All regulatory mechanisms should be evaluated with regard to the various statutory objectives. Efficiency properties can be thought of as technical, allocative, or dynamic in character. Technical efficiency is served by encouraging the minimum cost production of any given bundle of goods and services. Allocative efficiency is served by setting prices that encourage production of a bundle of goods and services that is optimal in terms of both quantity and quality and distributes those goods and services to those customers who value them most highly. Dynamic efficiency is served by encouraging process and product innovation.

All regulatory mechanisms or approaches pose tradeoffs among these statutory objectives. For example, regulatory mechanisms that create incentives to minimize production costs may also create incentives to reduce service quality. Mechanisms that promote allocative efficiency through some form of price discrimination may also be viewed as unfair. Mechanisms that require sharing of the gains from innovation between utilities and consumers may impair the incentives of utilities to innovate in the first place.⁽⁴⁾

In most of the markets under its regulatory jurisdiction, the FERC currently uses cost-of-service regulation. This form of regulation has the advantage of producing prices that are just and reasonable in the sense that the revenues that a utility is allowed to recover are close to its accounting costs. However, cost-of-service regulation may do little to promote, and may actually inhibit the achievement of, technical, allocative, or dynamic efficiency.

Under cost-of-service regulation, prices fail to reflect marginal cost. Historical accounting costs, on which prices are based, may bear little relation to future replacement costs. In addition, utilities' common costs are often allocated to their different services in an arbitrary way under cost-of-service regulation. Because this results in prices in excess of incremental cost, inefficient entry may be encouraged. In addition, because prices are based on firms' accounting costs, cost-of-service regulation may provide utilities with incentives to cross-subsidize by shifting costs from their unregulated services to those that are regulated. On the other hand, in some cases utility prices are set too low because they do not reflect the opportunity cost of reallocating the scarce good to another consumer who values it more highly. These prices fail to clear the market and cause services to be distributed among consumers in an inefficient way, such as rationing.

Under cost-of-service regulation, utilities' incentives for technical and dynamic efficiency are blunted because cost and profit changes are automatically passed on to customers through changes in allowed prices. Utilities have reduced incentive to minimize costs, innovate or expand product variety because they receive little or no long-term benefit from such efforts. Cost-of-service regulation may also give utilities incentives to over- or undercapitalize, depending upon whether the allowed rate of return is above or below the market cost of capital.

In assessing whether some other form of regulation would produce a better outcome than cost-of-service regulation, the FERC must carefully consider the characteristics and problems of the specific industry being regulated. For the industries under the FERC's jurisdiction, some of the more commonly cited problems with cost-of-service regulation may not be very severe, or, if they are severe, may not be solved by a change to some forms of incentive regulation. Other forms of incentive regulation, however, may have the potential to create significant public benefits in the industries under the FERC's jurisdiction.

FERC jurisdiction extends to natural gas pipelines, oil pipelines, and wholesale electricity generators. In t hese industries, sunk capital costs are large portion of total costs. Since utilities have no control over such costs once incurred, various alternatives to cost-of-service regulation cannot correct for any historic over- or undercapitalization, nor can they reduce utilities' current capital costs. However, various alternatives to cost-of-service regulation could remove future incentives to overcapitalize.

Utilities have some control over their operation and maintenance expenses, but these represent a small portion of annual costs. Also, the industries under the FERC's jurisdiction do not seem to present as substantial opportunities for technological innovation as do other industries. Consequently, the potential gains from moving to a regulatory scheme that provides better incentives for cost minimization and innovation may be more limited than some other regulated industries.

On the other hand, cost-of-service regulation may have a detrimental effect on allocative efficiency in the industries under the FERC's jurisdiction, and there may be a significant benefit to adopting a regulatory scheme that produces more efficient pricing. In addition, there may be opportunities for market growth in some areas, so there may be potential benefit from improving utilities' incentives to expand service.

The NOPPS focuses nearly exclusively on some form of price cap rate regulation as the primary or indeed the only alternative to cost-of-service rate regulation. Such a perspective, in our view, is too limited. In general, regulators should specify which of the problems with current regulation it hopes to solve and consider the full array of regulatory approaches to restraining market power designed to improve upon the existing regulatory mechanism.

