Expert advisers: Why economic forecasters can be useful even when they are wrong

Abstract

The dilemma posed by expert consensus can be summarized as follows. On the one hand, it seems perfectly reasonable to give special weight to an agreement reached by those who have studied a topic in great detail. On the other hand, does the very same specialization that confers expert status also mean that the group is unable to consider all alternatives equally? In other words, do the shared analytic models and other practices that expert groups rely on provide an enhanced understanding or mean that they focus only on those aspects of the problem that fit neatly into their pre-conceived way of thinking? The problem for policy-makers and those who would rely on experts is thus to identify which experts they need to consult. If the boundary is drawn too tightly, reaching consensus may be easy but its practical applicability may be highly restricted. If the boundary is too porous, then consensus may never be reached and the epistemic quality of deliberations may suffer as a result of irrelevant or unfounded concerns. If we accept that non-coercive decision-making is appropriate, then the problems of expert consensus are twofold: first, the members of the expert group must be identified and, second, the relationship between these experts and the wider society must be clarified.