Intro Sandra Knispel: The World Bank’s goal is to end poverty. It was founded in 1944 at the Bretton Woods Conference to rebuild Europe and other parts of the world devastated by World War II. The Bank is independent and gives loans to countries solely based on need and merit of the project. But that’s theory say political scientist. In reality, they argue, some powerful multinational corporations are pulling the strings.
Joining us today is Randy Stone, a professor of political science at the University of Rochester, who is one of the two political scientists who found undue corporate influence at the World Bank. Welcome to the Quadcast, Prof. Stone.
Randy Stone: Thank you, Sandra.
SK: So, before we even get to your charge let’s backtrack for a moment. Can you talk a little bit about what the World Bank is supposed to do? How it is supposed to work in theory?
RS: Well, the World Bank is a development organization, with about 170 members, which engages in project-based lending. So, it identifies a wide range of different projects; it might be about water safety, or health, or education, or they may build a large dam. And this project is intended to produce economic development and improve quality of human life. It has a rather rigorous evaluative process to go over these projects and make sure they achieve their objectives and that’s where it sometimes runs into a little trouble.
SK: So, theoretically, a really noble idea.
RS: Oh absolutely. It’s repurposed itself a couple of times. Of course, it was originally intended to eradicate the devastation of World War II, to rebuild. But, it has subsequently repurposed itself as an all-purpose development organization.
SK: You and Rabia Malik, who got her PhD in political science at Rochester and is now a post doc at New York University Abu Dhabi, wrote the research paper together that appeared in the Journal of Politics. It’s titled “Corporate Influence in World Bank Lending” —how do multinationals actually get involved with World Bank projects?
RS: Well, there are two major ways. Projects are implemented by organizations, usually private firms, and about three-quarters of those firms are located in the country in which the project is taking place, and the other quarter or so are multinational firms who operate as contractors for those projects. But, firms are also interested in projects they’re not directly involved in. If they are downstream in some sense, for example, an aluminum smelter, which is planning to use the electricity that’s produced by a dam, would be interested in the completion of the dam, even though it’s not directly involved in the project.
SK: The Bank has a way to ensure the quality of the projects it funds. But that control, which entails withholding loan payout, is not necessarily in the interest of corporations—can you explain why?
RS: That’s correct. Now, I should emphasize that the World Bank is one of the most transparent international organizations. There are many other development banks that do not publish nearly as much as the World Bank does. As a result, we were able to do this project on the World Bank. They do detailed evaluations of every project. These project reports, ICR reports, they’re called, range from 20 to 200 pages in length, and then the Independent Evaluation Group goes back and audits all of these projects again, and in some cases, downgrades the projects. So, there are checks and balances. It’s quite a detailed process they go through to make sure, first of all, that they’ve accomplished what they’ve set out to do with a particular project, but secondly that they haven’t violated any one of the large number of mandates the World Bank has taken on over the years; for example, not to displace indigenous peoples, not to cause environmental damage, not to violate human rights, and so on and so forth. All these things are involved in the evaluation process. The difficultly comes in when a multinational corporation, which is interested in the completion of the project so that it will get paid, because it’s a contractor, or because it wants the electricity from the dam—[but] the firm is not interested in the World Bank’s mandates. The environmental sustainability, for example, doesn’t enter into the profit and loss calculus of the firm. And so, the firm may be interested in just rushing something into completion while the Bank staff is concerned about the collateral damage that may be done.
SK: Contemporary models of foreign direct investment rely on large firms and their technical know-how for such large foreign projects. But you write that you find “no evidence that multinational corporations’ involvement as contractors improves the performance of Bank projects.” What did you find instead?
RS: This is an extraordinary finding, I think. Contemporary economics tells us that the firms that engage in foreign direct investment are the most highly productive firms with the best technology. And you would think that engaging these firms—bringing their know-how, their technology and their organizational acumen into the picture—would help to lead to better outcomes in projects, and it’s striking that we find that that’s not the case. It does lead to better evaluations; it does not lead to better performance. And so, the key here was that we took these 4,200-some projects with reports that might be a couple of hundred pages long, we combed through them with a team of research assistants and coded point by point: what were the objectives of the project, and to what extent does the World Bank staff assess the objective as having been met? And we create an index, and we compare that index to the headline grade that the project receives. So, the project gets a B+, but according to our assessment, they only completed 66 percent of the project—so there’s a gap there. So then, we wanted to see what is the effect of having a multinational corporation involved in implementing the project on the assessment gap between the evaluation and the underlying performance, and we found a strong correlation. Particularly, when the multinational firm was involved in a particularly important management role.
SK: So, why is it that you, with your team of researchers you would find that assessment gap and nobody at the World Bank seemed to notice, or do they not care?
