Syndicate content

Are the Knowledge Bank’s staff under-specialized?

Adam Wagstaff's picture

Ideas often come from unexpected quarters. Last week, Ricardo Hausmann came to the World Bank to talk about his work on economic complexity. I missed the seminar, but afterwards read his Atlas of Economic Complexity: Mapping Paths to Prosperity. (I had actually already looked at the stunning – but rather confusing charts – of his coauthor Cesar Hidalgo after reading Tim Harford’s great new book Adapt: Why Success Always Starts with Failure.)

On the face of it, the Atlas of Economic Complexity doesn’t have a lot to do with the topic of this blog post – whether World Bank staff are under-specialized.  But bear with me, and I hope I’ll convince you otherwise.

A reinterpretation of Adam Smith

Hausmann and his colleagues take a new – and very graphical – look at what international trade reveals about a country’s wealth. Products, they argue, embed knowledge. The knowledge underlying some products is rather basic, and can be mastered by one individual. Other products embed a lot of very complex knowledge that no single individual could ever hope to master.

Giving a modern twist to Adam Smith’s idea that the secret of the wealth of nations lies in labor specialization, Hausmann and his coauthors argue that the key to the division of labor is that it gives us access to a quantity of knowledge that none of us would be able to hold individually. The secret to a successful economy is that “it weave[s] vast quantities of relevant knowledge together, across large networks of people, to generate a diverse mix of knowledge-intensive products. Simpler economies, in contrast, have a narrow base of productive knowledge and produce fewer and simpler products, which require smaller webs of interaction.” Just to make sure the reader has really got the point, they add: “Because individuals are limited in what they know, the only way societies can expand their knowledge base is by facilitating the interactions of individuals in increasingly complex webs of organizations and markets.”

As knowledge accumulates, it needs to be parceled out across more and more people. The constraint is not “explicit” knowledge, which can easily be transferred from one person to another and one country to another, but rather “tacit” knowledge which is costly to acquire and costly to transfer. So people specialize, each person acquiring a limited set of deep skills and a limited number of chunks of deep knowledge – the key is to make sure these skills and knowledge are mutually coherent so the person can function effectively. But a successful economy needs not only people who know how to do things, but also people who know who these people are (“know-who”) and people who know where they are (“know-where”).

What’s the relevance of this to the World Bank?

On the face of it, the World Bank is a highly specialized place. It has five networks, each subdivided into several sectors. The majority of people work in and are affiliated with (or “mapped”) to just one sector. The boards of these sectors are responsible for overseeing the professional development of their staff complement, deciding promotions, and setting broad priorities in terms of skill mix – how many economists there should be relative to specialists, operations officers, communications officers, and so on.

But my sense – this comes in part from my experience sitting on four sector boards – is that the Bank’s staff are becoming less specialized, not more specialized as Hausmann’s diagnosis of a successful economy might lead us to expect. Following Hausmann’s logic, we would expect to see just one or two Bank staff knowledgeable about topic X, another one or two specializing on topic Y, and so on. And given the global reach of the Bank, its sheer size, and its pulling power (it attracts very bright and able people from across the world), its staff would be global experts – not household names perhaps but certainly people who command a solid global reputation.

There are, of course, some such people, but my sense is they’re becoming fewer and fewer. The trend, it seems to me, is away from specialization. Staff are increasingly expected to be able to do high-quality analytic work, task-manage lending operations, and lead policy dialogue with governments. There’s been a consequent blurring of profiles – economists are expected to have the sector-specific knowledge of a specialist, and specialists are supposed to acquire the economics of an economist. Both are expected to acquire the operational skills of an operations officer. And they’re expected to play these roles across multiple topics within their sector, often simultaneously. At promotion time, boards seem to look increasingly favorably on the all-rounder.

The trend away from specialization is evident too in the institution’s approach to “knowledge management”. There’s a growing frustration that knowledge flows too slowly around the Bank. The frustration is greatest over the slow flow of tacit knowledge. The institution’s rotation policy is a response to this frustration – unable to transfer tacit knowledge around the Bank as quickly as it would like, the institution has resorted to transferring people from one region to another every five years.

