Editor’s note: Bruce Jones of New York University was the lead author of the latest World Bank World Development Report. On July 27, 2011, Humanity coeditor Nils Gilman interviewed Jones about it. Jones spoke in his capacity as a scholar: the perspectives he voices are his own, not those of the Bank.

Nils Gilman: As background, please tell us what “World Development Reports” are, and how in general they get put together.

Bruce Jones: “World Development Reports” (WDRs) are the flagship research products of the World Bank. Every year the Bank chooses a topic on which to concentrate research focus—a topic on which it feels there has been enough policy or research development that it warrants a fresh look. Then usually a pretty big team is put in place, but led by one or two senior scholars who pull together the material. Most WDRs are longer than this one, and most of them have a series of separate chapters written individually by different people. For this one, to make it more accessible for policymakers, we tried to keep it a little shorter, and we had an integrated drafting: we didn’t pass out chapters, but did the drafting ourselves.

NG: Your official role for the project was “Senior External Advisor.” What did that entail?

BJ: I was integrally involved from the beginning with the two co-leads, Sarah Cliffe and Nigel Roberts, in designing the architecture of the research. Then I was one of the three principal drafters of text, along with Sarah and Nigel. There was a large cast of characters who contributed ideas in a number of different ways. Finally, I had the additional role of brokering the interrelationship between the Bank team, on the one hand, and the United Nations, on the other, because a lot of the topics were ones on which we had to bring in U.S. expertise or bore on UN issues, so we had to spend a lot of time doing consultations and engagements with the UN, and I helped organize a lot of that.

NG: How did the Bank decide to take on the topic of conflict and violence as a developmental issue?

BJ: There were two things. One, it’s been a topic that [World Bank President Robert] Zoellick himself paid quite a lot of attention to when he got to the Bank in 2007. (In 2008, he gave a major speech at the International Institute for Strategic Studies, setting out some of the key themes.)¹ Then there was a competition of ideas inside the Bank about what the topic should be, and this won out. There are usually three or four proposals, and this one won Zoellick’s attention.

 

Violence and Development

NG: Let’s begin with some of the substantive findings. What kind of violence is the Bank concerned with in this Report? As you were designing the Report, did you consider the variety of different ways in which different societies define violence, and variations in the legitimacy of different sorts of violence?

BJ: We basically excluded both interpersonal violence at a family level and criminal violence of a disorganized type. (We weren’t interested in robberies, or murders of a limited type.) For our purposes, the violence had to be, first of all, organized and, second of all, in some sense,a challenge to the state. That was our threshold: is the violence organized in such a way that it constitutes a challenge to the state, such that it interrupts the developmental pattern of the state? Whether it is organized into a rebel group fighting the government, an organized criminal gang that is challenging the rule of law, a terrorist organization—for us those were more minor questions than the first one: “Is it organized, and of a scale where it constitutes a challenge to the legitimacy of the state?”

What we found when we actually started to do the work—and this is very much an initial finding, and it needs a lot more work (but my hunch is that more work will strongly reaffirm this)—is that the distinctions between different categories of violence are much more important to scholars than to reality. On the ground, the way in which violence is organized at any given point in time is a tactical question, a function of competing interests or what have you, but it’s not the case that there are deep and fundamental differences in how different forms of violence (and how they’re organized) affect the process of development.

That’s a pretty controversial finding. The academic literature makes some sharp distinctions between different categories of violence, but we believe that when you look at the pattern of the evolution of realities on the ground, those categories are overplayed. We’ll see now in the literature and in the field more and more emphasis on breaking down these categories and trying to understand in a more granular sense the political economy of who’s contesting what, with patterns and categories of organization becoming a secondary question.

NG: What is the basic relationship between development and violence?

BJ: The vast majority of very low-income and fragile states have experienced some sort of episode of serious violence in the last twenty years. Fragile states that haven’t recently experienced violence is a category of two or three. What that means is that violence or the threat of imminent violence is the fundamental operational reality for virtually all developmental efforts in low-income states today. Development takes place in the shadow of violence.

