LexisNexis(TM) = Academic

1996

17 Energy L. J. 401=20

LENGTH: 15291 words

ARTICLE: MARKET = POWER IN=20 ELECTRIC UTILITY MERGERS: ACCESS, ENERGY, AND THE = GUIDELINES=20

by Robert J. Michaels *

* Professor of Economics = at=20 California State University, Fullerton; A.B. University of Chicago 1965; = Ph.D.=20 University of California, Los Angeles, 1972. The author is also = Consultant in=20 Economics and Finance, JurEcon, Inc., Los Angeles; and Adjunct Scholar = at the=20 Cato Institute. Earlier versions of this article were presented at = Rutgers=20 University's Advanced Workshop in Regulation and Public Utility = Economics, San=20 Diego, July 12, 1996; and at The Energy Daily's = Conference on=20 Utility Mergers and Acquisitions, Washington D.C., June 18, 1996.=20

SUMMARY:
... In electric utility = mergers the=20 antitrust agencies (DOJ and the Federal Trade Commission) defer to the = opinion=20 of the Federal Energy Regulatory Commission (FERC). ... = Instead, the government was content to state that the company had = "strategic=20 dominance." ... If those proceedings produce efficient reservation and = pricing=20 schemes, open access will have brought about the = competitive=20 market for delivery that is the necessary complement of competitive = markets for=20 energy and capacity. ... Open access = may=20 nullify monopoly power in transmission, but the FERC believes it = possible that=20 an open access utility can have monopoly power over = capacity or=20 energy. The FERC now presumes that open = access=20 utilities requesting market-based energy prices have no = monopoly power in new generation. ... Cost-benefit analysis is an = exercise at=20 the margin: If a merger has passed the open access = screen, does=20 further study of energy or capacity increase the = likelihood=20 that the FERC will reach a correct decision? That research will be = beneficial if=20 it provides information that the open access analysis = does not,=20 and if the accuracy and relevance of this information reduces the FERC's = uncertainty. ... The Commission has yet to produce publicly available = studies of=20 such basics as the relationship between energy prices = and=20 seller concentration, or the effect of open access on = those=20 prices. ...

TEXT:
[*401] =

I. Introduction

All mergers affect = competition,=20 some by creating superior competitors and others by creating potential=20 monopolists. The Antitrust Improvements Act of 1976 requires = prescreening of=20 proposed mergers to identify those that are likely to affect competition = adversely. n1 To implement that law, the U.S. Department of Justice's = (DOJ)=20 Merger Guidelines contain prescreening procedures that = attempt=20 a compromise between theoretical rigor, limited data, expeditious = processing,=20 and consistency. n2 The Guidelines single out for = scrutiny=20 those mergers that significantly affect numerical measures of supplier=20 concentration in relevant markets. In electric utility mergers the = antitrust=20 agencies (DOJ and the Federal Trade Commission) defer to the opinion of = the=20 Federal Energy Regulatory Commission (FERC). The FERC, = however,=20 often uses the concentration standards of the = Guidelines in its=20 decision process. As electricity markets change, the FERC's continued = reliance=20 on the Guidelines raises the likelihood that it will = deny=20 applications for competitive mergers or permit monopolizing mergers to = proceed.=20 In some industries a merger that increases supplier concentration may = make=20 collusion easier, but in industries with open access to = essential facilities any link between concentration and monopolization = will be=20 attenuated. The market at risk of monopolization in an electric utility = merger=20 is a market for facility access, rather than one for = goods or=20 services that the [*402] Guidelines might = suggest.=20 If access is critical and concentration is not, the = FERC Order=20 888 provides a new foundation for rational merger policy. n3 =

Section 203=20 of the Federal Power Act requires that the FERC investigate mergers of=20 jurisdictional utilities and approve those "consistent with the public=20 interest." n4 The Federal Power Commission (FPC) itemized that interest = in its=20 approval of a 1967 merger between Commonwealth Edison Company and = Central=20 Illinois Electric and Gas Company.

... The = Commission, in=20 determining whether this particular merger is consistent with the public = interest, has considered, among others, the following factors: the = effect of the=20 proposed action on the applicants operating costs and rate levels, the=20 contemplated accounting treatment, reasonableness of the purchase price, = whether=20 the acquiring utility has coerced the to be acquired utility into = acceptance of=20 the merger, the effect the proposed action may have on the existing = competitive=20 situation, and finally, whether the consolidation will impair effective=20 regulation either by this Commission or the appropriate state regulatory = authority. n5

In January 1996 the FERC issued a = Notice of=20 Inquiry (NOI) into revision of these standards. n6 Many respondents to = it=20 support continued reliance on the Guidelines as a = screen for=20 market power and broadly agree with FERC's existing procedures for = identifying=20 those markets in which a merger may affect competition. n7 =

Concurrently=20 with the NOI, the FERC is debating whether to continue to approve = mergers as=20 long as they do not harm competition (its existing standard) or if it = should=20 undertake proactive rulemakings and proceedings with the objective of = increasing=20 competition. n8 A 3-2 Commission majority recently set the proposed = merger=20 between Baltimore Gas and Electric Company and Potomac Electric Power = Company=20 for hearing because of concerns that "applicants may have overstated the = size of=20 the geographic markets in which generation concentration is measured." = n9 The=20 dissenting Commissioners recommended immediate approval, noting that the = majority intended to examine competition even though there were no = interven-=20 [*403] ors who disputed the applicants' claim of competition = in all=20 relevant markets. n10 In the since-abandoned merger between Washington = Water=20 Power and Sierra Pacific Power, a lack of intervention did not deter the = FERC=20 from setting competitive issues for hearing because of concerns about=20 post-merger transmission availability. n11 The FERC is proposing a = longer-term=20 role for itself in the pending merger between Public Service Company of = Colorado=20 and Southwestern Public Service Company, who intend to connect their = systems=20 with a high-voltage link that will begin operation in 2001. If it = approves the=20 merger, the FERC intends to order filing of a supplemental market power = analysis=20 six months before the line goes into service. n12 If the FERC finds = "competitive=20 harm associated with the operation or control of the new line," it will = consider=20 imposing an open season on the line or requiring construction of = additional=20 transmission facilities. n13

Since the mid-1970s, the FERC has = used the=20 tools of antitrust analysis to analyze competition in industries that it = regulates. n14 In addition to electric utility mergers, the Commission = has=20 defined relevant markets in dockets where a utility or independent power = producer is seeking market-based rates for wholesale sales. n15 In the = past, the=20 FERC staff constructed antitrust markets to determine competitive = situations=20 that might be affected by foreclosures of access to=20 transmission facilities. n16 Currently, the FERC is defining markets to = help=20 determine conditions under which it will authorize interstate gas = pipelines to=20 charge market-based rates for transportation. n17 It is also examining = markets=20 for gas storage in connection with market-based rate requests. n18 = Earlier in=20 the history of open access, the FERC used antitrust = markets to=20 evaluate applications by pipelines for market-based gas inventory = charges. n19=20 It has also employed them in market-based rate proceedings for oil = pipelines,=20 considering the competition applicants faced from other pipelines and = from other=20 modes of transport. n20

The FERC's NOI on electric utility = mergers=20 offers it a unique opportunity to refashion its principles to comport = with=20 changes in power markets that will come with open = access. At=20 the same time, a wave of merger applications offers an opportunity for=20 case-specific application of the new principles. The FERC currently = assumes that=20 open access transmission tariffs for comparable service = could=20 mitigate or eliminate market power in transmis- [*404] sion. = n21 It=20 continues to evaluate capacity and energy markets, = however, by=20 the standards of the Guidelines. The FERC must = reconsider both=20 those methods and the continuing relevance of those markets in light of = its=20 developing open access policy. If the FERC makes = transmission=20 access the prime criterion for merger approval, it = directly=20 confronts the most likely source of anticompetitive effects. An=20 access-based approach is consistent with established = principles=20 of market definition and applicable in a range of regulatory = environments. This=20 departure from past practice is necessary because the competition that=20 regulators must protect is qualitatively changing. Mergers, power = exchanges,=20 retail wheeling and other new institutions will bring forth markets = whose=20 organization and scope we cannot envision today. Competition to design = new=20 transactions and create new markets is as important for economic welfare = as the=20 competition that exists in today's transitional industry. Open=20 access looks to both the present and the future in ways = that=20 the Guidelines do not.

The next section of this = article=20 describes the Guidelines and examines their relevance. = It=20 broadly concludes that the economic theory underlying them is unclear = and the=20 empirical evidence on their applicability is inconclusive. Section III = then=20 discusses the problems encountered in applying the = Guidelines=20 to electricity markets, in particular the difficulties of inferring = monopoly=20 power from data on the concentration of sellers. Even if the FERC uses = the=20 Guidelines with care, they will not be helpful for = evaluating=20 competition in an industry with open access. Section IV = suggests that the FERC's merger policy should deemphasize antitrust = markets for=20 capacity and energy. Instead, the Commission should = concentrate=20 on the intensive study of transmission access in the = merged=20 system, a subject on which it already has regulatory expertise. Section = V=20 expands on the critique of capacity and energy markets, = concluding that in today's transitional electricity industry those = markets are=20 particularly unlikely to improve the Commission's decision-making = abilities.=20 Section VI extends this discussion to more general considerations of = antitrust=20 activism in the current merger environment. In electricity, economists = have=20 identified problems that will often be better resolved by regulation and = other=20 litigation than by antitrust. Section VII draws some more general = conclusions on=20 the relationship between regulation and antitrust.

