Estimating Industry Multiples

Format:Print
| Language:English

Abstract

We analyze industry multiples for the S&P 500 in 1995. We use Gibbs sampling to estimate simultaneously the error specification and small sample minimum variance multiples for 22 industries. In addition, we consider the performance of four common multiples: the simple mean, the harmonic mean, the value-weighted mean, and the median. The harmonic mean is a close approximation to the Gibbs minimum variance estimates. Finally, we show that EBITDA is a better single basis of substitutability than EBIT or revenue in the industries that we examine.