. That wouldn't be the objective. But at the end of the day you should be able to sit back and look at the Bank's overall approach to a country and ask the question, “Is it addressing the needs of the poor?” or, you know, to phrase it slightly differently, “Is it, given what the country itself is doing, addressing the needs of the poor?” So it should be a broad view. Within that it may be that you end up saying, “Okay, we do need to have some projects which are in some sense very much targeted to the poor,” but we're not saying that every project should have a poverty orientation. We're not even saying that somewhere in every country's program the Bank must have a poverty-oriented project. What we're saying is that when we look at the country or at the Bank’s strategy in any particular country, that Lyn Squire January 16, 1992 – Final Edited 6 the issue of poverty has been thought about, and we've seen what the country's doing, and we've figured out, therefore, what we ought to be doing. WEBB: Yeah, I think I understand that. I think again it's a matter of attitude, not just with respect to [inaudible] Every time I get into a discussion of poverty, that’s really whether poverty is present in their thoughts. SQUIRE: Exactly. That's why I get a little concerned when—and we do this, and the Board [of Executive Directors] wants to do this when they want to know what percent of our lending is for poverty purposes, and I think that's not a very fruitful way of seeing, and in some countries that percentage may literally be zero, and yet you could feel very comfortable that the right things are being done to help the poor people in the country. And in another country where the poor may really be suffering, a large part of our program might have a very strong poverty orientation, a direct poverty orientation. So in the first country we have the poor doing really well and the Bank not, you know, on these scores, not showing much in terms of its lending as being poverty oriented. In the other country the poor are doing very badly and it’s showing more things directed towards the poor. It's got to be the Board view which prevails, but the Board is hung up on measurement, you know, it’s how many dollars are allocated to the poverty objective. WEBB: We need examples. SQUIRE: Of what? WEBB: Of what you’re talking about. One kind of country might be doing [inaudible] might be doing the other but, even though they don’t get [inaudible] KAPUR: [inaudible section] China? SQUIRE: Well, I can't--I mean, every sort of country context is so complex it’s difficult to quickly come up with a single example. But presumably you're looking at--what period are you looking at? WEBB: Fifty years. KAPUR: In any case, it does strike me that they must have [inaudible] China. To a great extent poverty is a political issue in countries because of political at the top [inaudible] on the poverty per se. And the Bank for a long time has publicly maintained, at least until recently, that [inaudible] political institution and that’s all the [inaudible] Has it affected the way intervention has done? SQUIRE: I just have great difficulty knowing exactly what the Bank’s position is on political issues now, but if I understand [Ibrahim F.I.] Shihata's paper on this, we are still taking a limited approach—it was still a very, very cautious statement about what we can do. It basically came down to say, “Well, look, critical factors influence the way in which economies operate so on and so forth and we must be aware of how the political situation Lyn Squire January 16, 1992 – Final Edited 7 is influencing those aspects.” But it's very much a focus on the economics of poverty, not a direct questioning of political motives or political actions. And, you know, if you look at some of the countries that have done reasonably well--including helping the poor- -I would certainly not describe them as democratic systems from the government point. So I've personally not felt that that’s interfering. As I said, I don’t get too excited about it. WEBB: Could we talk a little about the distribution weights? When, how and when were these program [inaudible] SQUIRE: Second half of the ’70s was the height of the discussion. WEBB: This was before ’75 or . . . I don’t remember because I arrived in July ’75. SQUIRE: My recollection is that—yeah, it would have been before ’75 that the discussion started for sure. WEBB: You were then working--were you already in that division? SQUIRE: I can't remember. I stayed with [Herman G.] van der Tak for some time after the completion of the white paper [Economic Analysis of Projects]. I know that it was a year or so. I would have thought that sometime in ’74 that . . [Interruption] SQUIRE: . . whatever the policy committee was, in ’78 or something. WEBB: Where was van der Tak? SQUIRE: We had a centralized department under Warren Baum which was responsible for quality control of all our projects. Every single project was reviewed by a committee chaired by Warren. At that committee he would have his economic advisor, who was Herman, his procurement adviser, his engineer or whoever else, and the team that prepared the project would actually come and sit at the table. They would go as a team. WEBB: And you would sit with van der Tak [inaudible] SQUIRE: It was fascinating. WEBB: Yeah. Very educational! SQUIRE: Absolutely. So, anyway, that’s where Herman was. WEBB: And it was--and then he was--was he asked to develop this or did it all come out of his own initiative? Lyn Squire January 16, 1992 – Final Edited 8 SQUIRE: He had been away for a year in, I think, in England somewhere, and may have exposed to some of these ideas, because when he came back and took that job, he obviously felt a need to rethink how we were approaching the economic analysis of projects, and I would guess he was the guiding force. WEBB: Is he still here? SQUIRE: He's retired from the Bank, but he still does things for it. He came