One of the most controversial issues associated with restructuring the electric industry is the determination and recovery of stranded costs. Regulated electric rates are designed specifically to cover a utility’s cost of doing business, i.e., to recover its operating costs and invested capital and to provide an opportunity to earn a reasonable return on its capital. On the other hand, market-based prices are indifferent to the costs incurred by any individual market participant. Therefore, as electric markets are opened to competition the level of revenue earned by a utility may no longer closely match the level required to cover its costs. Some may earn more than their cost of doing business, others less. The difference between costs expected to be recovered under rate regulation and those recoverable in a competitive market is termed "stranded costs." If market prices are lower than regulated rates, as many expect, utilities could be faced with investments that are unrecoverable in the competitive market.

The issue of stranded costs has several facets. One is to clearly define what is meant by "stranded costs." The utility’s cost obligations must be identified and quantified. Although a company may have "strandable" costs, future market conditions will dictate whether these costs are unrecoverable. Therefore, expected future market revenues must be quantified, as well. Other considerations are: Whether stranded costs should be recovered? If so, should 100% of the stranded costs be recovered or only a portion? Who should pay for the stranded costs and what mechanism should be used for recovery? Has the utility company made a bona-fide effort to mitigate its stranded costs? and, Were the original investments and expenditures prudent?58

Definition of Stranded Costs

To address the issue of stranded costs the Commission set up a focus group which included representatives from IOUs, EMCs, MEAG, Industrial end-users, CUC and CPG. Each party’s views and definitions of stranded costs, the reasons for the existence of stranded costs and the appropriate recovery mechanisms differed.

No consensus was reached on a precise definition of stranded costs. A definition should be developed within the context of a Commission proceeding on this subject. Other states have already addressed this issue. For example, the Public Utility Commission of Texas acknowledges that "the concept of stranded investment has become confusing because of the number of definitions and interpretations." They also recognize that no costs are actually stranded until such time as customers begin to switch suppliers. As such, they define two concepts, "stranded investments" and "potentially strandable investments." The definitions made by Texas are as follows:

Stranded investment is defined as the historic financial obligations of utilities incurred in the regulated market that become unrecoverable in a competitive market. In the past, utility investments, i.e. "Financial Obligations," have been made in the regulated market, the market in which utilities "historically" operated. In that market, utilities anticipated that investment would be recovered in rates charged to customers. These obligations may become "unrecoverable in a competitive market" because prices in a competitive market are uncertain, and as such, may be below regulated prices. If a utility cannot charge as much in a competitive market as it would have charged in a regulated market, a portion of the asset becomes "unrecoverable" or "stranded." Thus the change from a regulated to a competitive market can create stranded investment.

The term "potentially strandable investment" is used because no investment is necessarily unrecoverable. The portion of potentially strandable investment that will ultimately become stranded is unknown. Costs may become stranded because the customer leaves a regulated utility for a market-based source of supply. However, the level of strandable costs is not dependent solely upon the customer’s behavior. Rather, the quantity of potentially strandable investment is also affected by market conditions. Even if the customer continues to buy from their local utility, the difference between the previous regulated price and the new market price will drive the amount of stranded investment.

Determination of Stranded Costs

Stranded costs can be determined in a variety of ways. One approach is to measure the stranded costs associated with each generating asset or purchase contract. Under this method the market value of the asset is compared to the fixed and operating costs of the asset. The determination of market value requires numerous assumptions and sophisticated modeling and therefore results may vary widely.

The market value is usually defined as the potential revenue a particular asset or investment can earn in a competitive market place. This value is calculated in terms of the price of electricity in the marketplace multiplied by the energy sold from that asset. Most models of stranded cost calculate this value over the life of the asset. If the revenues collected do not cover the historical costs associated with that asset, the difference is "stranded." The critical assumptions required in calculating the market value are those concerning the competitive market price. First, since a competitive market is not currently in existence, there are no real prices to evaluate. Instead, there are models that will, under simplified assumptions, calculate the market price for whole regions. However, these models are simplifications of a very complex system. As such, the assumptions used and the simplifications made could drastically alter the results. These uncertainties and simplifications have resulted in estimates that are of such large variance that they appear to be unusable for making decisions about how to allocate stranded costs.