II. = Merger=20 Policy and the Merger Guidelines

A. The=20 Guidelines

The Guidelines provide the = antitrust=20 agencies with standardized procedures to evaluate the numerous and = heterogeneous=20 mergers that they must screen. Analysis under the = Guidelines=20 starts from the relevant product and geographic markets in which the = merger=20 might allow the exercise of monopoly power, defined as "the ability = profitably=20 to maintain prices [*405] above competitive levels for a = significant=20 period of time." n22 The relevant market is for "a product or group of = products=20 such that a hypothetical profit-maximizing firm that was the only = present and=20 future seller of those products ("monopolist") likely would impose a = "small but=20 significant and nontransitory' increase in price." n23 The product group = is the=20 smallest set for which the hypothetical monopolist (i.e., a perfectly = enforced=20 collusion that includes both the merged firm and all others in the = market) could=20 profitably increase price by the critical amount, "in most contexts ... = [an]=20 increase of five percent lasting for the foreseeable future." n24 The = relevant=20 geographic market is the smallest area in which a monopolist in the same = products could profitably impose that price increase. n25 Market = definition=20 focuses solely on the ability of purchasers to abandon more expensive = products=20 (demand substitution factors). n26 The likelihood that new sellers will = enter=20 the market in response to higher prices is considered separately. n27=20

Having found the relevant geographic area and products, the = antitrust=20 agency must determine whether that market can sustain the threshold = price=20 increase. A hypothetical monopolist's power to profit by raising price = depends=20 on the "likely demand responses of consumers" and the "profitable supply = responses [of market entrants] likely to occur in one year and without = the=20 expenditure of significant sunk costs of entry and exit." n28 Consumer = responses=20 include switches to substitute products and to the same product shipped = in from=20 other areas. Existing producers can respond by increasing output or = shipping it=20 across geographic lines (including national boundaries), and new = producers can=20 open up for business. Even with copious data, economists will have = difficulty=20 predicting and quantifying these responses. An agency with limited = resources=20 that must preapprove numerous mergers has little choice but to abandon=20 quantitative sophistication for a simpler way of screening out = questionable=20 deals.

For their screen the Guidelines use the=20 Herfindahl-Hirschman Index (HHI). The HHI is the sum of the squares of = the=20 shares of all sellers in the relevant market. n29 The HHI may be a more=20 sensitive indicator of monopoly power than a market share calculation = because=20 the HHI covers all firms in the market and large firms carry = disproportionate=20 weight in it. A market with N sellers of equal size has an HHI of 1/N. = n30 A=20 monopolized market [*406] has an HHI of 1 [or 10,000] and in = the=20 limit a market with numerous small sellers has an HHI approaching 0. If = the=20 market's post-merger HHI is below 1,000, it "will ordinarily require no = further=20 analysis." n31 A merger that raises a market's HHI by less than 100 = points but=20 keeps it under 1,800 is "unlikely to have adverse competitive = consequences," but=20 one that raises the HHI by over 100 points "potentially raises = significant=20 competitive concerns" even if it remains below 1,800. n32 A merger that = raises=20 the HHI by over 50 points in a market where it is already over 1,800 = raises the=20 same potential concerns, and a merger that increases such a market's HHI = by over=20 100 points is "likely to create or enhance market power or facilitate = its=20 exercise." n33 Applicants for the latter merger might respond by showing = that=20 industry characteristics make collusion difficult, that quick market = entry by=20 new competitors is likely, or that merger is the only way to achieve = lower costs=20 or to save a failing firm's assets. n34

B. Economic Theory and = the=20 Guidelines

The antitrust agency = screening a=20 proposed merger must make predictions about competition that are largely = beyond=20 the competence of present-day economics. Competition in the marketplace = usually=20 takes the form of unruly and complex rivalry among highly motivated = participants=20 whose actions will be difficult to foresee. Such competition is = desirable=20 because it promotes innovation and puts a stop to activities that do not = create=20 value. No available economic theory consistently and accurately predicts = the=20 path along which a continuously rivalrous market will evolve. Instead,=20 economists must distill from individual market situations those = principles from=20 which they can draw generally applicable predictions about competition. = The=20 method of choice is to focus on the equilibrium a market will attain in = the=20 absence of such disturbances as innovation or developments in related = markets.=20 n35 Equilibrium models start with assumptions about buyer and seller = behavior to=20 predict the price, output, profitability, and other measures of = performance that=20 a market will settle into. The economic theory that underlies merger = policy=20 compares the equilibria of pre-merger markets and post-merger markets = that have=20 a greater concentration of suppliers. Unfortunately for policy, small = changes in=20 theoretical assumptions (e.g., about how a seller responds to the = decisions of=20 competitors) can yield substantially different predictions about = equilibrium and=20 about how a merger might affect that equilibrium. [*407] =

The=20 Guidelines do not include a formal statement of the = economic=20 theory that underlies their market definitions and criteria for merger = approval.=20 n36 They do, however, describe the empirical consequences of the theory. =

Other things being equal, market concentration = affects=20 the likelihood that one firm, or a small group of firms, could = successfully=20 exercise market power. The smaller the percentage of total supply that a = firm=20 controls, the more severely it must restrict its own output in order to = produce=20 a given price increase, and the less likely it is that an output = restriction=20 will be profitable. If collective action is necessary for the exercise = of market=20 power, as the number of firms necessary to control a given percentage of = total=20 supply decreases, the difficulties and costs of reaching and enforcing = an=20 understanding with respect to the control of that supply might be = reduced. n37=20

Collusion, however, is only one of many possible = paths=20 that a market can take when individual sellers have some ability to = choose their=20 prices. The likelihood of collusion rises if the market's institutional=20 characteristics are conducive to forming and enforcing an agreement.=20 Characteristics enumerated in the Guidelines include = the public=20 visibility of prices and transactions, the homogeneity of products, and = the=20 existence of contract terms that might facilitate overt or tacit = collusion. n38=20 The likelihood of collusion also depends on the expectations of sellers=20 regarding how other sellers will react to their individual decisions. = Depending=20 on these expectations, market outcomes can range from aggressive rivalry = that=20 delivers very low prices to perfect collusion that delivers monopolistic = ones.=20 In only one case ("Cournot" expectations) are the output and price = effects of=20 mergers well-predicted by conventional measures of market shares and=20 concentration. n39 There is little empirical evidence that sellers in = actual=20 markets, including electricity, have such expectations about one = another. n40=20 The many plausible economic theories of oligopoly include models of = dominant=20 firms, price [*408] leadership, market-share growth = strategies,=20 rivalry in producing innovations, and game-theoretic worlds in which=20 super-rational competitors can arrive at numerous possible end-states = depending=20 on the details of what is assumed. n41 The authorities probably err on = the side=20 of excessive caution when they base premerger approvals on a theory that = posits=20 collusion as the most likely consequence of seller concentration. n42 = The law,=20 however, may require such caution. n43 The change in a market's HHI = bears no=20 necessary relation to a merger's economic benefits and costs. More = efficient=20 (i.e., lower cost) production benefits both the merging parties and = society=20 because it reduces the resources necessary to produce a unit of output. = Against=20 this benefit, the economist weighs the likelihood that resources will be = misallocated because the post-merger market may be more prone to = collusion and=20 restrictive practices. n44 A merger that increases market price by more = than the=20 threshold of the Guidelines may still pass the = cost-benefit=20 test if it yields large enough production efficiencies. n45

C. = Economic=20 Evidence on the Guidelines

Of the many = possible=20 economic theories, only some contain logical demonstrations that = concentrated=20 markets will reach collusive equilibria and unconcentrated ones will = not. n46=20 Since the 1930s, economists have attempted to evaluate these theories by = measuring the existence and strength of statistical associations between = seller=20 concentration (e.g., as measured by the HHI) and market performance = (e.g., as=20 measured by profits or prices) over samples of industries. n47 These = studies=20 have not substantially narrowed the range of professional disagreement = about=20 these [*409] relationships and their interpretation. n48 = Some=20 believe that the evidence for a positive concentration-performance = relationship=20 is strong enough to serve as a basis for aggressive antitrust policy. = n49 Others=20 contend that the measured strength of that relationship is highly = sensitive to=20 the choice of data samples and statistical methods. n50 The = interpretation of=20 the statistical work is also unresolved. Positive correlations between=20 concentration and profit or price are consistent with a range of seller = behavior=20 that includes collusion as one of numerous possible cases. Such = correlations can=20 be evidence of efficiency rather than collusion if economies of scale = are=20 substantial. n51

Seeking more direct evidence on price, some = economists=20 have studied how seller concentration affects it in different markets = for the=20 same good. n52 In the largest work of this kind, Leonard W. Weiss and = his=20 associates examined 121 different sets of data, mostly for such services = as=20 banking, retailing, advertising space, and rail freight. n53 Of these = data sets,=20 between 61 and 73 percent exhibited a positive price-concentration = relationship,=20 but the chosen criterion for positivity is open to question. n54 Other = studies=20 have found positive concentration-price relationships in some industries = (air=20 fares and bank loans) and negative or nonexistent ones in others = (brewing,=20 baking, and grocery retailing). n55 There is no unique link between the = price=20 effects of concentration and the HHI standards of the=20 Guidelines, but under plausible assumptions = concentration that=20 would trigger scrutiny under the Guidelines often = produces a=20 price increase of less than five percent. n56 [*410] =

However=20 uncertain the economics, screening mergers by a numerical standard may = still=20 yield more economically rational decisions than the making of = case-specific=20 judgments. If so, alternative HHI criteria will have differing = consequences.=20 Research on the relationship between the HHI and industry performance = has found=20 that neither the danger level of 1,800 nor the safe harbor of 1,000 are=20 associated with any notable departures from competitive outcomes. n57 = Even if=20 the HHI is a useful warning of collusion in certain market situations, = no=20 concise characterization of those situations is currently available. n58 = The=20 originators of the Guidelines selected 1,000 for a safe = harbor=20 HHI "as much as a political anchorage [against more restrictive views] = as=20 because anyone thought that nicely round number was just right." n59 = Commenting=20 on the FERC's merger policy, the DOJ states that "changes in market = structure=20 such as market share and market concentration are indicators of the = potential=20 for the merged firm to behave anticompetitively." n60 In reality, the = evidence=20 favoring the Guidelines is sparse.