with me recently to Mexico. WEBB: We should ask him where he was in England [inaudible] sources of that. SQUIRE: You see, because at that time all the [I.M.D.] Little and [James A.] Mirrlees and all that stuff had been going on in Oxford. I don't know whether Herman was there at Oxford or what. And I was at Cambridge doing my Ph.D. when all that was going on. I had lots of contact with those guys. WEBB: You lived with the [inaudible] SQUIRE: Sure. WEBB: It was part of the . . SQUIRE: Absolutely. WEBB: I see. So it would have a more general topic. SQUIRE: All that Herman and I did--I mean, it's an exercise in exposition. I mean, you know, trying to think through what Little and Mirrlees were saying, what Amartya Sen was saying, and figure out what it might mean for the Bank, putting it a little bit more in kind of Bank language. WEBB: Now, you say you were working at [inaudible]. Was this a year away during your period with him? SQUIRE: No. Before. WEBB: It was all before. Well, when you met him you didn't know that [inaudible] SQUIRE: No. It may not have even been in England. I think it was. WEBB: Okay. And when you met him and started working at figures, was Herman very much on [inaudible] certain people? Was there a sense of change, of innovation here that changed things [inaudible] Lyn Squire January 16, 1992 – Final Edited 9 SQUIRE: I don’t know [inaudible] characterized by, but I would assume that’s a fairly radical--it’s a discontinuity. WEBB: I mean, ex-poste it was. But was there sense then that it was? SQUIRE: I'm really sort of scratching my head about that. There was very little economic guidance to staff at that time. There were guidelines, but they were pretty thin, and, you know, quite apart from distribution weights, there was a clear need to revise, revitalize, improve, bring up to date the guidelines that we had. WEBB: Did you have a sense that you were taking part in something radical? SQUIRE: Yeah, in the sense that it was a fairly major shift. WEBB: Yeah. And did there begin to be quick reactions? SQUIRE: Unbelievable. WEBB: Both good and bad? SQUIRE: I mean the split was very clear. Those who, if you like, intellectuals in the Bank, many who--the research department, for example--appreciated the merits of the economics that were implicit in what was being suggested. But then operations—not one hundred percent but certainly this was the dominant view--operations clearly felt that this would entail additional work for little benefit—it was an obvious cost for them, and it was not clear to them that there would be an advantage . . . KAPUR: What about the . . SQUIRE: Sorry. Let me just finish up on that point, if I may. The point being that there was a view that the economic analysis of a project was not the decisive factor in determining whether the project goes forward; therefore, you know, refining the analysis is not an obvious—an obviously wise use of human resources. WEBB: Do you remember any very forceful statements about this issue that had an effect on--at that time? SQUIRE: You mean effect in the sense that . . . WEBB: By Warren, for instance? Burke Knapp? SQUIRE: No. WEBB: Someone who was fairly articulate [inaudible] Lyn Squire January 16, 1992 – Final Edited 10 SQUIRE: People in the center were lukewarm to positive, but were really, I think, really quite . . . WEBB: Warren? SQUIRE: Obviously it went forward under Warren’s . . . WEBB: When he was . . . Do you know anything about Burke Knapp's thinking, policies? SQUIRE: Yeah, this is not something which is a great interest of his. He had his own, critical but perfectly comfortable [inaudible]. He wasn't very closely involved, is the point I’m making, but certainly not saying, “Okay, we can’t do this.” So I would say he was positive but somewhat distant. KAPUR: To jump the gun, but in the reaction to Little-Mirrlees, you look back on that analysis, there seemed to be a distinct view held among operations people that this never a concept that could be really operationalized. Was that a view held then also, that other than the relative advantages but the sheer effort needed was just not . . . SQUIRE: No. There's two parts to my answer on that. One is—I think there was definitely a view that the distribution weights were problems because of course it was a problem, first because it introduced statistical subjectivity into the analysis, because there's no scientific way to arriving at the weights and because actually identifying the beneficiaries of many of our projects is a genuine practical problem. So I think there was definitely a view that that part of the analysis was either impossible or at best going to be very, very difficult. But--and I think it's unfortunate that this somehow got lost in the debate on distribution weights--we were talking about improvements in the basic way in which we approached the evaluation of projects, whether or not we had distribution weights, and I don't think we’ve ever really been questioned on the feasibility of that. It kind of got lost in the . . . WEBB: I’m sorry--whether or not the other aspects of the proposal? SQUIRE: Yeah, I mean the question was . . WEBB: The non-distributional aspects? SQUIRE: Right. The feasibility, the operational feasibility of that was--I don't think was seriously questioned. I mean there may be some people who did, but not across the board by any means, whereas there was a fairly serious concern that the distribution weight system was just not feasible. WEBB: But did anyone ever suggest ways to improve the—objectively to [inaudible] deposit, what manner of revenue relative but what one could look or relative view, attach Lyn Squire January 16, 1992 – Final Edited 11 an equal value to a one percent change [inaudible] Was it ever part of the discussion or versions of the distribution weight proposal? SQUIRE: So a dollar change? KAPUR: Right, so I mean in the sense that relative to a dollar of genuine costs, or rather than a