Future or forecasted market price is defined as the price that can be expected to be charged for the output of the generating unit in a deregulated market. The future market price may be forecasted over an hourly, weekly, annual or other defined period of time. The final actual market prices will differ from forecasted prices. As with other competitive markets, sellers will sell their plants’ output at the highest price that they can receive. The seller is not expected to sell below the operating cost of the unit. Thus, a counterbalance of supply and demand is created at the marginal cost of the unit. That is, suppliers will continue to make electricity available to the market as long as they can earn more than their cost of production. The prevailing market price will determine if the unit is run or not run to meet the price and demand expectations of the market. The market clearing price is the price of the available generation that can be supplied against the expected demand created for that generation. Excess generation and low use will create low market prices while a shortage of generation and high use will create higher market prices.

Forecasted fossil fuel costs for coal, gas and oil will provide the competitive basis for existing generating units (including nuclear power), the economic viability of new units and the use of alternative fuels. Forecasted energy use will determine the extent to which current generation is used, the demand for new generating additions, the upgrading of existing units and the development of alternative power sources. In the future, the expansion of generation facilities will be dependent upon the market place. Estimates of the cost to install new generation as well as estimates of the operating cost of those units can be made without designating who will own the facilities. The choice of new generator type and the timing of construction will greatly impact the estimate of potentially strandable investments. A delay of a year or two, or a change from one generation type to another, will alter the market price and therefore change the stranded cost estimate. Because of the necessity to match on an instantaneous basis the supply or demand of electricity, an adequate supply of generation plus sufficient backup reserves will be required for reliability. The cost of this backup will be driven by the adequacy of reserves in the generating region, transmission constraints and the degree of any continuing obligation to provide service to all customers. These factors will form the basis of the forecasted market price.

Once determined market value is compared to the fixed and operating costs of the assets. These costs must be determined for each period of time over the life of the asset. In a purely regulated environment, these costs would be adjudicated in a regulatory setting in which all parties are allowed to present their points of view. The costs associated with an asset should include acquisition or fixed costs as well as the costs associated with the operation of the facility. In financial terms, this is the total of both the undepreciated assets remaining on the balance sheet as well the operating expenses. These costs must be established for each year until the asset has completed its used and useful life. The values assumed for costs will cause wide variations in the results of the strandable cost calculation. In those states that have attempted to calculate stranded costs, such as Texas, the assumptions for fuel cost have greatly affected the resulting strandable investment.

Given an estimate of both the market value and the cost profile of the assets, several methods may be utilized to determine stranded cost. One method of determining stranded costs is a net present value analysis. The revenues required to be collected over the remaining life of the generating asset to pay for the operating cost of the unit including debt cost are compared to the market revenues that may be charged for the output of the same unit over the remaining life. By comparing the projected costs, including debt cost, that must be recovered for each of the utility’s generating units to the projected market operating revenue that may be charged for each unit, it is possible to forecast which units are able to run and recover their full costs, and which ones are not. Thus, the net present value of costs of the generating asset which are in excess of the net present value of the forecasted market revenue become the stranded cost. This may be expressed as: NPV of the forecasted operating cost of the unit minus the NPV forecasted market price for the generating unit’s output.

The key to this approach, obviously, is making accurate forecasts of both future generation costs and the future market price of power. Errors in such forecasts could drastically affect the stranded cost estimates, either up or down. Another concern with this model is that some assets will not only cover their costs but earn revenues in excess of the operating and capital costs. These "negative" stranded costs should be netted against the "positive" stranded costs to form a more accurate estimate for the company as a whole.

