This discussion paper resulted in a publication in 'The World Bank Economic Review'.
Support for economic reforms has often shown puzzling dynamics: there are many examples of reforms that started off successfully but nevertheless lost public support, and vice versa. We show that learning dynamics can rationalize this apparent paradox, the reason being that the process of revealing reform outcomes is an example of sampling without replacement: every winner revealed reduces the number of unfilled winning places left, thereby making individuals who remain uncertain on their identity (reform winner or loser?) more pessimistic about their chances of benefiting from the reform. Consequently, learning considerations challenge the conventional wisdom that sequencing should be such that favorable reform outcomes are revealed first. Finally, we provide an explanation for why the gradual reform strategy worked well for China, while this is much less so for Latin American and Central and Eastern European countries.