III. = Structural=20 Analysis of Regulated Markets

A. Concentration and Competition=20

The common practice of calculating concentration = statistics is=20 intrinsically questionable if a market is under price, profit, and = service=20 regulation. Even if market structure accurately predicts outcomes in = unregulated=20 industries, it must be used with great care in industries where = regulators=20 determine important aspects of that structure. n61 The pitfalls of = structural=20 analysis are at their clearest in the antitrust context of Otter Tail = Power=20 Company v. United States. n62 The relevant markets in contemporary = merger and=20 market-based rate dockets are usually not those of Otter Tail, but the = methods=20 of analysis are descended from that case. More recent capacity and=20 energy market definitions are examined in Section V = below. In=20 Otter Tail, the Supreme Court accepted the government's structure-based=20 determination that a utility had monopoly power. The Court held that = Otter Tail=20 had violated the Sherman Act by refusing to wheel inexpensive federal = power to=20 newly-formed municipal utilities, thus using its transmission monopoly = to block=20 retail competition. In the relevant market for [*411] retail = service, the government found a monopoly because Otter Tail's service to = 465 of=20 the 510 municipal franchises in its territory gave it a market share of = 91=20 percent. n63 Otter Tail's refusal to wheel probably deterred some = franchise=20 changeovers, but the fraction of cities it served at the time of the = refusals=20 either tells nothing or misleads regarding its ability to act = monopolistically.=20 Instead of indicating market power, the company's high "share" of = franchises=20 could equally well reflect its inability to refuse service to them. n64 = At=20 trial, Otter Tail argued that the major determinant of municipal utility = formations was their legal priority to obtain inexpensive federal power. = n65=20 Unallocated blocks of this power, however, turned up only infrequently. = If the=20 only superior alternative to Otter Tail was so limited, the company's = continuing=20 service to the many cities that were without that alternative cannot = reliably=20 indicate its monopoly power. Otter Tail's refusal to wheel was an = exercise of=20 monopoly power over transmission, but the government did not call = transmission a=20 relevant market. n66 Instead, the government was content to state that = the=20 company had "strategic dominance." n67 If (by analogy to retail) Otter = Tail's=20 territory bounded the transmission market, the company actually had too = low a=20 share for monopoly power. It owned only 6.2 percent of transmission in = the=20 territory, the remainder being held by cooperatives, municipals, other = corporate=20 utilities, and the federal government. n68 By refusing = access=20 to certain lines, however, the company was able to thwart franchise = changeovers.=20 Otter Tail's low share of transmission still allowed it to harm those = cities=20 without alternative paths to inexpensive power. Under regulation, both = high and=20 low concentration lose whatever inferential value they might have in = unregulated=20 markets.

B. Open Access: Competitive Allocation = of a=20 Natural Monopoly

In a natural monopoly, the minimization = of total=20 cost requires that only a single facility of appropriate size be = constructed.=20 n69 If electric transmission between two points is a natural monopoly,=20 regulators promote efficiency by permitting construction of only one=20 high-capacity line and [*412] imposing a service obligation = on its=20 owner. In reality, there are engineering limits to the economic capacity = of one=20 line embedded in a larger system, and reliability may warrant = duplicative lines=20 of lower capacity. Operational problems aside, the line's owner need not = be a=20 reseller of purchased or self-produced power to end-users. If the line = operates=20 under open access rules, users and producers of power=20 (shippers) can have contractual rights to the line's capacity and = related=20 services. n70 If shippers can subdivide and trade those rights, exchange = will=20 allocate them to users who value them most highly, as it would in a = competitive=20 market. Natural monopoly is an attribute of transmission technology, but = the=20 market in which the line's services are exchanged can be competitive. = The fact=20 that a single line is the cheapest link between two points tells nothing = about=20 who should produce, purchase, or market through put in order to maximize = the=20 line's economic value.

Assume that a number of small, = separately-owned=20 transmission lines link a power producing area with a consuming area. = n71=20 Shippers contract for transportation with owners of the lines in an = environment=20 where information about opportunities and alternatives is easily = available. In=20 this market, a shipper's options for contracting with any one line are=20 unconstrained by its dealings with other lines. As described, the market = is=20 probably competitive enough to escape the attention of antitrust and = there is no=20 reason to institute rate regulation. n72 (It resembles inter-city = trucking.) Now=20 assume that a single line with the same total capacity replaces the = small ones=20 and each former owner contracts with it, at cost-based rates, for the = same=20 capacity it held previously. Under open access, = capacity=20 holders can repackage and retrade it among themselves and with shippers = who in=20 turn can further parcelize and reallocate it. If capacity holders cannot = collude=20 (e.g., because agreements are unstable or because antitrust is a = threat), the=20 competitive price for service and the competitive allocation of space = will again=20 prevail with a single line carrying everyone's power.

Rate = regulation of=20 capacity contracts may be necessary to deter the line's owner from = attempts to=20 withhold capacity or price discriminatorily. A regulatory = "use-it-or-lose-it"=20 provision is also needed to discourage attempts by holders of capacity = rights to=20 monopolize or collude. Under such a rule, unused capacity reverts to the = owner,=20 who must offer it for [*413] interruptible or short-term = firm=20 service. n73 The more capacity a collusion attempts to withhold, the = lower the=20 likelihood of interruptions on transportation offered for interruptible = service.=20 The line's owner is, in effect, a competitive market entrant who = instantaneously=20 acquires all unused capacity and has no choice but to make it available. = If all=20 available capacity thus reaches the market, each unit of it will fetch = the=20 competitive price.

"Use-it-or-lose-it" rules exemplify the = importance of=20 analyzing market institutions rather than statistics of concentration.=20 Concentration statistics for facility ownership or capacity holding are=20 calculable under any system of rules. Competitive allocation of = capacity,=20 however, will be more likely with open access than = without it.=20 The choice of relevant markets is itself institution-dependent. If the = line has=20 a single owner, the HHI for ownership is 1.0 [or 10,000]. A market for=20 ownership, however, is not where competition happens. n74 If rights to = the line=20 are well-defined and exchangeable, competition happens in a market for=20 access. n75 The state of competition for=20 access bears no necessary relation to the number of = entities=20 that have rights to use the line at any moment. The = Guidelines=20 state that ease of entry into a market can counteract the potential = effects of a=20 merger that would otherwise raise concerns. n76 If so, an HHI based on = current=20 user statistics loses significance because open access = opens a=20 market for reallocating rights to the line, one that is open to new = users and in=20 which existing users cannot contrive an artificial scarcity. =

Experience=20 with gas pipelines broadly validates the importance of = access=20 and the irrelevance of concentrated ownership. Since the coming of open=20 access, markets for both gas and its transportation = have=20 performed far more competitively than before while the concentration of=20 pipelines between most fields and city-gates has changed little. Events = have=20 outpaced the FERC's interest in determining when it should allow = market-based=20 pipeline rates because the evidence is that market-based rates are = already in=20 effect. Secondary capacity markets are growing, discounting is = widespread, and=20 "gray market" transactions effectively circumvent secondary market price = caps.=20 n77 The amount and depth of discounting between fields and city-gates is = independent of the concentration of pipeline owner- [*414] = ship=20 linking the points. n78 In gas, the evidence is becoming clear that = institutions=20 matter and structure does not.

IV. An = Access-Based=20 Standard

A. Access as a Merger Screen =
=20
Because the Guidelines lack empirical support and = are=20 problematic in regulated industries, they will probably not serve the = FERC well=20 in analyzing what many expect will be a wave of electric utility = mergers. The=20 Commission is likely to gain more insight by applying its existing = regulatory=20 expertise to the implementation of open access in = merged=20 systems. As competitive generation grows, utilities might best protect = their=20 remaining monopoly power by providing competitors with unequal = transmission=20 access. If so, the FERC should find a merger consistent = with=20 the public interest in competition if it determines that the post-merger = utility=20 will comply with all open access practices and = standards in=20 effect. If it analyzes capacity or energy markets, the = FERC=20 should take great care before applying the Guidelines = to them.=20 n79 If it finds risks that open access will be = administered=20 anticompetitively, the FERC should explore conditions on the merger. It = might=20 require the reformulation of questioned contract or tariff provisions = (beyond=20 the requirements of Order 888). n80 If operating practices carry risks = of=20 anticompetitive action, the FERC might require an Independent System = Operator=20 (ISO). If market imperfections will give the merged system an = unwarranted=20 competitive edge, the Commission might consider specifying = characteristics of=20 the power exchange institutions the merged company operates under. n81=20

Orders 888 and 889, and the Pro Forma Open = Access=20 Tariffs, specify institutions and practices intended to allocate = transmission=20 and related facilities competitively: an allocation process and = reservation=20 priorities for existing capacity, a requirement of efforts to expand = congested=20 facilities, a nondiscriminatory and uniform system of information = exchange, and=20 a mechanism for reassignment of capacity. The details of capacity = reservation=20 and transmission pricing necessary for competitive = access are=20 subjects of separate proceedings. n82 If those proceedings produce = efficient=20 reserva- [*415] tion and pricing schemes, open=20 access will have brought about the competitive market = for=20 delivery that is the necessary complement of competitive markets for=20 energy and capacity. n83 =