To reduce the forecasting error of future market value and future operating costs inherent in this analysis other approaches could be used to determine stranded costs. One way to reduce, but not totally eliminate, forecast error would be to estimate market price by developing a range of prices based on a range of reasonable assumptions for all key variables affecting market price, and then negotiate or adopt a number within that range. Such a simulation would require modeling the region under different competitive assumptions. Another alternative would be to look at wholesale markets for some indication of what market prices may look like in competitive retail markets. Another approach is to use actual market prices once the competitive market is in place. A mechanism based on actually realized market prices could eliminate over-recovery of stranded costs and obviate the need to forecast market prices. A variation of this approach would be to make an initial forecast which would be trued-up over a period of time as the competitive market develops. This would be time-consuming and require extensive review and analysis each year. Another approach is the divestiture model. Market price would be based on the actual sales price of above market value assets or on an appraisal of what the assets would be worth if sold.

Given the uncertainty of estimating future market prices, regulatory agencies currently addressing stranded costs are developing alternatives to up-front estimates of stranded costs based on market price forecasts. For example, FERC, in its Order 888 provides an alternative to market price forecasts based on the actual contract price negotiated by departing customers. In California, while the issue is not fully settled, it appears that actual prices achieved by the power exchange will be used to calculate stranded costs, thus providing the opportunity to true-up forecasts over a period of time based on real market price data. In Pennsylvania, utilities have proposed market price estimates based on modeling forecasts and assumptions about all of the above variables. In Rhode Island, the legislature set a fixed amount per kWh for stranded cost recovery, based on a negotiated settlement.

Recovery of Stranded Costs

If costs will be stranded in a competitive market the level, method and timing of stranded cost recovery must be determined. One step in this examination of stranded costs is identifying who is responsible for the cost being stranded.

Some claim that stranded costs arise directly as a result of faulty investment decisions, e.g., plant technology or source of purchased power, by the utility management. They believe that the formation of a regulated monopoly was never meant to be an entitlement program. It was intended to provide the utility a return on capital, for the exchange of a fair price structure. It was never the intent of the regulatory compact to guarantee that all investments will be fully amortized in all future situations. According to others, the only reason we have stranded costs is that we can replace the existing system with one that lowers prices. The attempt to emulate competitive prices by regulating the monopoly has failed. Hence, they argue the stranded cost burden should be borne by shareholders.

Others, primarily the utilities, offer the "regulatory compact" argument. Briefly stated this argument purports that: As a result of the obligation to provide electric service invoked by regulators in exchange for monopoly status, electric utilities have installed generating facilities and committed to contracts with other suppliers to ensure safe and reliable service. In exchange for accepting the obligation to serve, the utilities were guaranteed recovery of these investments plus, in the case of investor-owned utilities, an opportunity to earn a fair rate of return. As the industry contemplates moving from a regulated market to a competitive market there exists a potential for the utilities to be unable to collect these historically incurred costs. Therefore, the utilities should not be penalized when the rules of the game change and stranded costs should not be borne by the shareholders.

Some suggest that the burden of stranded costs should be shared. Any sharing of stranded costs must be based on principle and consideration must be given to the adverse impacts on any group selected to bear a portion of these costs. The potential exists for rate shock to residential and small commercial customers in a restructured electric industry, even though competition should, theoretically, benefit all customers. Smaller customers, who may have less sophistication and less bargaining leverage than larger users, will likely have the most difficult transition to make. To burden these customers with a greater share of stranded costs than they deserve may make it impossible for them to receive affordable and reliable electric service in a restructured environment. The stranded cost issue needs to be resolved in a way that does not eliminate, or significantly delay, any benefits that residential and small commercial ratepayers may gain from competition. At the same time, rate shock must be avoided. This means that it could be acceptable for consumers to bear some stranded costs if doing so will help avoid rate shock.