Applying Adam Smith to the internal organization of the World Bank

This is all a far cry from Hausmann’s reinterpretation of Adam Smith. By analogy with the Hausmann diagnosis of a successful economy, we’d expect to see an institution that calls itself “the knowledge bank” to be encouraging an ever higher degree of specialization by its staff. Staff would specialize in a specific discipline but also within it. One staff member might become a world expert on the economics aspects of some particular issue in the field of education. She wouldn’t work on one region – rather she’d work across the world, being brought in to build a particular component of a project, or to contribute to or lead a particular part of a broader study. She would become a global expert in her niche area. She wouldn’t be expected to task-manage a lending operation.

That job would be done by someone who would specialize in putting together a project and a project team – he would have specialist knowledge of the nuts and bolts of assembling a project, but would also have the necessary “know-who” and “know-where”. He might specialize regionally – getting a successful project assembled in Latin America takes a different set of skills, knowledge, contacts, and cultural sensitivities than is required in, say, South Asia. Or rather than specializing regionally he might specialize by type of project or lending instrument – he might specialize in program-for-results operations while someone else might specialize in development-policy loans.

A different set of people might specialize in policy dialogue and related activities, knowing the tricks of a successful policy engagement and knowing which technical specialists to draw on at the right time. The project specialists and policy-dialogue specialists would need to acquire knowledge specific to their task, but also to be good connectors of topical knowledge and of the people where this knowledge is stored. It calls for strong task-management and interpersonal skills, as well as an ability to think creatively about resolving issues. Of course, it would be inefficient for the project specialist or the policy-dialogue specialist to acquire too much knowledge about specific topics in development. They would need to know enough to see which pieces are required at a particular moment, and how they fit together. By the same token, it would be inefficient for someone who specializes on a particular development topic to acquire much more than a smattering of knowledge on other development topics, even ones that might appear to an outsider to be closely related to their specialty.

This would translate into an entirely different set of human resource management considerations and conversations. In contrast to market economies where specialization emerges through a process of trial and error with survival depending on success in the global marketplace, an institution like the Bank would need to do some corporate thinking on who plays each of the various roles. Knowledge management would look different too. Efforts to extract and spread tacit knowledge would be recouched and likely limited to two distinct types of knowledge-sharing: deep technical exchanges among the relatively few experts on the particular topic, with an emphasis on exchanges with experts outside the institution; and “lite” bite-sized knowledge-sharing activities to keep the project and policy-dialogue specialists sufficiently knowledgeable to be able to keep connecting together disparate pieces of knowledge and the experts holding this knowledge.

Some parting questions

I’d be curious to know how knowledge-based institutions that are subject to market competition organize themselves. Do they operate the type of model I sketched above? If they do, are there reasons why an international organization should opt for a different approach? Is the Bank really becoming less specialized? If it is, does it matter?


Submitted by T.V. Somanathan on
Thank you for that extremely thought-provoking application of the principles of specialisation to the World Bank context.

