According to the jury citation, the professor at Northwestern University (Chicago, United States) has been a key figure in the development of modern econometric methods that have transformed how economists infer conclusions from data, recognize the degree of uncertainty in their models, and evaluate public policies when the evidence is incomplete.
“The methods he developed assess the degree of confidence that can be placed in empirical measurement,” the citation states, and for this reason his contributions have made him “a critical conscience of measurement in the social sciences.”
Manski’s work “has uncovered some of the erroneous assumptions we economists are prey to, which make our predictions and understanding of behavior quite fragile,” says committee member Sir Richard Blundell, David Ricardo Professor of Political Economy at University College London (United Kingdom).
Manuel Arellano, Professor of Economics at the Center for Monetary and Financial Studies (CEMFI) of Banco de España, and secretary to the committee, hails Manski’s prominent role in promoting the use of surveys to incorporate economic agents’ expectations into economic analysis. Today, central banks such as Banco de España, Banca d’Italia, and the New York Federal Reserve conduct surveys that gather, for example, the probabilities people assign to their house prices or income increasing, or to being unemployed in the future.
Going with “deep uncertainty” over “incredible certitude”
“Most economists,” says Manski, “don’t deal with uncertainty. They would prefer to get firm answers to questions. And this is especially the case in studying public policy. The public wants to have answers to know whether a policy is good or not, and economists like to provide them.” The result is that conclusions in economics are frequently characterized by what he calls “incredible certitude,” with figures and percentages that lack robust empirical support.
Manski has pioneered the development of methods to study situations of “deep uncertainty” that arise in difficult public policy problems. “Instead of providing a point estimate of some quantity, like what the tax revenue will be under certain income tax policies, I might give a bound, an interval.” Ideally, the bound should be as small as possible, but this comes with a trade-off, Manski warns. “You can tighten the interval by making more assumptions: the more you assume, the narrower the interval can be, but that entails a danger. You can draw strong conclusions by making strong assumptions, but then they won’t be credible.”