Is There a Limit to the Accuracy of Equity Valuation Using Multiples?

Abstract

We investigate the reasons for there being an apparent limit to the accuracy of valuation using multiples, when it is implemented in a standard way. We find that most of the error comes from failing to correct known biases in earnings forecasts. Another part comes from mismatching observable characteristics of target firms but the only characteristic that matters is growth. A smaller part comes from mismatching unobservable but persistent characteristics of the target firm. We develop a procedure to eliminate these errors, which reduces the variance of valuation errors by more than 60%. The remaining error is mainly random noise.

Keywords: Equity Valuation, Valuation, Multiples Valuation

JEL Classification: G11, G24, D81