Submitted by Tony Thompson on
Adam, Thanks for a very interesting blog post which is particularly thought provoking. Here are a couple of thoughts provoked by your piece. 1. Are you familiar with "The Checklist Manifesto" by Atun Gawande? An excellent and readable little book that addresses collaboration in a world of increasing complexity. I would particularly recommend chapter 3 "The End of the Master Builder" which addresses how multiple specialist skills sets organize to collaborate effectively and efficiently in the context of a large scale civil engineering project. I believe there are definitely relevant lessons for us at the Bank 2. Integrators versus specialists. A while ago the Bank was using this integrator and specialist terminology. The integrators were in fact specialists (by your account) in sourcing and combining expertise in contrast to those who contributed deep expertise within narrow technical areas. In the Bank it seems that the one group believes the other group is dominant. Your experience on the sector boards is fascinating. In contrast to what you experience, the perception from where I sit is that the Bank continues to emphasize and reward technical specialization over all else including the integrator function (but then again I'm an integrator so that fits with my stereotyping.) Hopefully this blog will be a nudge to re-establish a balanced perspective which recognizes and rewards both elements equally. That would be a useful discussion. 3. Where does the Bank gets its specialist knowledge? Here I think it is useful to think in terms of firm-specific versus general human capital. The project and integrator functions tend to have some very firm-specific elements to them - complying with the Bank's processing guidelines, packaging different knowledge inputs into a single product/process, etc.. These would then seem to be the skills which the Bank should retain and develop on a long term basis. Deep technical skills in particular areas are, by definition, less firm-specific forms of human capital. These forms of expertise may then be best acquired through STC type arrangements unless there is some reason why the Bank needs to capture these skills and hold them internally. One would then expect the bulk of Bank staff to be specialists in firm-specific functions with the bulk of Bank consultants providing the specific technical expertise. One argument for internalizing these skills would be that it is the cheaper way to access them. Bank staff working in, say, the Financial Sector, may be cheaper than the equivalent consultants on the open market. Presumably they are willing to forgo some of their external market premium for the opportunity to be identified more closely with the Bank's mission, better job security, etc.. It would be interesting to know to what extent the Bank's staffing profile fits this kind of model. 4. Another angle to approach the issue of specialized skills is to think in terms of the value addition which comes from simply being a Bank staff member. The vast majority of us work at the Bank (or at least came to the Bank) because we wanted to associate ourselves in some way with the Bank’s mission to reduce poverty. Given that this represents in some way our individual goals, the question then becomes are those individual goals best met by staying within the Bank or by leaving and seeking to achieve those goals by some other means? For the technical specialist, what is the added value of belonging to the Bank? I suspect it continues to come, even in the world of Access to Information, from having privileged access to people and information. This, after all, is part of the value of the Bank brand. This privilege potentially provides the Bank’s technical specialists with an advantage over their peers working outside thereby providing value addition, not just to the individual, but also to the Bank. This is another way to look at the question of specialist skills, how we source them and whether we should contract certain skills sets for the long term or not. Many thanks again!

Thanks, Tony, for the kind words and these great points. 1) I have Gawande's book, and will take another look at the master builder chapter. 2) I'm not sure about the specialist vs integrator terminology. What I'm wondering is whether everyone shouldn't be a specialist of some sort. They could specialize in a particular aspect of the Bank's work (e.g. DPL's), a specific topic (e.g. civil service reform), a specific academic discipline (e.g. political science), or some combination (e.g. the political economy of civil service reform). The term "integrator" is consistent with a complete lack of specialization, isn't it? And "specialist" doesn't capture the idea that people can specialize in things other than disciplines and topics. 3) The firm-specific versus general skills distinction makes a lot of sense. You clearly wouldn't want to have an army of STC's task-managing Bank projects, but you might get away with outsourcing topic-specific expertise. My guess is you wouldn't want to outsource that completely, because it's through discussions between people with different specialties that activities take shape. Without anyone with the topic-specific knowledge in-house, you could end up contracting five different consultants before you get it right. A one-hour brainstorming with a bunch of in-house "boffins" could save a lot of angst and time. That said, you clearly wouldn't want to have someone in-house with one very narrow specialty who gets called on once every five years. So you'd encourage multiple specialties and/or have highly specialized staff only where their area of expertise gets used on a daily business. 4) This is related to the third point, right? Whatever they specialize in, Bank staff accumulate expertise in applying knowledge in the specific fora the Bank operates in. That certainly makes them valuable to the Bank, making the outsourcing option for non Bank-specific skills and knowledge look doubly unattractive. And yes I'm sure it's the opportunity to help make things happen that make technical specialists want to acquire the expertise in applying their knowledge. I guess the interesting thing here is the vast array of new opportunities that have sprung up outside the Bank that also give people the scope to acquire experience and skills applying their knowledge to development problems. But that's for another blog post! Thanks again, Adam