If utilities do not recover their stranded costs, the shareholders will bear the burden. The demographics of investors and the effect that absorption of stranded costs will have on the utility's ability to attract sufficient capital to provide safe, reliable and affordable power must be considered. However, utilities should not recover imprudent investments or stranded costs that can be or could have been mitigated prior to the market being opened to competition.

It should be noted that two Congressional proposals for restructuring the electric industry include recovery of stranded costs:

The state regulatory authority may require, as a condition for the purchase by any person or municipality of retail electric energy services, the payment of a charge to recover costs incurred by an electric utility that become unrecoverable due to the availability of retail electric service choice. (U. S. Representative Dan Schaefer)

Electric Utility companies that prudently incurred costs pursuant to a regulatory structure that required them to provide electricity to consumers should not be penalized during the transition to competition. A retail electric energy provider shall be entitled to full recovery of its stranded costs, over a reasonable period of time, through a non-bypassable Stranded Cost Recovery Charge imposed on its distribution and retail transmission customers. No class of customers will be assessed a stranded cost recovery charge in excess of the class’s proportional responsibility. (U.S. Senator Dale Bumpers)

(1) Legislation creates a "non-bypassable" customer charge with a transferable guaranteed right to collect for the utility;

(2) State utility commission determines and authorizes the amount to securitize;

(3) Utility sells the right to collect the customer charge through a trust in the capital market; and,

(4) Utility receives the proceeds.

Securitization can save on capital carrying costs by replacing the utility’s higher cost debt and equity with the lower cost bond rate based on the legislatively guaranteed revenue stream. In order for securitization to achieve this savings, investors have to be given assurances that the payment stream is secure; that is, more secure than ordinary utility debt. Securitization would replace the annual revenue the utility would have received with a lump-sum, up-front payment of cash. The money could be used to: (1) buy back stock and retire debt in equal proportion or replace higher cost equity; (2) restructure power purchase contracts from non-utility generators and others; or, (3) invest in new power projects.

The cost-of-capital savings come from lowering the risk to investors by increasing customer risk. Several problems are inherent in securitization:

• Securitization is an irrevocable ratepayer commitment to pay the security in full;
  

• The forecast error, that is the risk of guessing wrong, when estimating the amount of stranded costs is placed on customers. Stranded cost estimates are highly sensitive to relatively small changes in assumptions. Many of these assumptions are highly speculative guesses about how things will turn out in a competitive market; and,
  

• Finally, securitization results in regulatory and market bypass. Customer protections from the regulation and monitoring of costs are removed. Supervision of these costs is placed beyond the oversight of the market and regulation.

At least one state has opted to use securitization, that is, California. Five others have laws that allow securitization. Another state has concluded that: "Exit fees and securitization are recovery mechanisms that should not be adopted because of their anti-competitive characteristics."60 Before seriously considering securitization for Georgia it must be determined whether securitization is an appropriate policy tool to apply to the "stranded cost" problem and whether securitization is harmful to the development of a competitive generation market. 

Conclusions

In sum, even though a utility may have potentially stranded costs, these costs may still be recoverable in a competitive market place. The determination of stranded costs must be done on a total company basis, not just for selected assets. Unless it is based on actual market prices, the estimate of stranded costs is subject to considerable forecast error. Stranded cost recovery guidelines should be adopted to ensure that the allocation of sharing of stranded costs is equitable and that the time period for recovery is appropriate. No group should achieve a windfall at the expense of another. The focus group on "Principles to Abide by in a Competitive Market" developed the following guidelines:

The State of Georgia should consider the recovery of legitimate, unmitigatable stranded costs . . . Recovery of stranded costs should consider prudence reviews and past regulatory actions . . . The length of time for any transition period to increased competition should consider what is necessary to recover stranded costs. . . The State of Georgia, rather than the Federal Government, should set the rules to determine the recovery from stranded costs and investments. . . There should be no cross-subsidy and over-recovery of stranded costs.