The Ecconomics of Organizing Economists

8 July 2008 at 9:43 am Peter G. Klein Leave a comment

| Peter Klein |

Most regulatory agencies are staffed by a mix of attorneys and economists. Members of these groups do not always play well together. How, then, should such an agency be organized — functionally, putting the economists in a single unit, reporting to a chief economist, or divisionally, spreading the economists throughout divisions organized by legal issue, industry sector, geographic region, etc. and having them report to an attorney in charge of each division? An interesting application of the U-form versus M-form problem posed by Chandler (1962).

An analysis of organizational structure at US and European competition agencies by Luke Froeb, Paul Pautler, and Lars-Hendrik Roller (via Dan Sokol) finds that

the main advantage of a functional organization is higher quality economic analysis while the disadvantage is that the analysis may not be focused on legal questions of concern, and is less easily communicated to the ultimate decision makers. Likewise, the advantages of a divisional organization are decentralized, and faster, decision making; however, the quality of the economic analysis is likely to be lower and can result in less information reaching the ultimate decision makers.

Froeb, Pautler, and Roller suggest that hybrid forms, such as (a) functional organizations with strong horizontal links between economists and attorneys or (b) divisional organizations with strong vertical links between economists and attorneys, and managers trained in both law and economics, are best.

My own experience at the CEA confirms the importance of both the vertical and horizontal links. We economists were organized into a focused unit for major projects like the Economic Report of the President but were also assigned to ad hoc, inter-Agency teams working on specific policy issues (I dealt with spectrum auctions, the pricing of air traffic control, and foreign ownership of domestic telecom assets, among other things). I was typically the lone economist (though hardly the nerdiest member) on each team.