Access as a=20 merger screen is consistent with the FERC's view that open=20 access tariffs for comparable services mitigate or = remove=20 market power in transmission. n84 It can also be considered a relevant = market=20 whose product is the right of access to transmission in = the=20 merged system. n85 Although open access deprives HHIs = of much=20 meaning, if the FERC wishes to do so it can calculate them for existing = users of=20 the network or subsets of it. Whatever the calculated value, the merger = should=20 rise or fall according to whether the market for access = satisfies the Guidelines' standards for ease of = competitive=20 entry. n86 The FERC has acknowledged that existing markets for capacity=20 reassignment allow "transmission customers to compete with the owner to = some=20 extent in the firm transmission market." n87

An = access=20 standard makes exclusionary conduct and situations that are conducive to = it=20 easier to identify. For example, the Otter Tail Court would have = received better=20 guidance from a market for access rights than it = received from=20 the retail market that it accepted. Otter Tail participated in wholesale = pooling=20 activities over the upper midwest and wheeled extensively for other=20 transmission-owning utilities. n88 Its wheeling policies, however, = excluded=20 potential competition for transmission access from new=20 municipal systems. By excluding municipals, Otter Tail profited from its = ability=20 to foreclose their alternatives. To reach the "right" result from = antitrust=20 precedent, the Otter Tail Court had to rely on an uninformative = calculation of=20 franchise shares and to disregard shares that indicated that the company = lacked=20 monopoly power in transmission. Defining the relevant market as one for=20 access rights leaves fewer gaps in the logic. =

B.=20 Access Present and Future

The = electricity=20 markets of the future will probably contain commodities and services = that are=20 unknown today. Some transactions that are common today will probably = lose=20 importance as the industry changes. The market institutions of the = future will=20 also differ, both from today's institu- [*416] tions and = from those=20 institutions that many conjecture that they will see. An=20 access-based standard is well-suited for mergers in an = industry=20 whose future shape is so uncertain. The standard adapts well because=20 access is the right to transmission of any product that = might=20 be invented rather than only those products that are traded today. = Whatever=20 market institutions evolve, traders will require transmission=20 access to arrange their receipts and deliveries. An=20 access standard also transcends changes in the numbers = and=20 types of eligible transmission users. If retail customers gain wheeling = rights,=20 the criteria for openness will be the same that apply in markets that = currently=20 exclude them. The standard of competitive access also = adapts=20 easily to changes in the federal-state jurisdictional boundary and is = consistent=20 with differing state policies on retail wheeling.

In recent = merger=20 dockets, the FERC has usually singled out short-run capacity, nonfirm=20 energy, and transmission as relevant markets. n89 As = open=20 access becomes general and markets evolve, the = opportunities=20 for users to assemble their preferred power supplies from unbundled = components=20 will grow apace. Under the Guidelines, relevant = products are=20 determined by interchangeability. As markets evolve with open=20 access so will the boundaries within which products are = interchangeable. If the industry's institutions are changing = drastically, the=20 products relevant for today's mergers are less likely to remain relevant = than=20 they would if mergers were taking place against an unchanging = institutional=20 background. Market geography will also change. In recent mergers, the = FERC has=20 accepted wholesale markets that include only systems that are one or two = wheeling transactions away from the new company. n90 In at least one = recent=20 merger, this geography excluded from the market long-distance = transactions that=20 are actually taking place. n91 Under open access, only = cost=20 will limit the scope of power exchanges, and shares in such arbitrarily = defined=20 markets will lose any relevance they might have today. n92 If open=20 access makes currently uncommon transactions such as=20 transmission by displacement more feasible, it will also change market = areas in=20 unpredictable ways.

Just as predicting new institutions and = products is=20 inherently risky, so is predicting a merger's long-term effects by = examining how=20 it plays under today's market power standards in today's markets. = Merging=20 utilities often state an intention to compete by inventing new products, = new=20 types of [*417] transactions and new market institutions. = n93 The=20 power markets that cover the west today grew and changed over three = decades in=20 ways that no one could have predicted when transactions began. n94 One = might=20 justifiably question proposals that the FERC can expedite merger = processing if=20 it institutes a rulemaking to define a generally applicable set of = markets using=20 computer models of power flows. n95 Beyond the risk of error, when the=20 environment is changing so rapidly, computerized economics is inherently = contentious. In the 1980s, conflict over computer models in California's = resource planning process led to legislation that identified acceptable = software=20 by name and required annual reports by the Public Utilities Commission = to the=20 legislature. The legislation could not reconcile differences between = utilities=20 and intervenors over data used as input to the models, and the = subsequent record=20 "leaves an impression that disputes were fundamentally settled by = negotiation=20 rather than fact." n96

V. Capacity and Energy = as=20 Relevant Markets

Open access may nullify = monopoly power in transmission, but the FERC believes it possible that = an open=20 access utility can have monopoly power over capacity or = energy. n97 The FERC now presumes that open=20 access utilities requesting market-based=20 energy prices have no monopoly power in new generation. = n98=20 Regarding existing capacity, however, merger could conceivably create a = utility=20 with monopoly power even if neither of the parties alone possessed it. = Before=20 concluding that a market study of capacity or energy is = warranted, a cost-benefit analysis is in order. The FERC can impose = costs on=20 society by approving mergers that turn out anticompetitive and by = rejecting=20 mergers that would have increased competition. Foregone competitive = benefits and=20 deadweight losses from monopoly are amounts that economists treat = symmetrically.=20 n99 Cost-benefit analysis is an exercise at the margin: If a merger has = passed=20 the open access screen, does further study of=20 energy or capacity increase the likeli- = [*418] hood=20 that the FERC will reach a correct decision? n100 That research will be=20 beneficial if it provides information that the open = access=20 analysis does not, and if the accuracy and relevance of this information = reduces=20 the FERC's uncertainty. n101

Applying the = Guidelines,=20 however, is particularly problematic in markets where commodities, = geography,=20 traders, and institutions are all in flux. The = Guidelines may=20 pass a cost-benefit test for mergers in unregulated industries. They do = so,=20 however, not because they are effective discriminators, but because if = premerger=20 screening must be done there is no clearly superior alternative. The = antitrust=20 agencies defer to the FERC's expertise, and that expertise surely = includes=20 insight into the nature of power markets that goes beyond generalities = about=20 concentration. The FERC should use the Guidelines only = if it=20 has reason to believe them the best tool to apply to the markets it = understands.=20 n102 Before accepting the Guidelines, the FERC should = try using=20 some available data to test their applicability. The Commission has yet = to=20 produce publicly available studies of such basics as the relationship = between=20 energy prices and seller concentration, or the effect = of open=20 access on those prices. n103

Capacity and=20 energy markets are often defined by the service areas = and=20 interconnections of utilities, pools, and reliability councils. Those = areas have=20 usually taken their existing shapes because of politics or historical = accident=20 (e.g., past mergers) rather than as outcomes of market forces. They were = also=20 shaped by electrical technologies that once required companies to be=20 self-sufficient. Measures of size or concentration based on such = territories can=20 be unreliable and misleading, as they were in Otter Tail and are today = under=20 rules-of-thumb that limit trading distances to one or two wheeling = transactions.=20 Increasing competition will bring pressure on utilities to merge = irrational=20 territories or otherwise change their operations. n104 Whether or not = utilities=20 merge, the areas over which they transact [*419] will change = as=20 competition grows and regulation changes. n105 If economic and political = forces=20 will so change markets, the long-term relevance of any statistics that = use=20 today's territories should be demonstrated rather than assumed. =

Market=20 forces are also changing the identities of economic and uneconomic power = producers in ways that measures of concentration do not capture. = Increasing=20 amounts of generation are being constructed by non-utilities. As market = forces=20 increase the fraction of generation owned by non-utilities, they also = increase=20 the fraction of older utility-owned plants that will be economically = unable to=20 compete. n106 Measures of concentration become particularly suspect when = analyzing efficient plants because they will depend on market prices of = the=20 future that are intrinsically unknowable today. n107 Investments in = efficient=20 new plants have been a major force in expanding market areas. This = expansion=20 will continue as the proportion of efficient plants rise and open=20 access becomes universal.

Open = access=20 changes product interchangeability and conditions of market entry. As it = puts=20 neighboring utilities under more uniform rules, open = access=20 increases their freedom to experiment with transactions that mitigate=20 distance-related restrictions on market areas. Open = access may=20 also facilitate the entry of new producers because it guarantees them = the same=20 transmission rights as existing ones. If gas is an apt analogy, open=20 access that makes transmission more competitive can = also=20 increase competition in power production. n108 Even in the worst-case = scenario=20 of collusion in generation markets, open access can = help=20 destabilize restrictive agreements because it prohibits transmission = owners from=20 using exclusion as an enforcement device. With or without open=20 access, a transmission owner can attempt to operate its = system=20 to favor its own power production. n109 The transmission owner's = likelihood of=20 success, however, probably depends more on details of the implementation = of open=20 access than on the concentration of generators in its = area.=20 n110

The institutions that underlie market exchange determine = both the=20 possibilities for competition and the efficiency with which markets = operate. In=20 most industries, merger analysts can safely assume that future sellers = will be=20 operating in markets that resemble today's. In electric utilities, = mergers are=20 occurring amidst regulatory changes that will bring new market = institutions. For=20 mergers whose HHIs cause concern, the Guidelines call = for an=20 examination of how market institutions may affect post-merger competi-=20 [*420] tion. n111 If future market institutions are unknown, = however, the HHI may lose whatever inferential value it has in markets = whose=20 institutions are not changing.

