Valuation of capitalized interest on construction expenditures in the electric utility industry

The most common technique for compensating funds devoted to construction projects in the electric utility industry has been surrounded in controversy in recent years. The purpose of this dissertation is to analytically and empirically investigate how this technique, Allowance for Funds Used during Construction (AFC), affects the valuation of electric utility shares.The first chapter sketches the background of the controversy. Discussion by industry observers leads to major testable hypotheses. In Chapter 2 the value of an electric utility share is analytically derived using AFC and the most frequently mentioned alternative, direct inclusion of Construction Work in Progress (CWIP) in the rate base. Necessary and sufficient conditions for indifference between the two alternatives are demonstrated under certainty. Introducing uncertainty, it is shown that operating earnings and construction (AFC) earnings could be valued differently in the marketplace. Finally, it is argued that investors could continue to be indifferent between the two compensation techniques if the regulatory process provides rates which are adequately adjusted for risk.Chapter 3 outlines a theroetical valuation model developed by Litzenberger and Rao [1971]. It is an eclectic combination of the SharpeLinter capital asset pricing model and the Miller-Modigliani model for growth [1966]. Its unique feature is allowance for intra-industry differences in risk.In Chapter 4, two modifications are made to the theoretical valuation model. First, the earnings stream is split into two components — AFC and Operating Income (01). Second, the model is reformulated to incorporate systematic risk.Chapter 5 describes the source of the data and the transformations used in the empirical tests. Descriptive statistics are also provided.Chapter 6 presents the empirical results. The evidence is consistent with AFC credits being interpreted by the market as being of "lower quality" than operating income. Considerable attention is devoted to improving the specification of the valuation model.Chapter 7 enumerates the assumptions of the ordinary least squares technique and tests a representative model for violations.Chapter 8 summarizes the findings of the dissertation.Early in the history of electricity and gas service, an issue of equity arose over who should pay for expansion — current consumers or future consumers. Since most expansion was into new territories and current consumers received no direct benefits, the prevailing logic absolved current consumers from subsidizing construction expenditures. To achieve this goal CWIP was excluded from the rate base and thus current prices reflected only plant in service. Still, dollars devoted to construction had to be compensated. This was achieved by capitalizing interest. An interest factor was applied to the balance in CWIP and this amount was debited to the asset account. The credit side of the entry caused an increase in a nonoperating income account, Allowance for Funds Used during Construction (AFC).'In recent years, these credits have become an increasingly significant portion of reported income for many utilities (see Table 1-1). The growth in this phenomenon is due to two factors — increased construction budgets and longer duration of the projects. Both of these can be attributed in part to the construction of nuclear power plants in the late 60's and early 70's.