The basic thesis of the Report is that violence creates a fundamental and long-lasting commitment problem. Parties that have been through the experience of violence with one another have a strong rational reason to distrust one another for a long period of time into the future. And if you have a commitment problem like that, it’s very hard to organize a stable politics, it’s very hard to organize security, and that makes it really hard to organize economic investment. (Economics is slightly easier than these others, because you can risk a part of your savings, whereas you can’t risk a part of your life; but it’s still the case that the shadow of distrust, and the commitment problem that comes out of the experience of serious violence, constitute a major obstacle to patterns of cooperation.)

In an earlier phase of the Report, we wrote about this in game-theoretical terms. It’s the difference between an iterated prisoner’s dilemma and a single prisoner’s dilemma (PD). In a single-game PD, you have no rational pathway to cooperate; whereas in an iterated PD, you do—there’s a rational pathway to cooperation and institutionalization, though there is no guarantee that it will be followed.

So how does violence play in? If you believe that the person you’re playing with is likely to kill you in the next game, you’re not in an iterated PD. So violence forces people to treat their decisions as if they’re in single-play PD, whereas history should tell them that they’re in iterated, long-play PD. Now, because they have history as well, it’s not black-and-white: if you’ve been in a social interaction with a certain other grouping within the state for two hundred years, then the fact that there was an episode of violence at a certain point in time doesn’t mean you forget about that two-hundred-year history. But it does create a shadow of lack of confidence and lack of trust, which the Report argues makes it very difficult to structure the kinds of cooperation that are necessary to develop institutional patterns of cooperation.

NG: How far does the shadow of violence stretch? Or put another way, when can we expect countries to “graduate out” of a postconflict political pattern?

BJ: This is going to be a really important question for the next phase of the research. There’s pretty good evidence that you can trace the effects of slavery through several generations. If that’s true, then you can be talking about very long shadows.

On the other hand, we shouldn’t exaggerate the effects of these shadows. For example, you could also say that the shadow of the Civil War still hangs over the United States, and that’s obviously true, but it doesn’t affect basic institutionalization and development. So there are a whole host of other factors that determine how much it matters that that shadow endures: the form of the violence matters; and the intensity of the violence matters.

But in general we’re talking about generations, not years. Look at the breakup of Yugoslavia, and the effectiveness with which groups that invoked the historical memory of World War II atrocities chose to do so. This was not some abstract thing; it was a very powerful mobilizer. And that was referencing events from fifty years before, in an advanced society.

NG: Indeed. Even if current institutionalization seems to have put the shadow of violence to rest, political entrepreneurs can potentially mobilize the effect of that shadow for much longer thereafter.

BJ: Yes—and long after any international guarantees are active. Barry Weingast makes the point that political entrepreneurs who are willing to use violence have a real advantage, because if you look at risk calculation, once someone invokes the possibility of violence, risk calculations change.² Even if I think there’s only a 1 percent chance you’re telling the truth and there’s a 99 percent chance you’re full of shit, the consequence of being wrong on that 1 percent are so high that the rational decision is for me to follow you. So if you tell me that a bunch of Tutsi are coming over the hill to kill me, I might think you’re lying to me, but if there’s a 1 percent chance you’re right, I’m going to die. It’s a perverse form of Pascal’s wager.

In short, people willing to invoke the threat of violence have a considerable advantage in political mobilization, at least until extremely strong institutions take hold, and that takes multiple generations. The policy implication is that we can’t be thinking about security guarantees that last three to five years; we need to be thinking about guarantees that last a generation or more. And those sorts of guarantees aren’t going to look like the ones we do now: it’s not going to just be peacekeepers on the ground; you’ve got to be thinking about sustained security guarantees.

NG: The Report spends a lot of time talking about “external stresses” that can lead to adverse developmental outcomes. One of the sorts of “external stresses” that the Reportdiscusses at length is the violence and related developmental disruptions that stem from illicit economies and black markets. In your research findings, how do black markets and illicit commerce lead to developmentally debilitating violence?

BJ: That question is exactly where the research as we have it ends and the research agenda begins, because we know less about that than we should. You can see at least a couple of different pathways, however. One is that you have demobilization after civil war which leaves you with large numbers of mostly young men being organized and trained in certain social groupings; they often are recruited from certain social groupings and go back into their social settings with patterns of organization and patterns of the use of violence that leave them very easily susceptible to recruitment by organized criminal groups. This is the Colombia scenario, for example, and also the story for some parts of the demobilization in Guatemala. You have other scenarios, by contrast, where erosion of state control, or a long pattern of violence, leave you in a situation where law and order is essentially nonexistent. Here you have an open domain where organized criminal groups can insert themselves very easily into that space, and at low cost.