Some commentators hold that = experience=20 with the United Kingdom's mandatory and centralized electricity pool = market=20 provides evidence that higher seller concentration leads to higher = prices. n112=20 U.K. generators apparently act strategically to produce = energy=20 prices that sometimes exceed the operating cost of the most costly plant = bid=20 into the pool. n113 Market institutions there, however, may require that = price=20 exceed operating costs if suppliers are to meet their capital = obligations. n114=20 Norway has an unconcentrated generation industry and a voluntary pool. = Nearly=20 ninety percent of power in Norway is traded under contracts that are = independent=20 of the pool, making it hard to determine how closely prices track costs. = n115=20 This superficially less efficient bilateral market may produce more = nearly=20 competitive outcomes than the centralized exchange in the U.K. If = Norway's=20 performance is in fact superior, it could be due to either the larger = number of=20 sellers or to the absence of compulsory institutions that facilitate = coordinated=20 action by U.K. generators. In another variation, Chile's mandatory pool = buys=20 from producers whose HHI exceeds that of the U.K. generating sector, but = Chilean=20 prices track costs quite closely. n116 If institutional differences can = lead to=20 substantially different market outcomes, the relevance of concentration=20 statistics must be questioned. [*421]

VI. = How Useful=20 Is Antitrust?

Despite the ambiguity surrounding the=20 concentration-performance relationship in electricity, some economists = recommend=20 more active antitrust intervention in utility mergers. Writing for the = American=20 Public Power Association, Professor William Shepherd believes that = without=20 antitrust the current merger wave will lead to dominance of the industry = by a=20 small number of firms. n117 The poor performance of dominant firms in=20 unregulated industries is a complex issue, and Shepherd offers no = quantitative=20 evidence on the effects of dominance in electricity. n118 Some = economists favor=20 antitrust policy that prohibits or strongly conditions mergers that = violate=20 bright-line quantitative standards. n119 Others favor superimposing on = the=20 Guidelines a rule of reason based on the facts of each = merger.=20 n120

To its supporters, antitrust can attack both horizontal = market=20 power that arises when mergers concentrate generation and vertical = market power=20 that arises when an integrated utility compels customers to take = uneconomic=20 power by denying them transmission access to = alternatives. n121=20 As discussed above, the rationale for intervention in capacity and=20 energy markets is not compelling. Regarding = transmission,=20 antitrust activists doubt that open access will be = sufficient=20 to eliminate the risk of vertical market power. Their major concern is = that=20 transmission owners can manipulate availability and operating procedures = to=20 favor their own transactions over those of competitors. =

=20
... "Open access" is probably no major cure for = monopoly=20 power...The problem is that electricity services can be extremely = complex,=20 particularly involving long-run full-reliability services...The terms of = access can be a thicket of difficulties, in which "open = access" is actually pretty much closed to = competition...But=20 even if all conditions could be arranged perfectly, = access may=20 leave major degrees of monopoly power undisciplined. n122 = [*422] =20

As examples, utilities might alter the = transmission=20 available to generation competitors by strategically reconfiguring their = own=20 operations, by transacting for themselves to produce parallel flows on = lines=20 that are important for competitor transactions, and by strategically = scheduling=20 or neglecting maintenance. n123 Some of the transmission discrimination = episodes=20 cited in Order 888 could probably have been attempted by utilities in an = open=20 access regime as well. n124

The relevance of = antitrust=20 to operational incidents such as these is unclear. As noted in the = discussion of=20 Otter Tail, a utility's absolute size is of little consequence for its = monopoly=20 power over a single customer's transmission alternatives. Because = transmission=20 is a singular resource operating under technical circumstances that = regulators=20 cannot monitor perfectly, its owners will quite possibly have leeway to=20 manipulate it as alleged. They will, however, have this power whether = antitrust=20 oversight of generation is tight or lax. Instead of an antitrust issue, = this is=20 a situation in which efficiency may require that the ownership of = transmission=20 be separated from its control. If this separation is warranted, = formation of an=20 Independent System Operator may be in order. n125

Disputes over=20 transmission operation are to be expected during the transition to an = open=20 access regime. It is unlikely that filed tariffs and = service=20 contracts will cover all of the contingencies that turn out to be = relevant in a=20 market that has hardly existed before now. Beyond unforeseen events, = utilities=20 and wheeling customers may differ in their interpretaions of contract = terms. One=20 expects a disproportionate amount of litigation during a market's = formative=20 period because it is then that unforeseen opportunities and = unanticipated=20 conflicts will probably arise in the greatest numbers. Precedents set in = litigation give market participants knowledge of the operational = consequences of=20 their agreements and encourage them to contract more efficiently in the = future.=20 n126 Agents in any market can act opportunistically to avoid undesired = outcomes=20 or to seek gains beyond those expected from a contract. n127 Both owners = and=20 users of transmission have incentives to allocate benefits to themselves = and=20 shift costs onto others. Either side can initiate disputes that become=20 regulatory matters or lawsuits. Some disputes will be settled between = the=20 parties, and the settlements will become precedential, while regulators = or=20 courts will impose solutions in others. [*423] Antitrust can = be used=20 as a strategic tool to impede rather than foster competition, and treble = damages=20 may motivate private plaintiffs to turn contract disputes into antitrust = suits.=20 n128 If experience with gas pipelines is relevant, however, antitrust = may have a=20 minimal impact on the transition. Few pipeline antitrust cases reached = the=20 appellate courts, and those courts were often able to distinguish = efficiency=20 from opportunism. n129 Operational disputes arose between pipelines and=20 shippers, but few if any became claims that capacity was = anticompetitively=20 withheld. As its details became more transparent, open = access=20 facilitated new competition in interconnections, storage, market = centers, and=20 released capacity that further diminished the monopoly power of = pipelines. n130=20 Finally, there are no guarantees that the FERC and the courts will rule = in=20 accordance with an activist vision. Utility defendants often prevailed = in=20 pre-EPAct antitrust suits over transmission. As examples, in Borough of = Lansdale=20 v. Philadelphia Electric Company, the Third Circuit held that defendant = did not=20 have monopoly power in the relevant market because plaintiff could have = built=20 its own lines to the bulk power market. n131 In Town of Concord v. = Boston Edison=20 Company, the First Circuit made price squeeze cases more difficult by = requiring=20 a showing of actual competition between a utility and its wholesale = customer.=20 n132 In Cities of Anaheim, et al v. Southern California Edison Company = the Ninth=20 Circuit found a business justification for denial of = access to=20 fully-loaded transmission that would have allowed municipal systems to = purchase=20 Preference Power for themselves. n133 Defendant's offer of interruptible = service=20 on these lines was deemed reasonable because municipals had transmission = rights=20 that allowed them to import power from other regions. In a related case, = the=20 same court stated that the City of Vernon's request for similar=20 access that would only transfer benefits from the = transmission=20 owner to itself would "stand the essential facilities doctrine on its = head."=20 n134 It is not clear how EPAct and Order 888 might affect the antitrust = aspects=20 of these decisions. [*424]

VII. Conclusions=20

Premerger approval leaves the antitrust agencies and the = FERC=20 with the task of predicting when a proposed merger is likely to = significantly=20 reduce competition in a relevant market. Economics has not yet reached = the state=20 that it can make such a prediction with reasonable certainty, and the = costs of=20 error can be high for the economy. In electric utility mergers, = preapproval is=20 particularly thorny because markets are harder to define and quantify in = a=20 regulated industry than they are in an unregulated one. The=20 Guidelines are also hard to apply in electricity = because=20 technological and regulatory forces are dramatically changing the = environment as=20 the mergers go forward. There is good reason to expect that the markets = in which=20 the merged utilities will operate will be quite unlike markets of the = past, in=20 those cases where markets even existed.

A merger analysis must = look=20 forward, but with the knowledge that policies to ensure competitive = conditions=20 in today's markets will not necessarily do so in tomorrow's markets. = Merger=20 policy that operates under such uncertainty can best anticipate the = unknown by=20 first identifying the assets utilities can use to block new competition = and=20 determining how a merger will affect the gains from using them. Those = assets are=20 in transmission and not in generation. A merger is more likely to harm=20 competition because of its effects on transmission than its effects on=20 energy markets. Open access protects = both=20 competition in existing markets and competition to reshape markets by=20 introducing new products and trading institutions. The single most = relevant=20 market for both types of competition is the market for rights to use=20 transmission.

Despite the prominence usually given to statistics = of=20 energy market concentration in utility mergers, those = markets=20 are only relevant if relationships between concentration and price = assumed in=20 the Guidelines are valid in electricity. The=20 Guidelines are in reality a weak foundation for merger = policy,=20 both in electricity and elsewhere. The necessity of open = access=20 to foster the development of a competitive power industry has never been = in=20 doubt. The FERC should take seriously the possibility that open=20 access is also the only necessary policy.=20

FOOTNOTES:
n1. 15 U.S.C. 18(a) (1988). =

n2.=20 U.S. Department of Justice, Merger Guidelines, 47 Fed. = Reg.=20 38,493 reprinted in 4 Trade Reg. Rep. (CCH) P 13,102 (June 14, 1982)=20 [hereinafter 1982 Guidelines]; U.S. Department of = Justice=20 Merger Guidelines, 49 Fed. Reg. 26,823 reprinted in 4 = Trade=20 Reg. Rep. (CCH) P 13,103 (June 14, 1984) [hereinafter 1984=20 Guidelines]; U.S. Department of Justice Horizontal = Merger=20 Guidelines, 57 Fed. Reg. 41,552 reprinted in 4 Trade = Reg. Rep.=20 (CCH) P 13,104 (April 2, 1992) [hereinafter 1992 Guidelines]. = The=20 Guidelines also apply to mergers under Section 1 of the Sherman = Act (15=20 U.S.C. 1 (1988)) and Section 5 of the Federal Trade Commission Act (15 = U.S.C. 45=20 (1988)). See 1992 Guidelines 0.

