Grossman-Stiglitz paradox

The Grossman-Stiglitz paradox argues that markets cannot be completely informationally efficient (see types of market efficiency). In particular, if markets would be completely efficient and people cannot earn an extra reward by actively managing their portfolio, then investors would stop their activities and new information would no longer be incorporated.

The paradox was proposed by Sanford Grossman and Joseph Stiglitz in a paper in 1980, a number of years after Eugene Fama proposed the efficient market hypothesis.

Grossman-Stiglitz paradox

On this page, we discuss the Grossman-Stiglitz paradox in more detail.

Grossman-Stiglitz paradox definition

The paradox starts from an important implication that follows from the efficient market hypothesis set out by Eugene Fama. In particular, this hypothesis states that markets are efficient and securities’ prices reflect all available information. Depending on the type of market efficiency, this can be information from past price data (weak), all public information (semi-strong), and even insider information (strong). At the same time, it is true that collecting and analyzing information about securities requires resources. Investment professionals need to be paid and spend resources gathering information and trading securities.

The paradox is then: Why do people spend time on researching securities? Also, if people don’t need to commit resources because markets are efficient, then how do prices reflect all pertinent information?

Stiglitz and Grossman argue that even when the market is in equilibrium, a return must be earned for information collection and analysis. Thus, the market must be inefficient to some extent if abnormal returns can be earned after information processing and transaction costs. Thus, active management can be profitable. This doesn’t mean it is easy, however, as most investment professionals have a hard time beating their benchmark.

Summary

We outlined the Grossman-Stiglitz paradox. The paradox argues that markets cannot be informationally efficient in equilibrium because new information needs to be reflected in security prices which requires effort for which market participants need to be compensated.