Get access

A Generalized Econometric Model and Tests of a Signalling Hypothesis with Two Discrete Signals



    Search for more papers by this author
    • Graduate School of Business Administration, New York University. I am grateful to Milton Harris, Robert Korajczyk, Joe Williams, and an anonymous referee for their insightful comments. All remaining errors are, however, mine.


To test the major prediction of a signalling hypothesis-that the market price is monotonic in the signal-the price response to the signal must be measured. Since a signal is an outcome of a rational decision rule of the signaller, the market can infer the true type of the signaller from the signal. This necessitates estimation of the price response to the signal, conditional on the rational decision rule. Thus, the empirical models (e.g., event studies in corporate finance) that estimate the market price responses to signals without conditioning on the rational decision rules are misspecified if viewed as tests of the prediction of a signalling hypothesis. This paper builds a generalized econometric model with two possible discrete signals, derives the rational decision rules, presents a simple estimator of the price response to a signal, and illustrates its use in testing a recently expounded hypothesis that firms signal their true value by forcing or not forcing an outstanding convertible bond.

Get access to the full text of this article