One family of regulatory alternatives to cost-of-service rate regulation is incentive rate regulation, which includes price-cap regulation as well as other mechanisms. Incentive rate regulation specifies allowable rates but differs from cost-of-service rate regulation by severing or weakening the connection between a utility's rates and its own costs. The various incentive rate mechanisms are described in the following service.

Another family of regulatory alternatives to cost-of-service rate regulation relies on rules that govern behavior rather than specification of allowable rates. One important example of a rule-based approach are rules that govern the formation and operation of so-called competitive joint ventures. The relevant rules preserve an independence of action among joint venture participants with respect to initial capacity decisions and subsequent production decisions. In this manner, a measure of competition is created while retaining the economies of scale associated with joint venture assets.

Another example of a rule-based approach is regulation that allows two-part tariffs or other volume-dependent price schedules. Such nonlinear pricing may permit a firm to exercise market power while preserving long- and short-run incentives to provide optimal quantities. Of course, nonlinear pricing may provide lower unit prices to large customers than to others. This may be deemed to be unfair, particularly when different-sized customers compete with each other.

3. A SIMPLE PRICE-CAP MECHANISM IS NOT NECESSARILY THE BEST FORM OF INCENTIVE RATE MECHANISM

Incentive rate mechanisms weaken or sever the connection between utilities' future costs and prices in order to provide these utilities with incentives to minimize their costs and, under some mechanisms, to price more efficiently. There are three basic types of incentive rate mechanisms -- price-cap regulation, yardstick competition, and stand-alone cost caps. Under price caps, a utility's initial rates are set and subsequently adjusted by means of a price index based on exogenous inflation and, in some cases, an exogenous productivity adjustment. Under yardstick competition, a utility's allowed prices are bsaed on the costs of other companies in the industry. Under stand-alone cost caps, a utility may charge any prices that do not recover revenues that exceed the entry costs of a new competitor. The FERC should consider all three incentive rate mechanisms, rather than focusing solely upon price caps.

Each of these mechanisms can be paired with other supplementary mechanisms to increase pricing flexibility and regulatory fairness. Tariff menus allow the utility to offer a set of tariffs to customers. Selective discounting permits a utility to negotiate below-tariff rates with individual customers. Profit-sharing returns a fraction of a utility's excess earnings to its customers.

The statutes under which the FERC regulates pipelines and electric utilities require that all rates and charges be "just and reasonable." The purpose of this requirement is to protect consumers from exploitation and to ensure the efficient provision of adequate service. Just and reasonable rates have therefore been defined as rates that are high enough to maintain and attract capital and low enough to protect consumers against the market power of the regulated firm, see, e.g., City of Chicago v. FPC, 458 F.2d 750-51 (D.C. Cir. 1971). The starting point for ratemaking has traditionally been the regulated firm's cost of service. The statutes, however, do not prescribe cost-of-service ratemaking or any other methodology. FPC v. Hope Natural Gas, 320 U.S. 591, 602 (1944).

Even where the market power of the regulated firm makes reliance on competition inappropriate, the FERC has the flexibility to approve rates that depart from cost-of-service pricing in order to give regulated firms the incentive to reduce costs and operate more efficiently. See Permian Basin Area Rate Cases, 390 U.S. 747 (1968) (approving incentive pricing aimed at stimulating natural gas exploration). In Permian Basin, the Court held that the FERC may consider long-run consumer interests in setting rates. Id. At 798. See also Mobil Oil Exploration v. United Distribution Co., 111 S.Ct. 615 (1991). Thus, the FERC may approve rates that depart from cost of service rates in order to encourage firms to take actions (e.g., lower costs or efficiently expand capacity) that will benefit consumers in the long run through lower rates or better service.

Under price-cap regulation, a utility's prices are constrained below a ceiling set by the regulator. The price caps are adjusted periodically by a benchmark measure of cost inflation and, in some cases, a measure of expected productivity change in the industry. Each service's price can be capped individually or can be combined with other services into an index basket. Under the latter approach, there is an overall constraint on the allowed revenues from services in the basket, but there are no constraints on individual price changes.