dollar of genuine costs, so a percentage. WEBB: It’s fair. SQUIRE: You mean I should probably just say a one percent increase in any recent increase in income is the same for everyone. WEBB: Yes. SQUIRE: Really? So if I'm a very rich guy, my one percent is going to be a thousand dollars. If I'm a very poor guy, my one percent is going to be maybe a dollar. WEBB: That’s right. SQUIRE: We’d value those equally? WEBB: Yeah. SQUIRE: No, I can’t see any logic why we would do that. WEBB: Well, it's a . . . SQUIRE: That would be a grossly inegalitarian system. WEBB: More gross than not having any distribution support? SQUIRE: Yes, because you're putting a greater weight on the rich guy’s income, not the poor guy’s. WEBB: But you are anyway if you don't have distribution weights. SQUIRE: You're basically saying that we don't care who gets the income. So, if you're very poor and I'm very rich, you get a dollar, I get a dollar, the same weight. WEBB: But if you have a development cost where the distribution of income is going to change, then in effect everyone is going to have--but not necessarily, but let’s say that everyone is getting the same proportion. SQUIRE: Okay. Right. Lyn Squire January 16, 1992 – Final Edited 12 WEBB: It would be that case. SQUIRE: Yeah, if everyone gets the same proportional income you could make distribution constant, right? Then had you come with the kind of distribution weights that we were proposing and applied it to this constant shift in income distribution, you then would not come out looking very good because we were attaching our weights to the absolute increase in everybody's income. Cost of distribution means that the relative increases stay the same, so the absolute increase for the rich guy is much greater than the absolute increase for the poor. We would have been much more radical than that. WEBB: Well, of course, I understand that. But the way things are described this distribution of the weights is broad enough [inaudible] SQUIRE: That statement's not correct. WEBB: Okay. What is actually done in practice is something like that. If people look at growth without considering distribution, in fact when they look at the distribution, they say, “Well, that’s really is changing, so growth is [inaudible] neutral.” “Neutral” there— “it’s not getting worse” is what people are saying. SQUIRE: Not in the context of projects. WEBB: No. But it's--the distribution is worsening. People are saying, “Well, it's not getting worse.” They’re valuing it as if it were . . SQUIRE: It sounds good. WEBB: . . as if it were neutral, whereas in effect--of course, as you say, I mean if you actually violate that would be a grossly inegalitarian--what I'm saying (and what the Bank is saying) is what is considered neutral is in effect grossly inegalitarian. I agree. SQUIRE: Okay, but . . WEBB: But once you put it in terms of proportion it would like making explicit at the project level what people do anyway at the macro level, and it then will give you a lever to point out, “Is this what we really want?” And it would get around the subjectivity. It doesn't really get around it, but . . SQUIRE: Even what we do in practice has a subjective element because we're implicitly assigning weights, that’s right. WEBB: It’s always subjective, okay, but what it does is, because people are saying you can't do weights because its subjective, then you do the other macro projection which is [inaudible] brings to the micro level the subjectivity that people are living with [inaudible] Lyn Squire January 16, 1992 – Final Edited 13 SQUIRE: All these kinds of arguments—not this specific one, maybe--but all these kinds of arguments, this was the debate in that second half of the 1970s. WEBB: The debate continued for . . . SQUIRE: This was a very long drawn-out process, yeah. The actual guidelines which came out, and which do include income distribution weights, may well have been ‘79 or ‘80. KAPUR: Circulated for five years? SQUIRE: I think this was over a five year period. KAPUR: I'm sorry. You said that [inaudible] no, actually the distribution weights were applied. SQUIRE: That statement is correct. KAPUR: I’m sorry. Which one? SQUIRE: Okay. Let me start: two statements. One is that the distribution weights became part of the Bank guidelines. So they were in a sense formally approved by all the bodies that have to approve such guidelines, but, two, they were rarely used in practice. KAPUR: And so they're now—it seems they might be dropped? SQUIRE: They’ll be dropped. KAPUR: Even on the formal guidelines? SQUIRE: We will make the formal guidelines consistent with practice. WEBB: Can you say rarely, never? SQUIRE: No, I can't say that. There were a couple of cases, you know, pseudo- experimental cases where there was that. KAPUR: But even the--I guess, this is not directly relevant [inaudible]--even the Little- Mirrlees, you know, shadow prices, how was that--was that actually also--I mean, there’s a big feeling that that was also kind of [inaudible] SQUIRE: I think that this is an important point because while there were some people who felt uncomfortable even with that, in general that was much more acceptable, and I think for a while we had a fairly serious effort of trying to figure out what shadow prices would be in various countries and making sure they were implemented at the project level. I don’t know if you have talked to Anandarup Ray; he was sort of the key guardian Lyn Squire January 16, 1992 – Final Edited 14 of it at that point, took a fairly serious stance on this. So there may well have been quite a bit of that done. Now, since he left the economic advisor’s position or whatever it was called, I think we’ve seen a definite deterioration in the quality of our economic analysis of projects. That's not just the use of shadow prices; it’s also things like being conceptually clear about what constitutes a benefit or what constitutes the cost. I think this was the thrust