n3. Order No. = 888,=20 Promoting Wholesale Competition Through Open Access=20 Non-discriminatory Transmission Services by Public Utilities and = Recovery of=20 Stranded Costs by Public Utilities and Transmitting Utilities, III = F.E.R.C.=20 Stats & Regs. P 31,036 (1996). [hereinafter Order No. 888]. =

n4. 16=20 U.S.C. 824 (b) (1994).

n5. Commonwealth Edison Company and = Central=20 Illinois Electric and Gas Company, 36 F.P.C. 927 (1966).

n6. = Inquiry=20 Concerning Commission's Merger Policy Under the Federal Power Act, 74 = F.E.R.C. P=20 61,082.

n7. See, e.g., Comments of the U.S. Department of = Justice, FERC=20 Docket No. RM96-6-000, filed May 7, 1996, particularly the Appendix; and = Comments of the American Public Power Association, Docket No. = RM96-6-000, filed=20 May 7, 1966. A broader summary of comments appears in Justice/FTC Merger = Guidelines Recommended as Basis for FERC Review, Inside = FERC,=20 May 13, 1996, at 5. Others have made the case that uncertainty about the = industry's transition necessitates using the Guidelines = as an=20 interim policy. Richard J. Pierce, Jr., Antitrust Policy in the New = Electricity=20 Industry, 17 Energy L.J. 29 (1996).

n8. FERC's = Massey=20 Hashes Out View of Mergers, Sees Two Focal Points, Electric Power Alert, = Dec.=20 20, 1995, at 34.

n9. Baltimore Gas and Elec. Co. and Potomac = Elec. Power=20 Co., 76 F.E.R.C. P 61,111, at 61,572 (1996); slip op. at 13. =

n10. Id.,=20 Commissioners Santa and Bailey, Dissenting, slip op. at 4.

n11.=20 Washington Water Power Co. and Sierra Pacific Power Co., 73 F.E.R.C. P = 61,218=20 (1995).

n12. Public Serv. Co. of Colo. and Southwestern Pub. = Serv. Co.,=20 75 F.E.R.C. P 61,325 (1996).

n13. Id. at 62,045.

n14. A = summary=20 appears in Robert J. Michaels & Arthur S. De Vany, Market-Based = Rates for=20 Interstate Gas Pipelines: The Relevant Market and the Real Market, 16=20 Energy L.J. 299, 312 (1995).

n15. See also, = Kansas City=20 Power & Light Co., 67 F.E.R.C. P 61,183 (1994), and Enron Power = Enterprises=20 Corp., 52 F.E.R.C. P 61,193 (1990).

n16. See also, Pacific Power = &=20 Light Co., 26 F.E.R.C. P 63,048 (1984).

n17. FERC, Market-Based = Rates=20 for Natural Gas Companies 38 (1995) (staff paper).

n18. See = also, Koch=20 Gateway Pipeline Co., 66 F.E.R.C. P 61,385 (1994).

n19. See = also, El=20 Paso Natural Gas Co., 49 F.E.R.C. P 61,473 (1989).

n20. See = also,=20 Opinion No. 360, Buckeye Pipe Line Co., L.P., 53 F.E.R.C. P 61,473 = (1990).=20

n21. Order No. 888, at 63. The FERC also considers whether = affiliate=20 abuse or reciprocal dealing can create barriers to entry that enhance = market=20 power in a merger.

n22. 1992 Guidelines 0.1. = For=20 overviews of market definition under the Guidelines, = see John=20 R. Morris & Gale R. Mosteller, Defining Markets for Merger Analysis, = 36=20 Antitrust Bull. 599 (1991); and Joseph J. Simons & Michael A. = Williams, The=20 Renaissance of Market Definition, 38 Antitrust Bull. 799 (1993). =

n23.=20 1992 Guidelines 1.11, quotations and parentheses in = original.=20

n24. Id. 1.11. The 1984 Guidelines 2.11 = specified a=20 period of one year.

n25. 1992 Guidelines 2.21.=20

n26. 1992 Guidelines 1.0.

n27. Id. =

n28.=20 Id. 1.0, 1.32.

n29. Id. 1.5. The Guidelines = multiply a=20 decimal HHI by 10,000.

n30. A five-firm market where one firm = has a=20 market share of .96 and the other four each have shares of .01 has an = HHI of .92=20 [or 9,200]; one where each has a .20 share has an HHI of .20 [or 2,000]. = In the=20 first case it may often be reasonable to view the large seller as a = monopolist=20 who can make decisions without concern for reaction by the fringe = producers.=20

n31. 1992 Guidelines 1.51.

n32. Id. = 1.51(b).=20

n33. Id. 1.51(c). An HHI of 1,800 occurs in a market with the = equivalent=20 of between five and six equal-size sellers.

n34. Id. 2-5. =

n35.=20 See also, the 1992 Guidelines 1.0 specify that the = analysis of=20 a merger's effects on market price is to be made "assuming the terms of = sale of=20 all other products are held constant."

n36. A "heuristic = theoretical=20 representation" of this market model appears in Robert D. Willig, Merger = Analysis, Industrial Organization Theory, and Merger=20 Guidelines, Brookings Pap. Econ. Activity = Microeconomics 281,=20 286 (1991). For further discussion, see Robert J. Michaels & Arthur = S. De=20 Vany, supra note 14, at 302.

N37. 1992 = Guidelines 2.0.=20 The Guidelines also consider a merged firm's power to=20 unilaterally meet the price increase threshold. Id. 2.2. The distinction = between=20 unilateral and coordinated action is unclear because under the=20 Guidelines, a firm with unilateral power over price = will be=20 sufficient in itself to constitute a relevant market. Lucile S. Keyes, = The=20 Horizontal Merger Guidelines of 1992, 10 Rev. Indust. = Org. 143,=20 151 (1995).

n38. 1992 Guidelines 2.1. See also, = Kevin=20 J. Arquit, The Boundaries of Horizontal Restraints: Facilitating = Practices and=20 Invitations to Collude, 6 Antitrust L.J. 531 (1993); The interpretation = of a=20 given restraint is often unclear. See also, Keith J. Crocker & = Thomas P.=20 Lyon, What Do "Facilitating Practices' Facilitate? An Empirical = Investigation of=20 Most-favored-Nation Clauses in Natural Gas Contracts, 37 J. Law & = Econ. 297=20 (1994).

n39. Gregory J. Werden, Horizontal Mergers: A Comment, = 81 Am.=20 Econ. Rev. 1002 (1991); George A. Hay & Gregory J. Werden, = Horizontal=20 Mergers: Law, Policy, and Economics, U.S. Dept. of Justice Antitrust = Division=20 Discussion Paper EAG 93-1, at 3 (1993). Views in these articles are not=20 necessarily those of the U.S. Department of Justice.

n40. Frank = Gollop=20 & Mark Roberts, Firm Interdependence in Oligopolistic Markets, 10 J. = Econometrics 313 (1979); and Gyoichi Iwata, Measurement of Conjectural = Variation=20 in Oligopoly, 42 Econometrica 947 (1974). In electricity, data from the = U.K.=20 pool are inconsistent with some versions of the theory. See Catherine D. = Wolfram, Measuring Duopoly Power in the British Electricity Spot Market, = Massachusetts Institute of Technology Dept. of Economics Working Paper = (1995).=20

n41. Dennis W. Carlton & Jeffrey M. Perloff, Modern = Industrial=20 Organization (2nd ed. 1994), ch. 5-8 and 10. For a statement that these = models=20 are seldom adequate to capture economically important aspects of = competition,=20 see Franklin M. Fisher, Games Economists Play: A Noncooperative View, 20 = Rand J.=20 Econ. 113 (1989).

n42. Others believe that the=20 Guidelines provide such easily satisfied criteria that = an=20 excessive number of anticompetitive mergers slip through their screen. = See=20 Keyes, supra note 37.

n43. "The Guidelines = reflect the=20 Congressional intent that merger enforcement should interdict = competitive=20 problems in their incipiency." 1984 Guidelines 1. =

n44.=20 William Landes & Richard Posner, Market Power in Antitrust Cases, 94 = Harv.=20 L. Rev. 937 (1981); and Oliver Williamson, Economies as an Antitrust = Defense, 58=20 Am. Econ. Rev. 18 (1968).

n45. The Guidelines = allow=20 flexibility on the percentage price increase depending on "the nature of = the=20 industry" but do not further describe the relevant attributes of that = nature.=20 1992 Guidelines 1.11.

n46. These collusive = equilibria=20 can be explicit or "tacit." The latter term, though frequently used, is=20 difficult to define. For example, is there a tacit collusion if a market = is=20 served by two sellers who have no restrictive agreement but have = independently=20 chosen not to compete aggressively?