RS: I’ve talked to a number of people at the World Bank about this and many of them are surprised by our findings. Some of them are not surprised by parts of our findings and are surprised by other parts. I actually had one interview with a World Bank official, where he was insisting that, “No, no this doesn’t happen, there’s no revolving door at the World Bank. You know, you sometimes hear, these revolving doors where these lobbyists end up working for the government, and government people end up being hired by the firms that lobbied and so forth, no, no, no, that doesn’t happen in the World Bank, it’s too attractive working at the World Bank.” And then he thought for a minute and then he said, “Hmm, although I have seen examples of outside contractors, not staff of the World Bank, but outside contractors who did that,” and then he thought and said, “Yeah, I can think of several examples.” And then he started backtracking. The thing about the World Bank is that it is a very complex organization: 20,000 employees, nobody has an overview of all aspects of their operation, so it really takes someone who comes from the outside to get that bird’s eye view of what’s happening.
SK: So, when you come in now as the researcher, I would assume that it’s shocking to at least some of the managers to hear that. How are you being received?
RS: Well, it varies. I mean there’s always a little bit of skepticism, I mean most World Bank employees—professional staff—are economists, and they can think of lots of econometric reasons [why] our results might be faulty. And so, they start asking: “Well did you try controlling for fixed effects for countries?” and I said, “Well, yes, we did actually.” “And did you control fixed effects for time?” “Well, yes we do have time fixed effects.” “Well, did you try this? Did you try that?” And after a while they start scratching their heads and they think, “Well, maybe this is a real finding, and now we have to explain it.” We did find some people in the Independent Evaluation Group in the World Bank—kind of a watch dog auditing agency, semi-independent, really, not fully independent—but they were able to fill in a lot of the gaps for us about the incentive structures. Why is it that people within the Bank might have an incentive to look the other way if a project really wasn’t performing very well? But they, you know, want to get it funded. Bank officials really have an incentive to have their projects go to completion and be successful. Like Garrison Keillor would put it—all children are above average, right? So, they have this incentive to push loans and get things to be successful, and if the project gets a bad evaluation, it reflects badly on them. They have a very rigorous internal assessment process for promotions and raises and so forth, and so the quantitative score that you get is important to your career. And so, there have been cases that they can talk about where someone has come in and disputed their assessment when they claim that, well, this ICR report wasn’t done well. The project really wasn’t completed, or wasn’t successful, or had some sort of negative consequences. The Bank’s staff often push back very hard, so that helped us to understand that if they had an excuse to do so, World Bank staff would have an incentive to look the other way. Then the question is, well, why do they have an excuse to do that? And we think that’s where the lobbying efforts by the multinationals come in.
SK: So, to be clear—the incentives actually make the Bank perform potentially worse because there’s always the incentive to make things look better.
RS: It at least relaxes the incentives to enforce the conditionality. So, the idea is the World Bank lends money in return for getting work done. And one of the conditions that the developing country that’s receiving the loan has, it’s supposed to put in some of their own money, perhaps, or they are supposed to facilitate the project in certain ways. And if they don’t do that then the disbursement of the loan can be held up. The Bank has an incentive to build a bit of a reputation for actually enforcing these things so it can get some cooperation from the developing countries.
SK: Let’s go back to these string pullers. Who are these large multinationals who exert this undue influence? Because you found that they’re more likely to come from certain countries than from others…
RS: That was one of our striking findings. You might think, well, multinational corporations are multinational corporations and where these multinational corporations are involved those might be projects that are just different, or countries that are just different, or something. And so, we thought, well, we should look to see whether multinationals from different countries are treated differently. And in fact, we found that American multinational corporations and Japanese multinational corporations have these very strong effects on disbursement and on evaluations, and that that’s not true for German, or French, or British multinationals. We thought, well, that could be because there are just more multinational corporations from the United States and Japan that are involved, but actually, there are more French multinationals that act as contractors in our data than either American or Japanese ones. So, we think that the reason for this is that the United States and Japan are the two largest shareholders in the World Bank—they’re the two most powerful countries, the United States quite a bit more powerful than Japan in this respect. [It] is the country that initiated the creation of the Bretton Woods institutions, the international monetary fund in the World Bank, but in the 1980’s—at a time when Japan was growing very rapidly—their government set a goal of increasing their representation in the international financial institutions. And the United States was willing to give them a bigger share of the World Bank in return for not giving them a bigger share of the IMF, and so there was kind of a deal that was struck there. So, Japan ended up being quite influential in the World Bank. You can see that in the direction of policy advice that the World Bank adopted in the 1990’s, and you can see that certainly in their voting shares. So there’s reason to think that if two countries are going to stand out it would be the United States and Japan.