In a separate piece of work we did here at NYU on drug flows (largely around Afghanistan), we concluded that what you see in the global drugs market is that production doesn’t cost very much, and transshipment doesn’t cost very much.³ The real cost is in securing the flow.Drugs therefore flow to where it is cheapest to procure security. There has to be enough institutional development to provide some security, but it has to be really cheap and really corrupt. In that case, organized criminal networks can move at low cost and at speed into places where violence has (already) debilitated social order. In that case they’re not creating it; in that case, they’re using it.

A very different category would be Mexico. Here you have the creation of organized criminal groupings in essence within the state over a long period of development. What causes the violence now is a democratization process, a process of democratic transition where the PRI [the Institutional Revolution Party that governed Mexico for decades] says, “Okay, no longer are we going to be in bed with you guys, we’re going to kick you out,” and they say, “Hmm, thank you, not so much,” and fight back. That’s a very different form of contestation.

 

A Twenty-First Century Problem?

NG: One of the first subchapters talks about violence and conflict as a “development problem” that does not fit “a twentieth-century mold” of development. How would you describe that “twentieth-century mold of development” and why didn’t violence and conflict fit into it?

BJ: It might have been more accurate to say a “postwar mold of development.” There were really two dimensions to development in the postwar era. On the one hand, you had the World Bank itself and other official development institutions, and on the other, you had bilateral assistance. In terms of bilateral assistance, development was a tool of Cold War politics, pure and simple: development was a tool for buying alliances and for reinforcing the Western alliance against the Soviet bloc. Now, this wasn’t quite true of the official development institutions like the World Bank: there you had a model that was mainly organized around poverty reduction.

Overall, the model was premised on what’s referred to as “progressive modernization”—the idea was to help states access “modern” modes of production and “modern” modes of institutionalization, so that they could deal with their development challenges and reduce poverty. This model was premised on a certain understanding of the nature of statehood. You were looking at a lot of nascent states that came out of the colonial experience relatively stable internally, relatively poor, with relatively weak institutions. So the developmental model was one that said, hey, if you’ve got stable states with weak institutions, if you develop those institutions, then they can develop the private sector’s access to markets, and so on.

Of course there were amendments to that model over time, but it remained a model that, at its core, presumed state stability.

It’s only in the post–Cold War period that people start to pay a lot more attention to internal violence. The internal violence was always there, we just hadn’t paid much attention to it. After the Cold War, we start paying a lot of attention to it. And it gradually works its way onto the radar of the development community. But the development community has been surprisingly slow to understand quite how many poor people live in societies affected by conflict and quite how small the caseload is of stable states with large numbers of poor people.

NG: Why do you think the development community has been slow to make that recognition?

BJ: Development is a slow and complex business. I don’t mean that as a pejorative: it’s a hard process, and it takes a long time to see the effects of interventions. The consequence is that you can end up “failing slowly.”

Now, if you look at the way that organizations change—certainly public organizations—highly visible public failure is a crucial driver of change: if you look at why peacekeeping changed dramatically in the first part of this decade, if you look at why humanitarian assistance has changed dramatically, it’s because they have had very visible, high-profile failures.

With development, there’s been a lot of failure, but it’s been slow, and it’s been largely hidden from public view—not because anyone is being secretive, but because these are long, slow, complex processes. So I don’t think the development community has been confronted in a forceful, visible, impactful way by the changes and the failures and the kinds of ways that can drive deep institutional change, in a sense, until now.

The “now” is really an accumulation of slow, grinding failures in a lot of domains. Events in the Middle East helped catalyze a sense of “what the hell were we doing?” but a number of factors helped to catalyze a sense that what we’ve been doing really is not working anymore. But it’s taken a long time to get there.

NG: Have the Millennium Development Goals (MDGs) played a role in this growing realization? The MDGs created a universally agreed-upon benchmark for success when they were made official in 2000, and it’s becoming ever clearer that there’s no way we’re going to get to those benchmarks on the proposed schedule, that is, by 2015. Has that contributed to making the failure of development stark and undeniable?