of the Little and Mirrlees critique a couple of years ago. KAPUR: But was it inevitable, something which is almost inevitable given the institution that, the top management, is more concerned about macro issues than at the colleague level? People’s attention shifted almost? SQUIRE: I mean, at the height of adjustment lending there was, what, 25 percent, 30 percent . . . KAPUR: Yes, but again the numbers don't capture the energies [inaudible] concerned. Ernie [Ernest] Stern [inaudible] going through his chron files. SQUIRE: That was a different--well, I think the reason is somewhat different. I think it had to do with the change in people who were sort of the guardians of the Bank's economic analysis of projects. When Anandarup left the system changed, the degree of central control and the degree to which the center was a watchdog on these issues became much less. KAPUR: That had been when, the change? SQUIRE: In the early ‘80s, I guess. KAPUR: That would have been the retirement of Warren Baum also. SQUIRE: It may well have been. WEBB: Do you have annual reports or other material on the issue on distribution weights over this long period? You must have been going to meetings all the time. Who were some of the key people who were involved in it? SQUIRE: Well, the chief economists at the time were all involved this. WEBB: The regional chief economists. SQUIRE: Right. Some of them. WEBB: Guy [Pfeffermann] was one? [inaudible] SQUIRE: [Dragoslav]Avramovic was, Jean Baneth, Stanley Please. I mean those are the kind of characters that I remember. Lyn Squire January 16, 1992 – Final Edited 15 WEBB: And what other staff? SQUIRE: And then there was—I mean, you know, the real project guys. WEBB: Baum all the time? SQUIRE: No. No, Warren left it all to van der Tak. I mean I’m sure that he, van der Tak, kept him up to date and so on, but Warren was not--he didn’t go to those kinds of meetings. They were meetings of a slightly lower level of staff. WEBB: Was van der Tak always, in all this time, a real missionary with respect to this? SQUIRE: Yes. He was out there trying to get people to buy into this, into the approach. WEBB: As a group, how were the economists, the chief economists? SQUIRE: They sort of--I think they were expecting very much Operations to give us some difficulties. WEBB: So there was a core of [inaudible] SQUIRE: Well, basically we had support from the two central staffs. I forget exactly how it was organized, but there was Hollis's [Chenery] whole vice presidency, right, which was the—you know, we called it economic . . KAPUR: Policy planning. SQUIRE. Yeah . . [End Tape 1, Side A] [Begin Tape 1, Side B] SQUIRE: . . policy or economic analysis, and then all the middle-level managers who saw this as a serious cost and imposition. WEBB: Was Tony [Anthony A.] Churchill a part of that? SQUIRE: Tony doesn’t jump into my mind as being a key figure in that. David Turnham is more important. WEBB: He was in agriculture. SQUIRE: Yeah, agriculture. WEBB: Is he still in the Bank? Lyn Squire January 16, 1992 – Final Edited 16 SQUIRE: No. KAPUR: He’s [inaudible] I'm wondering, when you moved to operations, did your views change? Did you have greater empathy with the chief economist view at that time, or . . . SQUIRE: When I immediately moved—when I first moved to operations we did try to go through the process of having country economists do the shadow prices, have that passed through to the project economists. There was consistency on that front, but we never persuaded the projects staff to seriously estimate the impact of their project on beneficiary incomes. So that part was never done. WEBB: What region was this? SQUIRE: This was on Indonesia. Now, I forget exactly what I did. I guess I went to—back to work with Ben [Benjamin B.] King in the department of economics. King was the head at the time. When I returned to Operations, first to West Africa, I thought that there was a commitment to the pure economic analysis of projects but a deterioration of interest in distribution weights; they just never got off the ground. KAPUR: This was around ‘84? SQUIRE: Probably about ’84. And in the department, the West Africa department, I instituted a process where we had the economists in the department review every project that came through, both to see what sort of shadow prices were being used (and it was very serious). We had guidelines, country guidelines for what was a minimum effective set of shadow prices and to just check whether people were correctly identifying those. And it was difficult to find exactly what underlies our expectation of the issue. They’re no longer as clearly presented in the reports, to the Board, for example, as there used to be. So you really have to go back and talk to the guy and get into his files if you’re going to figure out, you know, exactly what goes on. So I think the process is very much less transparent and much less focus of attention given. I mean, if you go back to what I was saying earlier about Herman, when Warren had his meeting and Herman would be sitting there, right, he would review the economic analysis. He was questioning the guys there about that analysis--I mean, we really don’t have that kind of project control now. WEBB: Doesn’t it occur at the regional level? SQUIRE: In principle it could. In practice it rarely does. KAPUR: I was wondering whether Ben King, whether he had been in sort of doing the project analysis in the ‘50s--was he ever an activist? Lyn Squire January 16, 1992 – Final Edited 17 SQUIRE: Again, I don’t believe Ben King was--I don’t know. I'm not going to say he wasn't. KAPUR: Right, but it doesn’t strike you. WEBB: The big problem at the regional level was that the guy who would be doing this, his boss would be [inaudible] SQUIRE: The incentives are all wrong. WEBB: Yeah. SQUIRE: That’s the difficulty of someone who has my job, for example, the lead economist's job, because there you’re really trying to make