SK: So, basically your data is showing very clearly (or what you found in your data analysis) that the two countries who have the most say in the World Bank also are the two countries where the large companies directly benefit from that?
RS: Right. We can’t measure the size of the benefits the firms are getting, but we can see that there are very substantial differences in disbursement rates in these projects. So, for a multinational firm from the United States that has management responsibilities—that’s kind of the best case for exerting influence—that causes about a 10 percent increase in the disbursement ratio, that is, the ratio of money disbursed to money committed in a project.
SK: Let me be a devil’s advocate here. Why does it actually matter if multinationals pull the strings? Isn’t that just sort of a way of letting the market regulate itself?
RS: Well, the thing is that the firms have different incentives than the Bank. They have different interests. They want to make a profit. They’re not interested in the environmental impact. They’re not interested in the human rights impact. There was one particularly egregious case that we talk about in the paper, which was a dam that was under construction on the Paraná River between Argentina and Paraguay, and this was a dam which was a very expensive project. Even President Menem of Argentina, who was not known as a paragon of clean government, referred to it as a monument to corruption. So, at the height of the case it came under scrutiny by the Inter-American Court of Human Rights because it was displacing indigenous people, which was one of the things the World Bank is supposed to avoid. Doing so the bank insisted that a program be incorporated to compensate the indigenous people whose land was being flooded. And the managers of the project bought up the land from the indigenous people so that they could collect the compensation payments. So, it was a thoroughly corrupt deal, but the American firm that was operating as the general contractor on the project got fully paid. It was a company called MWH, which is based in Denver.
SK Now, the Bank has certain tools in its tool kit to—at least theoretically—make sure that the projects are of high quality. And so, suspending disbursements of loans is a very important and powerful tool in that tool kit and helps also the bank to maintain its reputation. Of course, that could mean risking the success of a particular project. Can you explain a little bit this conundrum and how the bank sort of tries to balance the two sides?
RS: That’s right. Well, in any particular case if you suspend the funding, chances are the project won’t be completed, or it won’t be completed well. And so, the Bank realizes that it’s shooting itself in the foot every time it pulls the trigger. So, it’s reluctant to do that, and almost all projects are completed eventually. They get delayed, but the average disbursement rate is almost 90 percent. Very often projects are delayed and delayed, and there are negotiations, and then eventually the money is disbursed. Eventually things are done so it’s costly to actually employ the sanction of cancelling a project for non-performance. On the other hand, if you never do it then you’re not credible, right? Then you can’t get cooperation, and you can’t accomplish much of anything, so the Bank has to be able to employ that sanction when it’s appropriate. So, they’ve created an incentive. It actually counts against the evaluation of the project if it is more fully funded than is justified by the accomplishment of the objectives. That counts against the project. At least, [as a staff member,] if you’ve got a bad project, you can cut your losses by not fully disbursing it. Now, if there’s pressure to disburse the loan there’s a multinational corporation, which presumably has lobbied its congressman, and the congressman has presumably contacted—or probably a congressman’s aide has contacted—somebody in the US Treasury, which oversees the US Executive director, who sits on the board of the World Bank. Somebody in the US ED [executive director]’s office contacts someone in the appropriate department and says, “Well, this is a project we’d kind of like to see go through.” No one along the way knows how important it was to the multinational, how important it was to the congressman, how important it was to the US Treasury, but it sounds kind of ominous. You don’t want to get in trouble with a congressman, or the US Treasury, right? So, everybody goes along. So there’s incentive to disburse. Well, everybody in the office knows this is a project we’re supposed to disburse, so we’d better give it a pretty good evaluation, because we have to justify the disbursement rate. So, we think that’s where the bias comes in.
SK: You write that you don’t have the smoking gun, so let’s go back to your methodology. What did you look at and how did you come to this conclusion without the proverbial smoking gun? What are you basing your claims on?
RS: Right. So we don’t have is an interview with somebody in the World Bank who said, “Well, I did this because I was contacted by the US Treasury.” We don’t have an interview with somebody in US Treasury saying, “Somebody in, you know, the xyz congressman’s office contacted me and that’s why we did this corrupt deal.”
SK: You’d need a subpoena for that one.
RS: Right. That’s the kind of evidence that you’re not likely to get. I did actually make a FOIA request for correspondence between US Treasury and Congressional offices but I never heard back.
SK: We should quickly say FOIA, Freedom of Information Act Request. Which is often used by media, but anybody, any citizen can basically request this. How long has it been since you filed that?