BJ: The MDGs have changed in terms of what they were designed to do. They were never designed to be programming tools. They were designed as mobilization tools. But they became programming tools, and you’re right: as we get closer to 2015 it’s becoming very clear that we’re not going to meet them. And then you go a little deeper and it’s pretty clear that, to the extent that we’re meeting them in some cases, it’s not because of official development assistance, which is where the MDGs ended up focusing, even though they were born with a broader connection. Instead, it’s because of policy change and economic integration into the global economy, China, and India, and so on: that’s what’s generating huge amounts of growth. So if you look at the development business in the light of the MDGs, it’s not very pretty.

 

The Place of the State

NG: Let me ask about another conceptual point. The concept of “human security” has been a key buzzword in development circles since the mid-1990s. The current WDR, however, makes a distinction between “human security” and “citizen security,” with a lot more focus on the latter. What’s the salience of this distinction?

BJ: “Human security” as a concept came about via the “Human Security Committee” that the Japanese and a few others pushed in the early 1990s in an attempt to broaden the concept of security away from narrow military and physical security toward all sorts of things that could undermine the life chances of people: economic issues, environmental issues, and so on.⁴

The problem with the human security concept is that it’s simply so broad as to be everything under the sun: poverty is a human security issue, injustice is a human security issue, environmental degradation is a human security issue . . . well, okay, everything’s a human security issue: it doesn’t tell you much.

We were trying to come up with a way of talking about how “hard” security—the enforcement of law, the use of force to impose the will of the state, the killing of people or the protection of people—can be oriented toward either the protection of the state, or toward the protection of citizens, in order to emphasize that those are very different things. Just using the word “security” doesn’t give you that distinction; we couldn’t find a term that did, so “citizen security” (or “security that is oriented toward citizens”) was what we came up with. So it’s an attempt to be narrower than “human security” but still get at the idea that the security tools of the state shouldn’t just be aimed at protecting the assets of the state, but should also be used to protect citizens and their rights. You could also use the term “democratic security,” but that was a bit grand, and we were trying to say something modest and practical about the orientation of security services.

NG: The place of the state in this WDR is one of its most interesting aspects. Traditional development—the same post–Cold War model of development we were speaking about earlier—really focused on the state and the nation-state as the object of development. But one of the striking things about this WDR is the way it decenters or deemphasizes the state, both in the characterization of the problem and in the kinds of solutions it proposes. There’s a lot of emphasis on supranational institutions, regional cooperation, and subnational entities. It’s not that the nation-state gets ignored, exactly, but it’s nonetheless a pretty striking departure from the Bank’s traditional view of the state as the central focus of developmental action. How does the Bank square these findings with its charter as an international organization, that is, as a device of collaboration between member states?

BJ: Well, we’ll see how they square it! Remember, WDRs are independent research publications. They may be welcomed by the [Bank’s] board, but they are not products of the board. Now, this one got a lot of positive reception from the board, so I’m not suggesting the board was distancing itself from it, but it’s important to bear in mind that the board is not necessarily adopting this as a framework.

With that said, a couple of thoughts. First, we’ve obviously seen a substantial progression and advance of globalization, which means that for any economy that’s trying to figure out how to grow and deal with development issues in the contemporary moment, the question of access to global markets is fundamental, in a way that wasn’t true even ten years ago. (This is less true for the poorest states; but for all but the poorest states, it’s a fundamental question.) So, the reality is that not everything is going to happen within the boundaries of the states. Second, to say that the state can protect itself from all the different stresses we’re talking about is to deeply ignore how much these stresses emanate from subregional and global processes. And the corollary is that you have to address these issues at that level.

Now, how much that gets translated into practice remains to be seen. Some of this is controversial. If you look at the question of subregional administration—the notion that the solution to some of these issues is to pool administrative services at a subregional level—that’s the kind of thing that’s easy for us sitting here at the Bank to recommend, but the politics of actually trying to do that at a national level are tough. But, it has happened in some places.

We know it’s going to be controversial, but these issues can only move by being put on the table. It will take some time to gestate but hopefully will generate some debate and discussion, and eventually some effects.

