When regulators grant changes to utility rates of return, they estimate growth on the basis of gross domestic product (GDP). But do utilities have any chance of growing at the same pace as GDP?...
Paradox of Thrift
Economic barriers complicate T&D modernization.
distribution system reinvestment levels (relative to depreciation) for 141 investor-owned electric utilities. The relationship to depreciation is an important one: a reinvestment level greater than one ( i.e., reinvestment / depreciation > 1.0) suggests that a utility eventually will raise rates (presuming no change in per-customer usage of existing customers) to economically recover the investment. Reinvestment levels less than one suggest a utility is substantially disinvesting in the network. In short, they are sweating the assets.
Note that reinvestment levels for all elements of the industry (first quartile, median and third quartile) declined throughout the 1988 to 1998 era (see Figure 1) . Regulatory uncertainty in this era resulting from transition-to-completion initiatives and rate freezes from merger activities led many utilities to invest more into unregulated assets and less into traditional T&D infrastructure. Industry reinvestment levels later were relatively flat and at historically low levels during the 1998 to 2003 era. During this time, investments were focused on back-to-basics strategies in the wake of the Enron and California debacles. Interestingly, the lowest (fourth) quartile of the industry was at or below a 1.0 rate in the early 2000s; in effect, utilities in this lowest quartile were disinvesting in their systems.
Since 2003, total reinvestment rates have risen industry-wide. This is consistent with the increasing rate-case activities in many jurisdictions. This has multiple causes—the rising cost of materials, the increasing attention to reliability investment, and the industry’s nascent modernization investments.
How Much to Reinvest?
This introduction will raise logical questions from experienced industry observers. Are replacement rates relatively high ( i.e., reinvestment/depreciation >1) or relatively low (say, less than 1) simply because the depreciation levels are inversely high or low?
Figure 2 presents a wide sample of the three-year average distribution replacement rates (replacement and reinforcement CAPEX/depreciation) for over 100 U.S. electric utilities. The scatter diagram shows that replacement rates are relatively uncorrelated to the level of depreciation (as measured by annual depreciation expense per customer). Figure 3 summarizes the data in Figure 2 and highlights this point further. It shows that the average reinvestment rate (as measured by the reinvestment/depreciation) is substantially identical for lowest, second, and third quartiles for utilities when they are categorized by depreciation per customer. The implication of this figure is fundamental but vital for corporate and regulatory policy makers—utilities reinvest at the level of capital resources available in their rates, not necessarily at the level of their system needs.
Although this may seem to be an obvious outcome, its documentation is an essential foundation and its implications are vivid. Rational electric utilities reinvest only at the level of the capital resources available to them in their rates. The amount of depreciation embedded in rates, and thus available to fund vital infrastructure reinvestment, should therefore be a topic of great concern for utility executives and regulators alike.
Understanding the level of depreciation in rates is central to determining the economic resources available to modernize the T&D network. The starting point is to recall that depreciation in a regulated utility doesn’t strictly follow normal accounting