RS: It’s been about two years, so it’s well past the stage where I should have heard back something, but I never heard anything. So, the kind of evidence that we do have is quantitative evidence. We have statistical correlations. We have a large data set. We have 4,200 of these projects and so we’ve got 4,200 observations in our analysis, and they cover a very large number of countries over the years 1994 till 2013. And what we can do to try to improve our inferences based on this data is a bunch of robustness checks. So, as I mentioned, we can control for the identity of the countries receiving the loans to see whether that’s driving it. Maybe certain countries are just harder to do projects in, so we’ve controlled for that. We’ve controlled for time fixed effects, so we’ve controlled for the fact that there may have been change over time in the World Bank’s procedures, or the kinds of projects it was engaging in. We control for the type of the project. We’ve got information about sectors for projects.
SK: Because some sectors are just harder?
RS: They might be. Some sectors might be harder. We didn’t find systematic patterns, actually. It seems like the evaluation process accommodates the differences in the projects sufficiently that we didn’t see any strong effects, because the objectives of each project are tailored to that type of project. The rubric for indicating whether there’s success or not is presumably related to how difficult it is in particular projects. We also controlled for different evaluation regimes, because the World Bank has gone through four different ways of coding their evaluations in the time period that we studied, so we wanted to make sure we controlled for that. We used a placebo test. We substituted foreign direct investment flows into the country for our measures of interest to see whether it’s just foreign direct investment that is associated with this and not the activity of particular multinational corporations, and we didn’t find any similar effect with the placebo treatment. One other thing, we thought it might be the case that it’s actually countries that are favored by the United States, or US allies, who get treated better than non-US allies, and it just happens to be the case that US allies get more investment from US multinational corporations, right? That could cause a spurious correlation. Also we tried that and we found that, we didn’t find any evidence that US allies were treated any differently than non-US allies. It was really due to the multinational corporations, so that allows us to discredit a number of alternative interpretations of our findings. That doesn’t mean that we’ve proven that our findings actually have the interpretation that we claim, but many other interpretations that we could think of are not consistent with the evidence.
SK: How did you get the idea to look at the role of multinationals in the first place? Did somebody tell you hey, there’s something untoward going on? Did you suspect that from data?
RS: Well, no. I’m just kind of a suspicious person. But I am working on a broader project on multinational corporations and their influence over multilateral institutions of various sorts. So, I have a paper about the influence of international banks over the IMF, and I’ve got a paper about the influence of multinationals over World Trade Organization disputes. And so, I was looking to see whether I could find something similar in the World Bank. And initially I thought that there probably would be something more to the geo-politics story, that US allies would be treated differently. I was surprised to find how strong the results were on multinational corporations and how robust they were. And then, of course, once we had that result we wanted to try everything we could to try to prove it was wrong, because, of course, reviewers are going to try to do that. And we weren’t able to, so we thought, well, this is the story, so let’s tell the story we’ve got.
SK: So, then what ought to change? Can the problem be fixed?
RS: Well, the difficulties that we find are first of all that the United States is too powerful in the international system. It has too much influence over international organizations; and secondly, that money has too much influence in American politics. You put those two things together, and it’s very difficult for international organizations to behave in an unbiased way, to try to function the way they’re written down on paper to function. It’s probably inevitable the United States will continue to exert an extraordinary degree of influence over multilateral institutions. My previous work suggests that it has very extensive informal influence in the International Monetary Fund, for example. So, I think the most likely direction to reform is to reform the influence of money and politics in the United States. That’s also a tall order because, of course, it’s the people who have the money and are already very influential in our political system who often get to write the campaign finance laws, but I think that is a key place for reformers to start to work. The World Bank, to its credit, has tried to reform, and it has instituted remarkably transparent procedures so that we were able to generate the data to use to do our analysis. We couldn’t do an analysis like that on the Asian Development Bank, or the Inter-American Development Bank, or the African Development Bank. They don’t provide the same quality of data on their websites, and that’s a positive thing [on the part of the World Bank]. There’s been a lot of pressure by non-governmental organizations over the years for transparency at the international financial institutions and that has really yielded a lot of dividends. So, sunlight is the other solution. If you can’t reform the campaign finance laws, and you can’t keep the United States in a box, because it is after all the most powerful country in the world, then revealing what’s going on is probably the best antidote to these kinds of patterns of cozy politics.
SK: So, keep shining a light on it?
RS: We keep working at it!
SK: Thank you, Professor Stone, for joining us for the Quadcast.
RS: Thank you, Sandra.
SK: That was Randy Stone. He’s professor of political science at the University of Rochester, an expert on international relations, comparative political economy, and Russian and European politics. Thank you for listening for the University of Rochester’s Quadcast. I’m your host, Sandra Knispel.