N47. The sources of data do = not=20 necessarily correspond to relevant markets for antitrust analysis. For = example,=20 the frequently-used U.S. Census of Manufactures provides data on profit = margins,=20 concentration (four-firm measures rather than HHIs), and other variables = for=20 hundreds of Standard Industrial Classifications. These classifications = are=20 defined for statistical reporting purposes and will only by coincidence = also be=20 relevant antitrust markets. Cf, Sam Peltzman, The Gains and Losses from=20 Industrial Concentration, 20 J. L. & Econ. 229 (1977); F. M. = Scherer, The=20 Causes and Consequences of Rising Industrial Concentration: A Comment, = 22 J. L.=20 & Econ. 191 (1979).

n48. Paul A. Pautler & Robert P. = O'Quinn,=20 Recent Empirical Evidence on Mergers and Acquisitions, 38 Antitrust = Bull. 741,=20 768 (1993); Richard Schmalensee, Industrial Economics: An Overview, 98 = Econ. J.=20 643 (1988).

n49. See, e.g., William G. Shepherd, Reviving = Regulation -=20 and Antitrust, 7 Elec. J. 16 (1994); Samuel C. Thompson, Jr., A Proposal = for=20 Antitrust Merger Reform: Repudiating Judge Bork in Favor of Current = Economic=20 Learning, 41 Antitrust Bull. 79 (1996).

n50. Gregory J. Werden, = A Review=20 of the Empirical and Experimental Evidence on the Relationship between = Market=20 Structure and Performance, U.S. Department of Justice Antitrust Division = Economic Analysis Group Discussion Paper EAG 91-3 (1991). (This paper is = not a=20 statement of DOJ's official views.)

n51. Harold Demsetz, Two = Systems of=20 Belief about Monopoly, in Industrial Concentration: The New Learning 164 = (Harvey=20 Goldschmied et al. ed., 1974); John McGee, In Defense of Industrial=20 Concentration 95 (1971).

n52. More direct evidence on mergers = and price=20 is scarce. Pautler & O'Quinn, supra note 48, at 757, cite a handful = of=20 studies. The list is dominated by airline mergers that raised prices = after=20 substantially increasing concentration (e.g., at individual airports) = from=20 already-high levels.

n53. Concentration and Price (Leonard W. = Weiss ed.,=20 1989).

n54. Most of Weiss's other data sets showed no = statistically=20 significant relationship, as opposed to a negative one. Nearly all sets = were=20 subjected to several regression analyses that attempted to isolate the = effect of=20 seller concentration by standardizing for other variables known to = affect price.=20 The choice of variables to "hold constant" depends on theoretical = relevance and=20 data availability, and normal practice is to publish several alternative = formulations. Weiss somewhat arbitrarily classifies a data set as = showing a=20 positive association if at least one of these formulations yields a=20 significantly positive effect for concentration, regardless of what = other=20 formulations using the same data show. Id. at 266.

n55. See = citations in=20 Pautler & O'Quinn, supra note 48, at 772.

n56. Weiss, supra = note 53,=20 at 277.

n57. Noel D. Uri & Malcolm Coate, The Department of = Justice=20 Merger Guidelines: The Search for Empirical Support, 7 = Int'l=20 Rev. L. & Econ. 113 (1987).

n58. Statistics of prosecuted = collusions=20 point to their greater prevalence in declining industries rather than = growing=20 ones, and in industries where owners of (relatively small) businesses = rather=20 than employees make the decision to collude (e.g., highway contractors). = Peter=20 Asch & J.J. Seneca, Is Collusion Profitable, 68 Rev. Econ. & = Stats. 1=20 (1976); Jon M. Joyce, Effect of Firm Organizational Structure on = Incentives to=20 Engage in Price Fixing, 7 Contemp. Policy Issues 19 (1989).

n59. = William=20 F. Baxter, Antitrust Policy, in American Economic Policy in the 1980s = 600, 610=20 (Martin Feldstein ed., 1994). Baxter offered no corresponding = explanation of how=20 1,800 became the danger-level HHI or how 50 and 100 point increases in = it were=20 determined to be critical.

n60. Comments of the U.S. Department = of=20 Justice, supra note 7, at 8.

n61. Keith S. Watson & Thomas = W.=20 Brunner, Monopolization by Regulated "Monopolies': The Search for = Substantive=20 Standards, 22 Antitrust Bull. 559 (1977); Michaels & De Vany, supra = note 14,=20 at 309.

n62. 410 U.S. 366 (1973).

n63. The government's=20 calculation was unorthodox. Cities with populations ranging from 20 to = 15,000=20 each counted as one franchise. Numerous cooperatives not included in the = computation served areas that did not give official franchises, and the = three=20 largest cities in the territory were islands served by another corporate = utility. Otter Tail actually sold under 30 percent of the retail = kilowatts in=20 its territory. See Andrew N. Kleit & Robert J. Michaels, Antitrust,=20 Rent-Seeking, and Regulation: The Past and Future of Otter Tail, 39 = Antitrust=20 Bull. 689, 709 (1994).

n64. Watson & Brunner, supra note 61, = at 566.=20

n65. The government agreed with Otter Tail. Kleit & = Michaels, supra=20 note 63, at 708.

n66. Although defining markets for various = types of=20 bulk power and transmission is now standard procedure, it was not then. = Otter=20 Tail was active in a well-developed regional bulk market that no one = considered=20 relevant despite the desire of municipal utilities to trade in it. At = the time,=20 modestly-sized Otter Tail was the fourth largest wheeler in the country. = Kleit=20 & Michaels, supra note 63, at 699.

n67. U.S. v. Otter Tail = Power=20 Co., 331 F. Supp. 54, 60 (D. Minn. 1971).

n68. Kleit & = Michaels,=20 supra note 63, at 699.

n69. It is difficult to determine a = facility's=20 efficient size if market growth is expected or if there is uncertainty = about=20 future demand for the facility's services. Michaels & De Vany, supra = note=20 14, at 338 (discuss the likelihood that pipelines will be undersized to = capture=20 monopoly rents).

n70. A line's capacity at any instant will be=20 determined by both its engineering specifications and the state of the=20 surrounding system. The text abstracts from the important question of = defining=20 property rights when the underlying capacity of the line is changing, = and from a=20 discussion of markets for ancillary services that necessarily accompany=20 wheeling. On these topics, see William W. Hogan, Electric Transmission: = A New=20 Model for Old Principles, 16 Elec. J. 18 (Mar. 1993); Order No. 888, at = 198.=20

n71. Here again the text abstracts from technical problems in = operating=20 parallel lines.

n72. This description is incomplete regarding = entry of=20 new sellers. In a competitive market newcomers may have higher costs = than=20 existing sellers (e.g., because their lines must traverse rougher = terrain). In a=20 competitive market, however, existing sellers are unable to take actions = that=20 raise the costs of new ones.

n73. This usage differs from the = FERC's.=20 Order 888 authorizes short-term service on unused electrical capacity as = in the=20 text, but defines "use-it-or-lose-it" as meaning that the holder of a=20 temporarily unused capacity right will lose the right itself, a = provision absent=20 from the Order. See Order No. 888, at 166.

n74. The FERC staff's = analysis of market-based pipeline rates generally reaches pessimistic=20 conclusions using statistics of ownership. The FERC Staff Paper, supra = note 17,=20 at 38.

n75. Rodney T. Smith, Arthur S. De Vany, & Robert J.=20 Michaels, Defining a Right of Access to Interstate Gas=20 Pipelines, 8 Contemp. Policy Issues 142 (1990).

n76. 1992=20 Guidelines 3.0. The Guidelines' = emphasis on=20 "committed" entry that entails sunk costs appears inapplicable to = markets for=20 access rights.

n77. Michaels & De Vany, = supra note=20 14, at 330. The FERC recently issued a notice of proposed rulemaking to = remove=20 price caps on released capacity. See FERC Initiates Rulemaking to = Consider=20 Revisions to Capacity Release Program, Foster Natural Gas Report, at 1 = (Aug. 1,=20 1996).

n78. Paul W. MacAvoy, Michael Doane, & Michael = Williams,=20 Federal Energy Regulatory Commission Order No. 636 as = the=20 Penultimate Regulatory Reform of the Gas Industry (John M. Olin = Foundation=20 Working Paper, No. 38, 1995).

n79. The case against using = capacity and=20 energy markets for any purpose appears in Section VI = infra.=20

n80. That Order requires the reformulation of unnecessarily=20 discriminatory coordination and pooling contracts. Order No. 888, at = 177, 261.=20

n81. Order No. 888, at 279 sets down the FERC's criteria for = Independent=20 System Operators. Some economists believe that exchange institutions and = the ISO=20 cannot be separated without a loss of economic efficiency. See William = W. Hogan,=20 A Wholesale Pool Spot Market Must Be Administered by the Independent = System=20 Operator: Avoiding the Separation Fallacy, 8 Elec. J. 26 (Dec. 1995).=20

n82. Capacity Reservation Open Access = Transmission=20 Tariffs, Docket No. RM96-11-000, F.E.R.C. Stats. & Regs. P 32,517 = (1996);=20 Inquiry Concerning the Commission's Pricing Policy for Transmission = Services=20 Provided by Public Utilities Under the Federal Power Act, F.E.R.C. = Stats. &=20 Regs., Regs. Preamble P 31,005, 59 Fed. Reg. 55,031 (1994).

n83. = Transmission pricing is complicated by a lack of consensus about = efficient rate=20 designs and the administration of transmission allocations in the = presence of=20 real-time power flows. See Hogan, supra note 70; Shmuel Oren et al, = Nodal Prices=20 and Transmission Rights: A Critical Appraisal, 8 Elec. J. 24 (Apr. = 1995); IPALCO=20 Enterprises, Inc., A White Paper for Restructuring the Electric Utility = Industry=20 (1995).

n84. Order No. 888, at 38; Southern Company Services, = Inc., 75=20 F.E.R.C. P 61,130 (1996). Beyond access, the Commission = also=20 considers whether affiliate abuse or reciprocal dealing can enhance = market=20 power. Id.

n85. The FERC has previously treated rights as = relevant=20 commodities, e.g., franchise competition as rivalry for the right to = serve an=20 area. See Southern California Edison Company, 40 F.E.R.C. P 61,371 = (1987).=20

n86. 1984 Guidelines 3.3.

n87. Notice = of=20 Proposed Rulemaking, IV F.E.R.C. Stats. & Regs. P 32,514, at 33,088. =

n88. See Kleit & Michaels, supra note 63, at 697. =

n89. John=20 S. Moot, A New FERC Policy for Electric Utility Mergers? 17=20 Energy L.J. 139, 144 (1996). In other mergers and power = marketing dockets, the FERC has treated transmission as relevant, = sometimes=20 breaking it into short-term and long-term. See Utah Power and Light, 45 = F.E.R.C.=20 P 61,095, at 61,284 (1988); Entergy Systems, Inc., 53 F.E.R.C. P 61,234 = (1992).=20

n90. Moot, supra note 89, at 148-149.