By weakening the connection between a utility's costs and its allowed prices, price caps provide utilities with incentives to minimize costs. Price caps can also provide utilities with incentives to engage in greater technological innovation, because they allow utilities to keep the profits associated with that innovation over a longer term. Because of its impact on incentives, price-cap regulation would be most beneficial in those industries where there are opportunities for significant reductions in operating costs, or opportunities for technological change. The industries under the FERC's jurisdiction would appear to present only modest opportunities for such benefits.

Price caps also reduce incentives for a utility to engage in cross-subsidization between its regulated and unregulated services. Because the utility's allowed rates are not based on its accounting costs, it has no incentive to shift costs from one sector to another. For this reason, price-cap regulation may be particularly well suited to those industries where utilities are participating in both regulated and unregulated markets or where partial deregulation is contemplated in the future.

Price caps can also reduce the direct costs of regulation if the exogenous cost adjustments chosen by the regulator provide reasonably good proxies for changes in the utility's costs. Therefore, price caps will be a good choice of regulatory mechanism to the extent that such proxies exist.

One potential problem with price-cap regulation is that it may induce utilities to reduce service quality in order to reduce costs and raise profits. A utility might cut back on maintenance expenses, resulting in lower system reliability. The FERC should consider whether such a strategy is possible in the industries subject to its jurisdiction and whether it is able effectively to monitor service quality.

In contrast to price-cap adjustments which adjust a utility's service rates using an exogenous rate of inflation, under yardstick competition the utility's prices are adjusted by an index that is based on the actual costs of similar utilities. These other utilities' costs serve as a "yardstick" to determine what is a fair cost pass-through. Like price-cap regulation, yardstick regulation breaks the link between a utility's own costs and its allowed prices, giving utilities an incentive to keep these costs down.

The primary difficulty with such a yardstick mechanism is deciding which other utilities to place in a particular utility's yardstick basket. If the costs of the utilities in the basket are not sufficiently similar to those of the utility subject to yardstick regulation, or those costs are themselves distorted by regulation, then yardstick regulation might lock that utility into unreasonable or unjust losses or gains. When considering whether to adopt yardstick regulation, the FERC ought to consider whether it is possible to construct reasonable yardstick baskets in a particular industry.

Another type of incentive rate mechanism sets price caps based on stand-alone costs. Stand-alone costs for one or a group of utility services represent those costs that a competitive entrant would need to undertake in order to provide the service(s) in question. Under a stand-alone cost test, regulated utilities would be allowed to set any prices so long as their recovered revenues for every possible subset of services, as well as their total bundle of services, do not exceed associated stand-alone costs. Under such a standard, a utility will act as it would in a competitive market. The Interstate Commerce Commission has adopted stand-alone costs as a guideline for determining whether rail rates for determining whether r ail rates for coal shipment are reasonable. Coal Rate Guidelines, Nationwide 1 I.C.C.2d 520 (1985).

Use of stand-alone cost price caps, however, creates two main concerns. First, stand-alone costs are a hypothetical construct and therefore cannot be directly observed; their estimation could substantially increase regulatory workload. Second, utility accounting profits under stand-alone pricing may be excessive if current costs of entry exceed the utility's embedded costs, and if pricing based on replacement cost cannot be supported as furthering the statutory goals.

In order to promote greater pricing efficiency under incentive rate regulation, utilities operating under such regulation can be allowed to offer tariff menus or selective discounts. A tariff menu is a list of tariffs from which a customer may choose a particular price and quantity pair. In order to ensure fairness under such a scheme, the regulator can regulate the level of one tariff on the lists, so that customers always have recourse to a rate deemed to be "just and reasonable." Any alternative that a utility voluntarily offers and a customer voluntarily selects will benefit both parties.

Selective discounts are bilateral agreements between a utility and specific customers; the aim often is to retain a customer who would otherwise leave. Such discounts are not generally available to other customers. So long as all customers have recourse to a fair tariff, the outcome of these negotiations can be deemed fair.

