Results
Shortcomings

COMPARABLE EARNINGS METHOD
FOR CALCULATION OF
UTILITY COST OF EQUITY

Description

The Comparable Earnings approach to cost of equity determination is based on the landmark Hope (1944) utility regulation case. The Hope decision stated that a utility's allowed ROE should be commensurate with returns which can be earned on investments in other firms having corresponding risks. The implication is that all one needs to do to determine what returns can be earned by other firms is to look at what returns actually have been earned by other firms over some suitable period in the past. The problem is how to select the appropriate other firms and historical period to look at without appearing to be arbitrary.

What are "other firms having corresponding risks"?

These could be either other utilities or non-utilities that are somehow comparable to the utility in question. However, because this model is based on accounting data (rather than market data), which can be distorted by regulation, it is more important with this model that non-utilities be considered as part of the comparable earnings analysis. Typical screening criteria might be as follows:

Sceening Criteria for Other Utilities

  • Primary country of business (US)
  • Beta
  • Debt as a % of total capital
  • Total assets
  • Electric revenues as % of total revenues
  • Annual construction expenditures
  • Nuclear generation
  • Financial strength rating (a Value Line measure)
  • Senior debt credit rating

Non-Utility Screening Criteria

  • Primary country of business (US)
  • SIC Industry code
  • Capital spending
  • Beta
  • Debt as a % of total capital
  • Total assets
  • Financial strength
  • Senior debt credit rating

In evaluating which screening criteria to use, several questions should be answered with respect to each one.

  1. Is there any theoretical reason why a relationship should exist between the criterium and earned ROE? For example, financial theory as represented in the CAPM model says that Beta should relate to expected returns, and therefore might also relate to actual returns.
  2. Can such a relationship be demonstrated to actually exist? One might speculate that companies with greater assets have greater market power and therefore can earn higher returns, but actual data might not support that conclusion.
  3. Does use of the criterium create a similarity that is intuitively meaningful? By using criteria such as SIC codes and selecting industries that have something in common with electric utilities (e.g. diversified natural gas or petrolium companies because they are energy related), you avoid having to explain how a newspaper publisher or automobile manufacturer should be considered "comparable" just because they are the same size and have the same Beta.

When applying the comparable earnings model, it may be instructive to compare the market-to-book ratios of utility stocks to those of industrial stocks. One should remember, however, that in the current environment the book value of some utilities may be overstated due to stranded investment that regulators have allowed to remain on the books.

During periods when most stocks are trading at market-to-book ratios significantly above one, the comparable earnings model applied to comparable industrials often will yield higher equity cost results compared to either DCF or risk premium models.

paulconsul@aol.com