 

The Role of Institutions

NG: Francis Fukuyama has recently written that the central challenge of (political) development is “getting to Denmark,” by which he means getting to institutions that are stable, democratic, peaceful, prosperous, inclusive, with low levels of corruption.⁵ Why is institutional transformation so difficult?

BJ: The first thing I would do is to amend Fukuyama by saying that while the long-term goal is to get to Denmark, you have to get there by way of Botswana. A big part of what thisReport is saying is that, if people go into Burundi and think they’re going to build Denmark, they’re going to fail: you’re overloading institutions, and you’re overloading the expectations, and it’s going to backfire. On the other hand, if you go into Burundi and you try to build Botswana, you might succeed, in a generation or two. Once you’re in Botswana, then you can aim for Singapore—and maybe Denmark after that.

NG: The Report emphasizes the centrality of institutions as a solution to violence. But isn’t violence, in some sense, always about institutional failure? Isn’t there then something paradoxical (if not quixotic) about an institutional approach to institutional failure?

BJ: Well yes, of course this gets a bit tautological; but on the other hand, if violence is a result of the failure of institutions, then institutions must also be the answer to violence.

It’s important to emphasize that the Report only gets to institutions in the second phase. There’s a lot of emphasis in the Report on the justice, security, and jobs of institutions as the core developmental messages. But well before that, there’s a very strong emphasis on political settlements, confidence building, trust, and commitment technologies. None of those are “institutions,” in the sense the Bank would normally use the term. And our point is that you have to do those things first; only once you’ve done a bunch of that and changed the character of confidence do you have the basis for building economic institutions.

NG: The question of political settlements brings up another crucial conceptual issue. TheReport calls, centrally, for “inclusive enough” institutions, in order to bring potential or actual peace disruptors into the political process and thus give them incentives to move out of the cycle of violence. But in practice, isn’t the debate always about precisely where to draw the line? It seems as if the question of whether to include a particular constituency in a political dialogue has always been the central sticking point, whether we’re talking about the IRA in Ireland, ETA in Spain, Hamas in the Occupied Territories, the Taliban in Afghanistan, or what have you. In all these cases, there’s a power that wants to exclude these groups from the dialogue, and often the refusal to give them a seat at the table is, in fact, a central cause of the conflict in the first place. So our question is: how does the concept of “inclusive enough” help decide when it is (or is not) legitimate to exclude a particular group or actor from the political dialogue? And what is the process by which one makes an adjudication of that sort?

BJ: Two things to say here. First, contrary to the trend of the last few years, it’s not the business of external interveners to make that call. (Which we’ve been doing: “Thou must not talk to the Taliban,” and so on.) When international interveners make that call, it usually creates more problems than it solves.

Now, clearly there are some groups that have placed themselves beyond the pale from the perspective of any other part of society, and there may well be times and places where including such groups in a settlement is counterproductive—that’s the concept of “inclusive enough.”

With that said, obviously you’re right that it’s contested. Essentially, this is a message to those groups who want to do the excluding. Our point is Cave! Hic dragones!—go ahead, but you’re going to find that you’re just stoking up the next war. This is the message to Sri Lanka: “Sure, you can get away right now with excluding the Tamils from the settlement, because you just defeated them; but you can’t exclude them forever, and if you try, it will come back to bite you.”

We’ve seen this again and again since the end of the Cold War: countries that try to move ahead on the basis of settlements that exclude significant groupings fail. States simply are not powerful enough to sustain such exclusions over time: there’s not the flow of international support that there was during the Cold War in terms of money or weapons to sustain that exclusion. So, the exclusionary strategy is going to fail.

Of course, that doesn’t mean that some leaders can’t profit from the exclusion in the short term, and so decide not to worry about the long term. But the argument here is that if you want to be on a stable developmental path, then you need to take an inclusive enough strategy, and stretch the definition of “inclusive enough.”

NG: Of course, if we’ve seen anything over the last decade, it’s that the international community has a really hard time keeping its fingers off of these things. What kinds of external intervention, if any, are appropriate, in developmental terms?

BJ: If you thought—the way we did during the Cold War—that violence was basically an extraordinary episode for a generally stable state, then it might be reasonable to say, during that phase, “Stay out, stay away.” But that’s not what we’re seeing.