Submitted by Tony Thompson on
Adam, Great points! What I liked about your idea that everyone is in some way a specialist is that it also applies to what was previously known as the integrator function. I see that has having the potential to be defined and developed as a highly specialized skill set in its own right which is what I believe you are proposing. There are examples of development agencies which have moved towards the end the spectrum in terms of contracting in technical expertize on an as-needed basis. USAID is a case in point. What you see there is a business model that is not slightly different, its substantially different from that of the Bank. For me the contrast usefully highlights some of the drawbacks of the alternative and some of the strengths of the Bank's model. One of the benefits I have enjoyed from my work at the Bank, and which I believe is an important asset to the institution, is the opportunity to accumulate cutting-edge knowledge acquired through engagement with client counterparts at the country level in real time when working on particularly development challenges. I found that in these cases, when dealing with people who had left the Bank even fairly recently, their knowledge of latest thinking was quickly becoming outdated. I don't think the Bank does enough to recognise the access to innovative thinking and opportunities that is inherent in the way we do business and to factor it into both its talent management and knowledge management systems. Looking forward to the next blog on the comparative merits of working inside versus outside the Bank! All the best, Tony.

Submitted by Wilson Magaya on
I have always followed some of the bank's way of doing things and I have found it well intriguing but very frustrating at the same time. This piece of writing has moved me to comment on a network i would otherwise observe rather than participate. Some time back I came across a paper from the bank on knowledge management, (around 1994) which i think was the watershed for the bank branding itself as a knowledge bank and of course this left me convinced that the bank staff are part of the prime knowledge workers of the world. Adam you speak of specialization from the points of know-how, know-who, know-where. I donot work for the bank, but I have made several observations and these are; Although you may feel there is less specialisation, the bank tends to hire specialized personnel, so on entry one is a field specialist in some subject matter, or they have that specialization and a skill such as networking, project management and so on and so forth. These skills do cover the how, who and where but it ommits the part that enables tacit knowledge to flow from one to the other through doing. there is the guy who knows how and doesnot do and then the guy who does bit know little of the how. that guy is missing in the organisation. Everyone is learned so is not easy therefore to teach. As Tony points out Bank staff are very knowledgeable but it is a pity they donot have an opportunity to be rewarded for teaching others what they know. All along I thought the WBI was a center where Bank staff (operations) take time to teach their own as well as outsiders but I see now that there is little formal staff on staff platforms for teaching and learning, "formal platforms" Your parting questions I feel are searching for answers that would be inappropriate to the bank but valuable to fire the discussion. I think it is incumbent upon bank staff particularly in WBI to take ownership of the definition of the bank as a knowledge organization and create networks of teachers and learners within the bank and beyond. The practical element of the whole process are the missing link. the bank has a lot of thinkers and scholars, it needs some doers.

Submitted by Anonymous on
Adam, This post came at a perfect time for me. I worked at the Bank for more than 5 years and now I’m doing a master’s that can be described as ‘generalist’. I pursued this in spite of the fact that almost everyone I worked with advised me to go “get a skill” or “an area of expertise”. The thing is that, as you point out, I didn’t see that being an expert was necessarily where things were moving. More than that, I don’t think that being an expert is the way to become a good task manager (which is what I would love to be one day). Another thing I saw is that the Bank is biased towards hiring subject-matter specialists and, even when they are hired to be project managers, their management skills are not given that much weight. You can see that also in the YPs that come in with either very narrow and specific professional backgrounds or PhDs. One of the most recurring conversations I had with engineers, anthropologists, and even doctors while I was at the Bank was the sense that they were losing their expertise, they would say “I’m becoming better as a task manager but worse as a [fill in the blank]”. I’m sure that’s not the case for all types of expertise, but there is something to be said when a mid-career transport engineer comes in and a 24 year old JPA has to be the one mentoring him for more than a year… it feels like a waste of resources. So, I really hope that your thoughts mean that the Bank is looking into these issues more seriously, and that conclusions end up being reflected in the recruitment and selection process, as well as in terms of career development. Just a few more, somewhat unconnected, points: i. There are a lot more incentives and career advancement paths for generalists and managers than for specialists, and that contributes to making people who either were experts at the beginning or had the potential to become one, to change their course; ii. I cannot agree more with your point about STCs, the Bank cannot rely on them for expertise exclusively, in-house experts are, more than useful, necessary, and iii. The Bank can be a “knowledge bank” but its value is also in the application of that knowledge to tackle very complex issues in the field, so I would put as much weight into the manager/practitioner side as into the more formal knowledge/academic side.