sure that the economic analysis is done well, but in a sense you’re fighting against the way the incentives are set up, which is to move a project through expeditiously. KAPUR: Jumping ahead again, I wondered if in your mind as you were working on the 1990 WDR—I guess you must have looked at the ‘80 WDR, between that time, between that one and this, what did you see as the difference in the evolution, both intellectually as well as sort of operational? Was there change? SQUIRE: On the distribution weights? KAPUR: No, this is on the WDR. Just on the whole approach. SQUIRE: I think the one major difference was that we had some real information to work with. The 1980 WDR didn’t. In 1990 we did not have a lot of information, but--I mean, the contribution, if that WDR made any contribution, the contribution was that we seriously tried to explore the database for making statements about the factors involved in distribution and so on and so forth. The 1980 WDR made similar kinds of noises, but it had really very primitive information because that was all that was available then. In terms--so I think that's the major difference. In terms of policy guidance or policy analysis, I think the two WDRs were really pretty similar in terms of their recommendations, but in Paul’s [Isenman], the 1980 one--and this was the particular direction which he gave it anyway—was implicit in the title Poverty and Human Resources. There's a lot of emphasis on that, somewhat less on the process of economic growth and the policies that might be governing economic growth. I'm not saying it was not there because it certainly was there, but in terms of emphasis in the 1990 one it was partly because of the work which we had to do with datasets we developed. The balance was slightly different because the way we saw it—you know, it’s basically all sort of labor market issues. You've got to have a growth process that creates the jobs. You've got to have people who have the skills, education, and the health to be able to respond to those job opportunities. So that’s how we saw it--giving much more Lyn Squire January 16, 1992 – Final Edited 18 equal emphasis on both sides of the issue, whereas Paul was much more on human resource, emphasized that side more. So I hope that’s clear. KAPUR: Yes. WEBB: Did you change the weights aspect? SQUIRE: Yeah. KAPUR: I was wondering, I mean--well, perhaps it’s, I have my own biases, but--on this issue of population, which is linked, I think one can say, very inextricably with poverty, [Robert S.] McNamara, you know, when he was in charge, you can see that urban growth policies correspond to McNamara’s [inaudible] population growth. You know McNamara’s constant [inaudible] And one sees that [inaudible] And yet when you actually see in the Bank that this is this done, it’s all very minimal, given perhaps from my point of view it’s [inaudible] And even now, especially in certain quarters of the Bank, either, you know, once you get the economic incentives right this will work out, perhaps with a lag difference [inaudible] Do you think--I wonder how you see that issue and how you see the thinking, especially [inaudible] SQUIRE: Well, first of all, I don't think there is, that we, that anyone’s established anything whatsoever between population growth and GDP productive growth. If you look at any of the simple correlations or more sophisticated regressions that’ve been done there has never been clear evidence of that one. Nor is there evidence, again, just looking at the numbers of the link between rates of population growth and fertility and efforts to reduce poverty. It doesn't mean to say we shouldn't worry about population growth, but the alarmist kind of view that it is really a key factor driving GDP growth and the rise in poverty I don't think has been established. I think what there is, is an increasing awareness that we really ought to be in the business of providing a package of basic services, education especially, female basic health care (we’re doing a lot of this) and family planning services. With that package, whatever you call that--and whether you're interested in doing that because it's important to population or whether it's a broader issue because it contributes to health and well-being, or even if you argue that this package actually is productivity increases, I don't know the difference. I'm less interested in what the particular rationale is, but there seems to be some broad agreement that that package is worth pursuing. I think there is also recognition that family planning programs are effective. If you have decided that you want to reduce population growth, there is reasonable consensus that there are ways of doing it. It's going to be less effective than going to the countries and so on and so forth, because it does have to be, not just (although it’s an important part) of the general development of the country, but of the country’s own plans on population. Lyn Squire January 16, 1992 – Final Edited 19 Of course, the vice president of research right now questions all this, the link between population and development. WEBB: We heard he has a view rather like yours, doesn’t he? SQUIRE: The first part he does. I'm not sure where he comes out in the end. WEBB: Ah. But now he's more skeptical, is he, about the importance of pushing family planning? SQUIRE: That's the sense I get, yes. KAPUR: The same thing that those old [inaudible] I had the sense that . . SQUIRE: Absolutely. I think that’s right. He's asking hard questions, and he's certainly criticizing sloppy thinking. You know, it’s the kind of thing which says, “Well, we're interested in the environment; therefore, at all costs, we must do things to protect the environment.” Now, no one in the Bank has that kind of view anyway, but he is questioning the values and people’s attachment to environmental goals. WEBB: As we begin to pick up—we hear some of the rumblings already about [inaudible] your