Estimates of the cost of equity for public utilities, 1961–1976

Abstract

Using a simple version of the dividend cash flow (DCF) model of stock valuation, the cost of equity for public utilities is often inferred to be equal to the sum of the dividend yield and the expected rate of growth in dividends. Witnesses who employ this approach generally extrapolate past growth patterns into the future and then assume that investors expect these trends to continue; no effort is made to actually assess the expectations of investors. This approach to estimating the cost of equity for public utilities is criticized for the failure to develop testable hypotheses as an inferential basis for testing the statistical reliability of estimates of the cost of equity. This article demonstrates an alternative to the traditional approach, based on the premise that reliable estimates of the cost of equity are derived only within a methodological framework that produces testable hypotheses. The Gordon model of share valuation is formulated in such a way as to show that there is a systematic and predictable relationship between the ratio of market price to book value of common stock and a firm's normal or expected return on equity. This relationship suggests an econometric model that not only tests the Gordon model of share valuation but produces at the same time, inferences concerning the cost of equity. Using this approach, year-end estimates of the cost of equity for electric utilities are determined for the 16-yr period from 1961 to 1976.