This blog post is co-authored with Mead Over at the Center for Global Development and is also posted at the CGD blog . The comments of Phil Keefer, David McKenzie, Michael Toman, Berk Ozler, Dominique van de Walle, Adam Wagstaff, David Wheeler and Michael Woolcock are gratefully acknowledged. These are the views of the authors alone.
These days there is a lot of discussion within development organizations and governments across the globe (including the World Bank) about how to assure a greater emphasis on development impact. While it would no doubt help if senior management gave stronger verbal signals on the ultimate goals of the institution, and more actively supported staff to attain those goals, such “low-powered incentives” have been tried before, and the problems seem to persist.
What seems to be missing is a way to more fully align the incentives of managers and operational staff with sustained development impact. Our experience is that many development professionals care passionately about having impact. However, currently development agencies appear to assess their staff primarily using easily observable bureaucratic and procedural measures, such as “the amount of money moved,” which may be poor indicators of longer term impact. (One is reminded of the old story of the guy who looks for his lost car keys at night under the lamppost—not because that is where he lost them, but rather because the light is better.)
Accountability for past results also seems weak. While client countries feel the effects of a staff member’s work for years or even decades, the development professional has often transferred to other countries and tasks.
We think it would improve the incentives in development agencies if development professionals were able to acquire a visible and abiding stake in the success of the projects in which they invest their time and effort throughout their career – and if that accumulated development impact were a factor in their professional recognition and financial compensation.
Here is our suggestion. The central idea is that each professional staffer in a development agency accumulates a “Development Portfolio” consisting of the number of “shares” of all the development projects or policies in which she has devoted time during her career. A professional’s current Development Portfolio is an objective and quantified version of her résumé, reflecting the development policies, problems and clients on which she has spent her time. Our innovation is to suggest that an outside party associate a virtual “value” to the shares of every project, so that any individual professional’s portfolio of shares could be valued and defined as her “Development Impact Wealth” (DIW).
How would a staff member acquire her DIW? At the beginning of each year, unit managers and project managers advertise their staff needs for each project or task in terms of weeks of effort, much as they do now. In addition, in association with each task, the managers designate a number of shares per staff week, which those working on those tasks will acquire as they complete their work. As staff members allocate their time among these tasks, they will be acquiring the shares associated with them. At this point in the project cycle the shares have not yet been valued. There is a portfolio of stocks, but it is not yet Development Impact Wealth.
How would individual development projects and policies be valued? This whole idea rests on the proposition that it is possible to score, at least approximately, the development impact of an individual development task. Classical development projects in the major development organizations are already receiving valuation scores shortly after project completion. (At the World Bank, the Independent Evaluation Group performs this service.) When done well, these “process evaluations” (as distinct from impact evaluations) are very useful for determining whether the project could have had the intended impact. Implementation of DIW could begin by using these scores to value the shares in each staff member’s portfolio as soon as the first post-completion evaluation report is issued. This information base could be expanded to include periodic updates to the assessments of past projects when new information becomes available, including from impact evaluations. The latter can be expected to play an increasingly important role, given the progress that has been made in our ability to credibly assess the impacts of development projects and policies. (We are building lampposts in places where the car keys might actually be found.) Ten years ago, when one of us outlined the travails of the fictional character, Ms Speedy Analyst working for the government of Labas and trying to solve the “Mystery of the Vanishing Benefits ,” impact evaluations of real development projects were quite rare. That has changed, although there is much scope for further progress, especially in making impact evaluations more relevant to the needs of development practitioners (as discussed here ). The efforts put into improving the quality and coverage of evaluations (both process and impact) will depend on what actions are taken based on the results. Tying staff and managerial assessment to evaluative efforts would probably help in supporting those efforts going forward, since the results will matter more than they do now.
How would a professional’s DIW score be used? At one extreme, an individual’s DIW score could simply be a means to honor professionals whose contributions over the years have had lasting impact for the client countries. For example, a development institution could publish each year on its website the names and DIW scores of its top ten staff members that year, with a retrospective look at their best projects. Or, in order to exert more leverage on manager and staff behavior, some or all of staff compensation could be based on the DIW. For example, merit increases could be tied to the changes in DIW since last year, while promotions are tied to the value of the entire portfolio.
How would this system reward work in difficult environments? With such a system staff members would have an incentive to allocate their time to projects which are likely to gain in value. If shares of all projects in all client countries are treated the same way, the introduction of the DIW would only exacerbate a problem already evident in development agencies – a preference to work in “easy” countries and on “easy” projects. To counteract that tendency and instead incentivize creative work in the hardest countries and on the most challenging projects, a staff member must be able to “buy” shares of hard projects less expensively (in terms of work weeks) than she can acquire shares of easy projects. The initial value of project shares per staff week could be set by the development agency to achieve this goal, but in a large enough agency (like the World Bank) it might be possible to establish an internal marketplace. The “time price” of a share of an easy project would rise (in terms of weeks of the staff member’s time), while the share price for difficult countries or projects would be driven down.
What if a project performs badly through no fault of its designers? The development professionals who supervise a project after its birth are often even more important to the project’s success than are those who design it. So they too should receive shares in the project commensurate with its difficulty. But all who work in development must accept the inevitability of so-called “country-risk” – when the political winds shift in the client country and the project is terminated despite its success. The DIW system can encourage staff risk-taking partly by granting more shares per work week for work on riskier projects (as described above). But in order to cushion staff members against some downside risk, perhaps every staff member could be allowed to exempt some of her worst performing shares from inclusion in her portfolio in any given year.