draft. SQUIRE: Mm-hmm. WEBB: I want to come to something else in the poverty story, which you a little bit linked to your very first comments, because I understand that out of your work, the WDR work, the most recent work, that a proposal was formulated that the Bank not only sort of make country poverty strategy evaluation central but also apply some conditionality or attach some conditionalities, maybe get into SAL [structural adjustment loans]-type lending. Is that right or not? SQUIRE: No. It’s almost right, except for that last part that you said. The idea is slightly different. We asked the question in the WDR: “What is the value of aid in reducing poverty?” And we go through the same things that we were talking about earlier. Aid is only a small part of what any country does. It's very difficult to actually identify any link between aid and the reduction of poverty. But what we were saying was that if a country is using its resources reasonably well and making adequate provision for human resources, all the things that you think of as being good both for general development and for reduction of poverty, if a country is doing that, then there's a good prospect that a dollar of aid will also be used well. If you have a country at the other extreme which is spending its money in all kinds of bad ways, with a big steel mill, on military expenditures or whatever, chances of a dollar of aid being used effectively are also reduced. So we were saying that--in the WDR, which is addressed to the world at large, not just to the Bank--we were saying that in thinking about aid allocations you should be judging what the country itself is doing, making decisions on that basis about the volume Lyn Squire January 16, 1992 – Final Edited 20 of the assistance to different countries. And then, in the policy paper, we got to talking about allocations within the Bank. So the point was not on SAL conditions, conditionality of poverty in the context of an operation, but rather saying that when we allocate either to Country X or Country Y, we will be thinking about what that country is doing in terms of its overall approach to reducing poverty. If we think the country is doing an effective job—these judgments are difficult, obviously--that would be an argument for increasing the allocation and vice versa. WEBB: Is this proposal written down anywhere? SQUIRE: It’s in there. WEBB: Part of this is . . . SQUIRE: We . . WEBB: This was made to the Board? As a proposal by whom? SQUIRE: Well, I presented it. WEBB: You personally presented it? SQUIRE: Yeah, but it had gone through all the review process. It's not so revolutionary because obviously we have a system--we've always had a system of allocating IDA [International Development Association], so otherwise how do you get to the outcome. And poverty was always, somehow or other, thought to be a dimension in that allocation. KAPUR: But in a sense you already are looking at a joint sort of [inaudible] with IDA allocative processes. Given that it's the fact the smaller countries, smaller populations, get more per capita here than [inaudible] process, which is independent of their poverty, and the fact that the donors already say X percent should go to Africa, you have two constraints already. And it's within that that you are making—it’s already a constraint on allocations. SQUIRE: We have a sort of system, or norms, for what a country should receive based on population, per capita income. I forget the exact amounts per annum. And then there’s scope for discretion around that. So the extent to which that discretion is actually used is a very good question, and on the big countries we’re not so fixed on the [inaudible] prospect of change. But that's the way the recommendation went to the Board. That's the way the Board discussed it and endorsed it. Excuse me: When I say that's the way, but it was not with these certain amount for China, certain amounts for India. KAPUR: Which is an [inaudible] In a conversation with Ravi Kanbur on poverty issues he sort of was alluding to the fact that the Bank, given the [inaudible] measurement of those poverty, and given that there’s Lyn Squire January 16, 1992 – Final Edited 21 really an underlying tension in the allocations, and that process has implicitly led to sort of underemphasizing absolute poverty because, you know, if you use that, then a country will just—they won’t report resources and so forth very near the poverty line, it can show it’s done very well, because it takes more effort to get those who are absent [inaudible] Do you think that’s a reasonable [inaudible] SQUIRE: I don't know. I know that Ravi does. Where do we get the information to test that proposition? I mean, there’s some a priori reason you might think that's what governments might do, but I'm not sure governments really work from a headcount to determine exactly what policy they’re going to pursue . . KAPUR: But the Bank certainly does in its [inaudible] Whenever you see numbers, you know, poverty-line, numbers below the poverty line, everything is based on that foundation. SQUIRE: Yeah, but that's only the starting point. And I don't think that when we implement projects, you know, we can't tell whether it's going to people below the poverty line or not. We just don't know. And I would not have thought that there was a, sort of by design, an effort to target those guys who were just below the poverty line in order to make a project look attractive. [both speaking at once] KAPUR: Right, no, it's just sort of implicit rather than . . SQUIRE: Sure. That may well happen. The one piece of evidence that we had is--for the countries where we had information on income distribution over time, it was that—well, two points. One is we looked at people under whatever our definition was of the poverty line, did the simple headcount stuff . . KAPUR: Right. SQUIRE: . . and drew whatever conclusions that showed. But then we looked specifically at