という事例を少し前にEconomic Logicが紹介していたことを、デュフロがオバマ政権のグローバル開発委員会入りするというニュースを目にして思い出した。

以下はEconomic Logicianによる前説。

Randomized experiments are all the rage in some circles, for example labor economics and especially development economics. The principle is simple: create some intervention in some market, randomly draw a group of economic agents that has access to the intervention, leave the others out, compare outcomes. In all that, you hope the behavior of the non-participants is not affected by the presence of the program to the others. This can be a heroic assumption, for example because market prices may respond for everyone to the intervention.



以下は該当論文(「Estimating Equilibrium Effects of Job Search Assistance」、著者はPieter Gautier, Paul Muller, Bas van der Klaauw, Michael Rosholm and Michael Svarer)の要旨。

Randomized experiments provide policy relevant treatment effects if there are no spillovers between participants and nonparticipants. We show that this assumption is violated for a Danish activation program for unemployed workers. Using a difference-in-difference model we show that the nonparticipants in the experiment regions find jobs slower after the introduction of the activation program (relative to workers in other regions). We then estimate an equilibrium search model. This model shows that a large scale role out*1 of the activation program decreases welfare, while a standard partial microeconometric cost-benefit analysis would conclude the opposite.



具体的には、渋滞効果(congestion effect)が生じた、ということのようである。


トラックバック - http://d.hatena.ne.jp/himaginary/20130111/how_randomized_experiments_can_go_very_wrong