Strategic Experimentation with Private Payoffs

TL;DR

This paper analyzes a strategic experimentation game with private outcomes, showing that private information leads to more experimentation than public cases, often exceeding social optimality, due to a new encouragement effect.

Contribution

It introduces a novel encouragement effect mechanism and provides tight bounds on over-experimentation in private payoff settings.

Findings

Experimentation exceeds benchmark levels with private outcomes.

Pure equilibria often surpass social optimality.

A new encouragement effect explains increased experimentation.

Abstract

We study a strategic experimentation game with exponential bandits, in which experiment outcomes are private. The equilibrium amount of experimentation is always higher than in the benchmark case where experiment outcomes are publicly observed. In addition, for pure equilibria, the equilibrium amount of experimentation is at least socially optimal, and possibly higher. We provide a tight bound on the degree of over-experimentation. The analysis rests on a new form of encouragement effect, according to which a player may hide the absence of a success to encourage future experimentation by the other player, which incentivizes current experimentation.