the bottom ten percent of the population to see to what extent they’d been participating in whatever growth there might have been. And we don't have a very large sample in the countries that we looked at, with one important exception. They, in fact, participated in percentage terms better than the average. KAPUR: Oh, I see. Which country was that? SQUIRE: I think it was Brazil. WEBB: These country poverty evaluations, they’re not meant to be the basis for [inaudible] the Bank has done some. There’s a reference here, a review of several such evaluations, the lessons. SQUIRE: Right. Lyn Squire January 16, 1992 – Final Edited 22 WEBB: Those tend to be--were they pretty good? Fair? SQUIRE: Yeah. WEBB: Do they come up to what you think is needed? SQUIRE: There were some, I thought, some exceptionally good ones. Let me mention two. One--the real issue here is data availability, okay? So one example, India, the India report I thought was very good and we have more data for India than virtually any other country. The other example is Malawi, where there’s very limited information. Barely enough to base a judgment on poverty assessment and take it to the point where a realistic judgment can be made on strategy. WEBB: [inaudible question] SQUIRE: No, these are Bank reports, the gray-covered; they go through the gray- covered process. I may have those. KAPUR: We can find them. WEBB: Don't worry because we have the reference. SQUIRE: This is Malawi, growth of poverty, March 1990. I need to keep this, but . . KAPUR: What's the ID number? WEBB: 8140? SQUIRE: 8140 dash M-A-I. And this is the India one. There are so many of the others, but this one you'll pick up all the information from interviews. Here, I thought, was some really good work on Malawi. WEBB: That was Kathie Krumm, wasn't it? SQUIRE: You know Kathie, don’t you? [inaudible] WEBB: Is she in the Bank anymore? SQUIRE: No. WEBB: That's a model for what could be done. SQUIRE: If people asked me what should be done, I'd say, “Take a look at those two reports.” WEBB: I told Devesh I did one of the first such reports; that’s how I got into all this. Lyn Squire January 16, 1992 – Final Edited 23 SQUIRE: Right. WEBB: Guatemala. It was just a one/one, one-man mission, but what was distinctive about it was this attempt to look at the whole picture. SQUIRE: Right. WEBB: A kind of a new and very judgmental, of course. [inaudible] Well, I think we've used up . . . SQUIRE: Okay, sorry, there’s one other--there's one question you didn't ask me which I think you should have, and that is: Having done the poverty WDR, why is it that in there we didn't talk about the income distribution weights, you know, taking it through to the project level, and how my thinking changed about this. WEBB: That’s a good question. SQUIRE: And the answer is that, one, I buy the argument that it's extraordinarily difficult at the project level to identify beneficiaries in a systematic fashion. So I bought into that operational argument. The second point is that I am much more persuaded now-- and this also has been the tenor of our discussions--about the importance of all kinds of policies, so the totality of what the government is doing is the real determinant of what's happening to the poor. So I'm not quite so hung up on the idea that we have to have the poverty weights, the income distribution weights, in our project analysis. But having said that, I still believe strongly that we should be looking across this whole package and making sure that the right things are being done for the poor. So I think I would back off where I was, where Herman and I were, in ’75. WEBB: That would really back off a long way, and not just from weights but from direct intervention in general with that argument. I mean, why are you—why aren’t there still hydropower projects if the country judges [inaudible] poverty. Why should the Bank have to replicate with a microscope what the country already did? SQUIRE: I think there is an argument about the comparative advantage of the Bank in certain areas, that’s right. WEBB: In addition to that, isn't there a sort of leverage argument? SQUIRE: There is. I think there is a comparative advantage argument at the Bank. We are being pushed by the NGOs [non-governmental organizations] to get lower and lower. I'm not sure that we can do that effectively or should be doing that effectively. On the other hand, I believe very strongly that we ought to be very much in the social sectors: helping with basic education, health and family planning, not maybe because we are necessarily the best people to implement that, but because I think the countries do need a Lyn Squire January 16, 1992 – Final Edited 24 bit of pushing in that direction, and if our being there helps move the expansion program a little more in the direction of providing those basic services as widespread as possible, I think that’s really a worthwhile kind of effort. WEBB: Are you saying there is not 100 percent fungibility, even in a country like India? SQUIRE: Yeah, because the point is that we're not talking just about our money, but usually what the Bank is doing is influencing the expansions in the sector or expansions in general, that's why . . . WEBB: Externally? SQUIRE: No. WEBB: Oh. The whole sectoral. SQUIRE: Budget. Yeah. That's where I think there is . . . WEBB: But influencing just by example, by persuasion? SQUIRE: It’s, you know, it’s persuasion. WEBB: Strong arm-twisting? SQUIRE: There may be some arm-twisting, arm-twisting persuasion, sure. KAPUR: I [inaudible] SQUIRE: Right. But that's--I think what we're doing increasingly with these time slice projects is a really, really, good idea. KAPUR: Refinance a fraction of public expenditure in . . SQUIRE: In social sectors, education, whatever it happens to be. But we have the opportunity to say something about all the expenditures in that sector. WEBB: You’ve got to be at the table for that sector to be on the