What we’re talking about, for the most part, are states that are stuck in long-term patterns of violence—patterns that are recurrent over time. In these cases, the “stay out” strategy just isn’t very credible or realistic. The international community is going to try to find ways to engage. So our analysis is trying to grasp the real world, where international actors are part of the environment, and we can’t pretend them away.

So the first thing we’re saying isn’t really to the development community. This is first and foremost about trying to navigate toward a political settlement, and then to guarantee and secure that political settlement: it’s about peacekeeping and related kinds of activities. And what we’re saying is, “If you’re preoccupied with reducing poverty and supporting development, then those are developmental interventions in these contexts,” and they should be supported as such and funded as such.

Once you’ve got the peace and security settlement, then there are other kinds of roles: there are economic issues, employment issues, institutional issues, services issues. But none of those are relevant in the absence of those basic political and security interventions. In these kinds of contexts, it just doesn’t make sense to draw a hard line between what constitutes security and what constitutes development.

NG: Assuming some of these recommendations are taken up, it suggests the need for a very different kind of institutional and collaborative relationship between the Bank and the United Nations and other international institutions.

BJ: The idea of joint operations goes exactly to that point. This is not about light coordination, or about some vague sort of sequencing. This is about recognizing that most of the levers we have do not fall neatly within the normal categories of development, security, politics, humanitarianism, and so on. If we’re serious about trying to end violence and help these states move out of a pattern of recurrent violence into something more stable, we’ve got to move past these narrow categorizations and stovepipes and toward something much more in the domain of blended and jointly owned and jointly commanded operations. That’s a very radical statement from the Bank’s perspective, and it’s almost as radical from the UN’s perspective, but we just think it reflects the reality on the ground.

Consider the G7+ countries, a group of seventeen of the world’s most fragile states, who have been saying for some time, “Look, development aid and the MDGs aren’t doing anything for us. We need investments in political settlements, we need investments in security institutions, we need investments in justice systems. Maternal health and all these things—fine. But in the absence of political stability, basic security institutions, and some sort of justice arrangements, we can’t get there, so start helping us with those political/security issues.”⁶ That’s going to be a big fight in the next few years as we get to the negotiations about what happens after the MDGs. The fact that the fight is being kicked off by and led by fragile states themselves is very important.

When you look at these recommendations, it’s easy to say that you’re going to run into issues of the Bank’s mandate, but this is one of those cases where theory and the law are behind the practice. In practice, the Bank and the UN have always found creative ways to do things. Neither of their mandates, nor the narrow legal procedure, would tell you that that’s the case, but in practice it’s doable. It’s really a question of leadership.

 

NOTES

1. Robert Zoellick, “Changing Trends in Global Power and Conflict Resolution,” International Institute for Strategic Studies, September 12, 2008, http://www.iiss.org/conferences/global-strategic-review/global-strategic-review-2008/keynote-address (accessed September 17, 2011).

2. Rui J. P. de Figueiredo and Barry R. Weingast, “The Rationality of Fear: Political Opportunism and Ethnic Violence,” in Civil Wars, Insecurity, and Intervention, ed. Jack Snyder and Barbara F. Walters (New York: Columbia University Press, 1999), 261–302.

3. Barnett R. Rubin and Jake Sherman, “Counter-Narcotics to Stabilize Afghanistan: The False Promise of Crop Eradication” (paper presented at the Center for International Cooperation, New York University, February 2008).

4. “Human security” was the focus of the UN’s 1994 Human Development Report, New Dimensions of Human Security (Oxford: Oxford University Press, 1994). Japanese Prime Minister Murayama Tomiichi endorsed the concept the following year in a speech at the United Nations, and three years later Prime Minister Obuchi Keizo;am declared human security a “key element of Japan’s foreign policy.” For background, see Bert Edström, Japan and the Challenge of Human Security: The Founding of a New Policy 1995–2003(Stockholm: Institute for Security and Development Policy, 2008).

5. Francis Fukuyama, The Origins of Political Order: From Prehuman Times to the French Revolution (New York: Farrar, Straus and Giroux, 2011), 14.

6. See http://www.g7plus.org/history (accessed September 17, 2011).