If utilities with substantial market power are able to reduce costs and engage in technological innovation, their resulting profits may be high. Under profit-sharing a fraction of these profits are given back to consumers by restricting the utility's allowed price increase in the subsequent regulatory period. While providing some benefits to consumers, profit-sharing also reintroduced many of the problems associated with cost-of-service regulation. Utilities subject to profit-sharing may have incentives to engage in cross-subsidization to reduce their accounting profits. In addition, measuring profits requires regulators to engage in the difficult conceptual exercise of measuring depreciation, as they do under cost-of-service regulation.

4. THE PROPOSED IMPLEMENTATION OF INCENTIVE RATEMAKING MAY WARRANT FURTHER CONSIDERATION

The FERC apparently intends to implement incentive ratemaking in a very cautious manner. First, regulated firms may submit incentive rate proposals on a voluntary basis but will not be required to do so. Second, rather than specifying industry-wide guidelines in advance, the FERC will allow regulated firms to design their own approaches to incentive ratemaking and will evaluate them on a case-by-case basis. Third, incentive rate proposals must meet a burden of proof which includes requirements that they be understandable by all parties and that consumer benefits relative to cost-of-service ratemaking be quantified. NOPPS, pp. 11-12. The voluntary nature, the case-by-case nature, and the burden of proof associated with the FERC's proposed implementation of incentive ratemaking each merit careful scrutiny.

The FERC's ratemaking authority permits it to review the rates established by regulated firms to determine whether those rates are just and reasonable. See, e.g., East Tennessee Natural Gas Co. v. FERC, 863 F2d 932, 941 (D.C. Cir. 1988). Since the FERC views the implementation of incentive rate regulation as somewhat experimental, it intends to proceed on a voluntary basis.

While we understand FERC's caution, it should be aware that a voluntary movement to incentive ratemaking could be problematic. Plainly, regulated firms will submit incentive rate proposals voluntarily only when it is in their own best interest. This creates two possible problems. First, inefficient firms can opt for the ongoing comfort of cost-of-service methodology, which means that incentive ratemaking may be eschewed precisely where its cost-minimization properties would be of greatest value. Second, regulated firms will opt for incentive methodologies that will earn them greater profits, even if an alternative methodology would create greater overall economic benefit, but potentially lower utility profits. Thus, a voluntary approach may reduce the benefits of a broader move to incentive ratemaking.

The FERC has the discretion to develop explicit industry-wide guidelines on what it considers to be appropriate incentive rate mechanisms for each industry under its jurisdiction. However, the FERC apparently prefers a case-by-case approach in which r egulated firms will present a wide variety of incentive proposals of their own design. This can be viewed as a process of cautious experimentation. To the extent that optimal design of incentive ratemaking varies from firm to firm within an industry, superior idiosyncratic solutions may be found. On the other hand, the costs of an idiosyncratic approach could exceed those of a more generic approach that prescribes a specific methodology throughout an industry, without providing sufficiently larger benefits. Because a more generic approach would avoid numerous contentious proceedings, the case-by-case approach to design of incentive ratemaking may create an unwarranted administrative burden. In addition, if the FERC were to specify specific types of incentive proposals that it would be willing to seriously consider, it would reduce utilities' uncertainty and would increase the likelihood that the regulatory schemes proposed would benefit consumers as well as utilities. Thus, the merits of a pure case-by-case approach to design of incentive ratemaking within an industry are subject to serious question.

The FERC anticipates that all incentive rate proposals must satisfy the twin requirements that they be understandable by all parties and that their benefits relative to cost-of-service ratemaking be quantified. This burden of proof favors the retention of cost-of-service rate regulation and may impede adoption of socially desirable incentive rate programs.

In order to support an incentive rate design, the evidence must show that the rate mechanism proposed is likely to effect the stated statutory goal, and that the increase in rates is "no more than is needed for the purpose." City of Detroit v. FPC, 230 F.2d 810, 817 (D.C. Cir. 1955); see also City of Charlottesville v. FERC, 661 F.2d 945, 949-951 (D.C. Cir. 1981); Public Service Commission v. FPC, 487 F.2d 1043, 1071-72 (D.C. Cir. 1973); City of Chicago v. FPC, 458 F. 2d 731, 749-752 (D.C. Cir. 1971). In the area rate cases, for example, higher rates for new gas dedicated to the interstate market were approved where the incentive rate was clearly linked to the goal of generating new gas supplies. Permian Basin at 797. In City of Charlottesville, by contrast, the court disapproved the inclusion of stand-alone tax costs in the rate base where there was no showing that the increased returns that this allowed the pipeline would result in increased exploration and development. 661 F.2d at 953-54. Thus, it must be shown that an incentive mechanism aimed at, for example, encouraging firms to reduce costs is likely to induce the regulated firm to undertake cost saving measures rather than merely allowing an increased rate of return.