page. SQUIRE: Right. Now, you think of the situation as it normally is, where we may be doing a single project in the transport sector. We could be doing that project, and the rest of the project expansions in the transport sector could be totally screwed. WEBB: But even those cases in the social sector? SQUIRE: Absolutely. That's why I think the time slice approach has a certain amount of attraction. We're not financing a particular project. What we're doing is we're looking at Lyn Squire January 16, 1992 – Final Edited 25 the government's entire expenditure program and, although obviously we can't go into every part of it in great detail, if we’re comfortable that it's going in the direction of basic health care or whatever we think is good, then we put up our money as part of the finance. KAPUR: [inaudible] sector investment [inaudible] SQUIRE: Yeah. WEBB: But if a country is doing relatively well in its general approach, you think that's generally not the case, generally there is room for improvement in each sector, and the Bank has the capacity to influence or should have? SQUIRE: Let me—I don’t know how we got so far off the subject, but it is an interesting issue. We are coming out of the adjustment phase--well, let me take Mexico because I know Mexico. We're not doing any more adjustment operations in Mexico. The balance of payments looks really good. But because of the process, there’s an urgent need in several sectors for a lot of public spending. It's also the case in Mexico that the key architects of the reform were really the central ministries and that's why the dialogue with the Bank was very good on adjustment operations. In the line ministries we still have a lot of the old, more traditional civil servants, and I think there's an urgent--not urgent, but there's a definite need for a dialogue with the line agencies. Time slice, sector adjustment operations gives us the opportunity to do that. WEBB: Time slice is what, a longer, long-term? SQUIRE: Could be two or three years. So there would be an annual review in the context of the budget and then probably an annual release of--or Bank funding would be released annually. WEBB: And basically it’s a SECAL [sector adjustment loan]? KAPUR: No, it's called a sector investment. WEBB: It actually has its own name? SQUIRE: If I do a SECAL, which is just an adjustment operation in a particular sector, right, I have to have a balance of payments rationale. WEBB: Oh, I see. SQUIRE: There has to be a balance of payments gap measured in some way that the Bank is coming in to fill. With a time slice or with any other investment project, we don't need that rationale. WEBB: [inaudible] Lyn Squire January 16, 1992 – Final Edited 26 SQUIRE: No, well, you're writing history! You don't have to worry—that’s right. KAPUR: Well, actually, I was thinking that irrigation in a particular state, do that time slice anywhere, in the state of Maharashta . . WEBB: So these are sectors? Is that [all speaking at once] KAPUR: They're called a sector investment, and that's how the database I showed to you, the one that has instruments of Bank lending, has SECALS and direct investments. SQUIRE: But the characteristic of it is that we are financing a time slice of the government's expansion program in a particular sector. WEBB: So, in other words, in the end the Bank really does need to do a lot of direct cultural investing . . SQUIRE: I think in some sectors we need to do it. WEBB: Because there is potential leverage there. SQUIRE: I think there's leverage, and it's difficult. WEBB: Difficult technically? SQUIRE: Well, yeah. I mean, the rural population which is just spread all over the place. How do you attract poor people to finance? How do you get the girls in Islamic countries to go to school? WEBB: So there's a technology the Bank attracts [inaudible] SQUIRE: In principle. WEBB: That's what you're saying. SQUIRE: That's right. Absolutely. And we're learning, as well, I must say. You know, maybe we’ll take an experience from one country and use it in another. The usual kinds of arguments for the Bank's involvement. WEBB: Well, Bangladesh is—we were there--I mean, they're the world’s experts on rural development. KAPUR: Rural credit, actually. SQUIRE: Yeah, but, you see, I didn't mention rural credit as being something that we want to necessarily be involved with in targeted intervention with poor people because Lyn Squire January 16, 1992 – Final Edited 27 I'm not sure that we've had much success with it. There may be ways of doing it. I feel more comfortable on the social sector interventions. Well, that’s a personal view. WEBB: You've been really, really helpful. Thank you very much. There will be more questions. [Interruption] SQUIRE: There’s a sense if you talk to people in the Bank now, they say, “Well, the Bank doesn't do work on labor markets.” You know, I think back, we actually had a division working on that. And I say, “I get a little bit upset when you say that.” I don't know exactly—I don’t know all the history myself, but certainly in this region we are now looking for people to do labor market type analysis. And this is something which is one of Larry [Lawrence H.] Summers’ personal interests as well, and there may be some rekindling of work in this area. But just as a phenomenon it’s interesting to see. There was this period when there was all this attention to labor markets but then, you know, it died. KAPUR: In the ‘70s, well, the ILO [International Labor Organization], and then in the ‘80s with employment. Then I guess when you had--the Bank began to, when sort of the adjustment programs which led to public sector retrenchment and compensating net and so on, and employment and labor markets began to be in touch with that, but still enough for that [inaudible] coming back in the ‘70s to income. SQUIRE: Okay, gentlemen. Well, I wish you luck. [End Tape 1, Side B] [End of interview] Lyn Squire January 16, 1992 – Final Edited