The year-long decline in the electric utility stock market has caught most market observers off guard. Picking the winners among electrics has become more difficult. Says Ed Tirello, long-time market savant and utility equity analyst at NatWest Securities, "Competition and retail wheeling have made the selection process nearly impossible short term."

To identify tomorrow's best industry performers, electric utility analysts have focused on generation. They believe low-cost producers with low industrial sales will do better than high-cost producers with large industrial sales within a region. Difficult to argue. Nevertheless, that's only half the story; the other half involves cost of capital and expected returns on functional assets. That's the half we address here. Looking at generation, transmission, and distribution as separate elements of corporate finance, we find that most electric utilities should quit the generating business.

Beta Phobia

Any article that puts the word "beta" on the first page will likely, and appropriately, be discarded. Stay with this one. The pain is short. The theoretical after-tax cost of equity (COE) capital for the electric utility industry currently runs at 11.1 percent, if you use the Capital Asset Pricing Model and, by inference, equity betas. So what? Well, theory holds that a stock price can only advance when a company's return on equity (ROE) exceeds its COE. Thus, if beta and COE are still relevant to stock price performance and worthy of our attention, we should be able to construct a predictive model. Graph 1 does just that. Market/ book ratios have been regressed against the quotient of ROE/COE. The graph looks good. Or, in less understandable parlance, the correlation coefficient is 63 percent and the R2 is 40 percent, which suggests that a company's ability to exceed its theoretical COE will determine, in large part, whether its stock price will rise or drop. So for now, let's stick with betas and COE.

The relevant question is this: Do generating assets carry a higher COE than distribution or transmission assets? If so, does the COE associated with generating assets exceed the allowed ROE, or can a utility with an allowed equity return of 11 percent focus on a business with a higher COE and hope for stock price advancement? Probably not. And yet that is just what a growing segment of the industry is doing (em focusing on the higher potential returns generated in the generation business. But how high do those returns have to be in order to lift a utility's stock price? Certainly higher than the theoretical COE, which we will now attempt to measure.

Measuring Functional Risk

We have divided the industry into three segments with high, medium, and low concentrations of generating assets (percentage of utility assets devoted to generating sector). Intuition might suggest that the segment with the highest concentration would post the highest COE. We have analyzed the utilities in each of the three segments and have removed leverage from their equity betas to establish asset betas (since we are focusing on asset classes). Then we calculated medians for each group (see Table 1).

The asset betas for all three segments are nearly identical,

Table 1

Electric Generating Segment

Risk vs. Concentration of Generating Assets

Median Asset Beta

High Concentration .31

Medium Concentration .29

Low Concentration .31

suggesting that the market does not currently differentiate between the risk characteristics of companies with 70 percent of their assets in generation and those with 70 percent of their assets in distribution. Going forward, the market probably should, or it could get caught again. More important, utilities should employ different asset betas in making capital investment decisions. The trouble is, we cannot directly measure the asset betas, or risk, of the functional utility segments. However, an indirect measurement will point you in the right direction.

Market Proxies

We have calculated betas over time for the three industries that arguably can serve as proxies for the three electric utility functions: 1) independent power (for generation), 2) gas pipelines (for electric

Table 2

Industry Segment Betas

Market Proxies (Medians)

Asset Equity Implied

Industry Beta Beta COE (%)

Independent Power 0.96 1.23 16.9

Gas Transmission 0.44 0.59 12.3

Gas Distribution 0.26 0.37 10.5

transmission), and 3) gas distribution (for electric distribution). Graph 2 shows the betas for these proxy industries over the recent past. The data is instructive. The independent power industry (IPPs), without a franchise, emerges as the riskiest. More interesting, IPPs have become riskier with the sunset of state-mandated contracts. The gas pipeline industry, with its various diversification schemes and vertical integration, was exceedingly risky 15 years ago. However, as that industry has begun serving more of a merchant function, the associated risk has fallen. The gas distribution industry (LDCs) operated in a relatively low-risk environment 15 years ago. Today LDCs carry even less risk, having survived Federal Energy Regulatory Commission Order 636 without much pain. (It is worth noting that a number of studies actually predict an increase in the theoretical COE for the gas distribution industry, since it can no longer count on pipelines as exclusive providers of gas. To date, however, this prediction has not come to pass.)

Table 2 shows the current asset betas and the resultant equity betas and COE for each of these proxy industries.

Implications

Taken at face value, this data implies that an electric utility with an 11 percent allowed ROE should not invest in generation, but in distribution. The flipside is even more difficult to stomach; only those companies with ROEs well above the industry average should invest in generation. An exception might exist for an electric utility that holds a very unique fuel or portfolio advantage, which it then can direct to

its generation business and can retain for its shareholders.

The stark reality is sobering: Many an electric company should exit some aspect of the generation business, unless allowed ROEs go up or that company possesses a unique skill set that can help it cut the COE.

But do we conclude that most utilities should simply abandon

a large component of rate base and employee headcount? No, although various intermediate steps can lower a utility's cost of service and COE. In sequence by ease of execution, a utility might consider a competitive bid for or negotiated sale of 1) its entire power purchase portfolio, followed by 2) its generating plant operations and maintenance function, and ending with 3) the outright sale of generating plant. The first to let go will reap the great- est value; the last the least. And those who don't may see their stock languish over the next few years.

In fact, future valuation is rarely this simple. Unforeseen factors will alter this predictive model. Suffice to say, however, that a significant portion of the industry is leaning toward generation as a stock price panacea at a time when true value may be produced by exiting generation. t

David P. Wagener is director of the Global Power Group at Salomon Brothers Inc., New York, NY.

Articles found on this page are available to Internet subscribers only. For more information about obtaining a username and password, please call our Customer Service Department at 1-800-368-5001.