Although the FERC should be mindful of the burdens implied in the case law, the FERC should not require precise quantification of benefits before allowing use of an incentive rate design. Cf. Permian Basin, 390 U.S. at 814 (Commission need not make findings in terms of absolute dollar amounts, but need only proffer "findings and conclusions sufficiently detailed to permit reasoned evaluation of the purposes and implications of its order.").

Respectfully submitted,

James F. Rill
Assistant Attorney General
Antitrust Division

Janusz A. Ordover Deputy Assistant Attorney General Antitrust Division _______________"/s/"_______________ Mary E. Fitzpatrick Assistant Chief Economic Regulatory Section _______________"/s/"_______________ John R. Sawyer Economist Competition Policy Section _______________"/s/"_______________ Michael A. Einhorn Economist Economic Regulatory Section

J. Mark Gidley Deputy Assistant Attorney General Antitrust Division _______________"/s/"_______________ Mark C. Schechter Chief Transportation, Energy and   Agriculture Section _______________"/s/"_______________ Roger W. Fones Assistant Chief Transportation, Energy and   Agriculture Section _______________"/s/"_______________ Michael D. Billiel Attorney Transportation, Energy and   Agriculture Section Antitrust Division U.S. Department of Justice Room 9104 Judiciary Center Bldg. 555 Fourth Street, N.W. Washington, D.C. 20001 (202) 307-6349

April 27, 1992

FOOTNOTES

1. Farmers Union Central Exchange, Inc. v. FERC, 734 F.2d 1486, 1510 (D.C. Cir.), cert. denied, 469 U.S. 1034 (1984). In that case, the court rejected oil pipeline regulations that imposed price caps set only to prevent "egregious price exploitation," because the FERC had failed to show that the resulting rates would satisfy the just and reasonable standard. Id. at 1509-10. The court rejected the FERC's use of anedoctal evidence, id. at 1508 n.50, but indicated that "empirical proof" that the actual rates would be just and reasonable could satisfy FERC's burden. Id. At 1510.

2. The exercise of market power may result in several types of inefficiencies. Allocative inefficiency arises when, as a result an exercise of market power, prices exceed marginal cost. Then, the value of additional output to society does not equal its cost of production, and an incorrect amount of the good or service is produced and consumed. Technical inefficiency arises when production and other costs are not minimized. A firm with market power is under less pressure to minimize the costs of producing a given quantity of output. Dynamic inefficiency arises when market power dulls the incentive for product and process innovation.

3. For a proper evaluation of market power, the prospects for ex ante competition, ex post competition, and potential entry must all be considered. Ex ante competition can occur in the context of proposed new service when potential suppliers compete for the right to serve before investments are sunk. The benefits of such ex ante competition might be conveyed into the ex post environment by long-term contracts without the need for traditional regulation

Ex post competition can occur among incumbent suppliers once investments are sunk. Such competition can be intramodal (the same product offered by multiple suppliers) or intermodal (substitute products offered by multiple suppliers) in nature. In general, ex post competition is assessed by defining relevant markets, evaluating structural conditions therein, and determining whether market conditions are conducive to the exercise of market power. See, e.g., Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines (1992).

Potential entry may discipline incumbent suppliers. Since industries under the FERC jurisdiction are generally characterized by sunk costs and economies of scale, neither "hit and run" entry nor small scale, nondisruptive entry is often plausible. However, potential entrants might still threaten incumbents by offering lower prices by means of long-term contracts, thereby protecting themselves from unprofitable pricing after entry has occurred.

4. The tradeoffs posed by cost-of-service rate regulation and incentive rate regulation are of special interest and are discussed at greater length below.