“Market for Manipulable Information"

Abstract: We study how investors, firms, and information sellers interact in a market with manipulable information. Our model builds on the framework of Admati and Pfleiderer (1986) and introduces two new features: information manipulability and investor heterogeneity. In the baseline model where investors care about actual characteristics, the average degree of signal manipulability has no effect on the equilibrium, whereas the uncertainty about signal manipulability plays a key role. Its contribution depends on firms’ incentive to manipulate the signals that are used to generate the score, which in turn depends on the equilibrium price sensitivity to the score. The optimal design of the score in this setting weights the precision of different signals against the endogenous uncertainty from manipulation. The introduction of mandate investors, who care about the scores on the characteristics and not the characteristics themselves, generates a new incentive for information sellers to inflate the scores. Pushing too strongly on the mandate could lead to reduction in the informativeness of the score and the equilibrium price, and could even result in mandate investors holding less of the desired stocks.