Ivo Welch

Ivo Welch appears in the book as one of the economists who first modeled information cascades. The model, developed with Sushil Bikhchandani and David Hirshleifer, explains how sequential choices can make other people’s actions look like valuable evidence, until individuals set aside their own information and the group converges on a wrong decision. Welch’s contribution is presented through this account of how a crowd can imitate rationally and still become uninformed.

The Cascade Model

The model begins with a choice between two new restaurants: an objectively better Indian restaurant and a worse Thai restaurant. Each person receives imperfect information about which is better; the information is useful but can be wrong. People therefore have reason to look not only at their own signal but also at what others choose, assuming that those others possess relevant information of their own.

How a Wrong Choice Spreads

The crucial condition is sequence. If the first diners happen by chance to receive misleading information and choose the Thai restaurant, later diners see a crowded restaurant and infer that it may be the better option. That inference can outweigh their own private signals, even when those signals favor the Indian restaurant. Once later choices rest on the observed choices of earlier diners, the group can follow the Thai restaurant despite its being objectively worse.

The mechanism is not simply thoughtless trend-following or a response to peer pressure. People imitate because they believe other people’s behavior contains information. That belief can make following the crowd reasonable for each person in turn, while the resulting pattern is wrong in aggregate.

Private Information and Collective Error

The model’s central danger is that, after enough people have made the same choice, a newcomer may rationally stop relying on private knowledge and follow the apparent consensus. But the cascade then ceases to gather the information individuals hold. Each person assumes earlier decisions reflect what those decision-makers knew; in reality, later choices may reflect only what people think earlier choices meant. Instead of combining private information, the sequence reproduces an inference about an inference.

The book distinguishes this account from an explanation that makes certain influential individuals central to the spread of new ideas. In the model associated with Welch, Bikhchandani, and Hirshleifer, early adopters matter because their decisions are seen first, not because they possess special influence or knowledge. Their model thus describes a cascade among people with equally useful private information, where timing alone gives early choices their power.