n91. Baltimore Gas = and=20 Electric Company and Potomac Electric Power Company, Commissioners Santa = and=20 Baily, Dissenting, supra note 9, slip op. at 4.

n92. One author = finds=20 that existing transmission rates so constrain energy = markets=20 that almost any merger becomes questionable under the=20 Guidelines. Carmen D. Legato, Electric Mergers: = Transmission=20 Pricing, Market Size, and Effects on Competition, 134 Pub. Utils. Fort. = 23 (June=20 1, 1996).

n93. For a sample of possible offerings, see Shmuel = Oren &=20 S. Smith (Eds.), Service Opportunities for Electric Utilities: Creating=20 Differentiated Products (1993).

n94. Robert D. Glynn Jr., = Offering=20 Customers Direct Access: Using Choice to Stimulate = Competition,=20 7 Elec. J. 52 (Dec. 1994); Robert J. Michaels, Wholesale Pooling: The=20 Monopolist's New Clothes, 7 Elec. J. 64 (Dec. 1994).

n95. = Comments of=20 the U.S. Department of Justice, F.E.R.C. Docket No. RM96-6-000, Appendix = at A5.=20

n96. Edward Kahn, Regulation by Simulation: The Role of = Production Cost=20 Models in Electricity Planning and Pricing, 43 Operations Res. 388, 395 = (1995).=20

n97. Baltimore Gas and Electric Company and Potomac Electric = Power=20 Company, supra note 9, slip op. at 12.

n98. Order No. 888, at = 63.=20

n99. The law's emphasis on incipient monopoly may require that = the=20 authorities weigh a dollar lost through a monopolistic merger more = heavily than=20 to a dollar saved through an efficient one. See note 43 supra. In its = rulings on=20 mergers, the FERC should specify (to the extent possible) the relative = weights=20 it is placing on the two outcomes.

n100. The text abstracts from = the=20 cost of performing and evaluating the additional studies. The costs of = delaying=20 a worthwhile merger probably outweigh the expenses on expert work by = applicants,=20 intervenors, and Commission staff.

n101. The text is a capsule = summary=20 of a complex process. See R. Duncan Luce & H. Raiffa, Games and = Decisions=20 (1957); Raiffa, Decision Analysis: Introductory Lectures on Choice Under = Uncertainty (1968).

n102. It is not clear how the Commission = should=20 weigh the absence in some recent proceedings of intervenors with = specific=20 concerns about prices, e.g., Baltimore Gas and Electric Company and = Potomac=20 Electric Power Company, Commissioners Santa and Baily, Dissenting, supra = note=20 10, slip op. at 4; Washington Water Power Company and Sierra Pacific = Power=20 Company, 73 F.E.R.C. P 61,218 (1995).

n103. Research using = utility and=20 market data will require ingenuity and rigor, and economists will = probably=20 differ on research methods and interpretation of results. Such research, = however, adds to useful knowledge at costs that are small when compared = with the=20 loss if the FERC decides a merger incorrectly. As examples of such work, = not=20 directly applicable to mergers, see Jan Paul Acton & Stanley M. = Besen,=20 Assessing the Effects of Bulk Power Rate Regulation: Results from a = Market=20 Experiment, 19 Appl. Econ. 663 (1987); Douglas Gegax & John = Tschirhart, An=20 Analysis of Interfirm Cooperation: Theory and Evidence from Electric = Power=20 Pools, 50 Southern Econ. J. 1077 (1984).

n104. As an analogy, = airline=20 route awards under Civil Aeronautics Board regulation were determined in = part by=20 politics, leaving the typical airline with an economically irrational = grid that=20 it could not sustain if required to compete. Airlines responded to = deregulation=20 by: [1] merging where route systems complemented one another; [2] = entering new=20 city-pair markets where they were more efficient than incumbents; and = [3]=20 altering established schedules to hub-and-spokes patterns.

n105. = DOJ,=20 May 7 Comments, RM 96-6, Appendix at A5.

n106. Lewis J. Perl, = Measuring=20 Market Power in Electric Generation, 64 Antitrust L.J. 311 (1996). =

n107.=20 The uncertainty of future market prices gives another reason to question = the=20 applicability of the Guidelines. What is the = significance of a=20 five percent increase when today's prices are set by a mix of = competition and=20 regulation, in markets whose institutions are still developing? =

n108.=20 Michaels & De Vany, supra note 14, at 328.

n109. DOJ, May 7=20 Comments, RM 96-6 at 11; John W. Wilson, Merger Policy=20 Guidelines for the Electric Power Industry, 9 Elec. J. = 14=20 (Jan./Feb. 1996).

n110. This matter is discussed further in = Section VI=20 infra.

n111. 1992 Merger Guidelines 2. =

n112.=20 See, e.g., Comments of the Staff of the Bureau of Economics of the FTC, = F.E.R.C.=20 Docket Nos. RM95-8-000 and RM94-7-001 (Aug. 7, 1995), at 3 and 8. =

n113.=20 Stephen Littlechild, Competition, Monopoly, and Regulation in the = Electricity=20 Industry, in Michael Einhorn (Ed.), From Regulation to Competition: New=20 Frontiers in Electricity Markets 125 (1994); Richard J. Green & = David M.=20 Newbery, Competition in the British Electricity Spot Market, 100 J. = Polit. Econ.=20 929 (1992); David M. Newbery, Power Markets and Market Power, 16=20 Energy J. 39 (1995); Frank A. Wolak & Robert H. = Patrick,=20 The Impact of Market Rules and Market Structure on the Price = Determination=20 Process in the England and Wales Electricity Market (1996). All = generators=20 bidding into the pool receive the same price as the marginal generator.=20

n114. In Great Britain only an energy price can = be bid=20 into the pool, leaving the question of capacity payoff. Plant owners = without=20 power purchase commitments may need to bid more than marginal cost if = they hope=20 to recover their investments. The pool energy rate = includes an=20 adder that is paid when capacity is short, but in practice, plants are = not=20 financed by gambling on the adder. See Michaels, supra note 94. =

n115.=20 Dan W. York, Competitive Electricity Markets in Practice: Experience = from=20 Norway, 7 Elec. J. 48 (June 1994); Jan Moen & Jan Hamrin, Regulation = and=20 Competition Without Privatization: Norway's Experience, 9 Elec J. 37 = (Mar.=20 1996). Nearly all generation in Norway is hydroelectric, a technology = for which=20 marginal costs are particularly hard to conceptualize.

n116. R. = Peter=20 Lalor & Hernan Garcia, Reshaping Power Markets: Lessons from South = America,=20 9 Elec. J. 63 (Mar. 1996). Chile's Interconnected Central System = accounted for=20 87 percent of that country's 1988 power production and had an HHI of = 3,400. The=20 HHI for generation into the U.K. pool in 1992-93 was 2,900. Both of = these mixes=20 include plants which must run independent of prices [e.g., nuclear] and = plants=20 which are only conditionally available [e.g., hydroelectric]. Pablo T. = Spiller=20 & Luis Viana Martorell, How Should It Be Done? Electricity = Regulation in=20 Argentina, Brazil, Uruguay, and Chile, in Richard J. Gilbert & = Edward P.=20 Kahn (Eds.), International Comparisons of Electricity Regulation 82, 116 = (1996);=20 Stephen Littlechild, Competition, Monopoly and Regulation in the UK = Electricity=20 Industry 4 (1993).

n117. William G. Shepherd, Applying Antitrust = to=20 Mergers in the Electricity Industry, Appendix A to Joint Petition of the = American Public Power Association and the National Rural Electric = Cooperative=20 Association for a Rulemaking Proceeding to Revise the Commission's = Standards=20 Applicable to the Merger of Public Utilities under 203 of the Federal = Power Act=20 (Jan. 17, 1996).

n118. Id. at 8. His examples of dominance = include Kodak=20 (which has exited mass-market cameras and lost its dominance in film), = IBM=20 (whose former dominance in mainframes did not reach personal computers), = and=20 Xerox (inventor of the copying process, but now a minor player in the = industry).=20 Dominant at one time, such firms can equally be viewed as examples of = how=20 competitive markets function in reality. Each started small, grew with = the help=20 of innovation, and lost its position to a new wave of innovators. =

n119.=20 Wilson, supra note 109, at 16. His citations on the relevance of seller=20 concentration cover only unregulated industries.

n120. Mark W. = Frankena=20 & Bruce M. Owen, Electric Utility Mergers: Principles of Antitrust = Analysis=20 (1994).

n121. Some analysts also believe that antitrust may be = helpful=20 in attaining goals usually considered beyond economic analysis such as=20 "fairness" and "freedom of choice." See, e.g., Shepherd, supra note 117, = at 6.=20 Likewise, economics may not subsume Wilson's concern that regulators = "cannot=20 adequately control the strategic power that these merged enterprises = will still=20 enjoy in dealing with would-be rivals." Wilson, supra note 109, at 16.=20

n122. Shepherd, supra note 117, at 23.

n123. Mark W. = Frankena,=20 Background for the 1996 Debate on FERC's Electric Utility Merger Policy, = 10=20 (1996).

n124. Order No. 888, at 128. Appendix C contains = summaries of=20 some incidents that have appeared in merger testimonies. In some cases = (e.g.,=20 App. C at 13, Item 5), the utility's actions were found consistent with = its=20 wholesale contracts and the antitrust laws.

n125. The ISO, = however, will=20 likely be designed in part by the transmission owners, introducing the=20 opportunity for manipulation because of specialized knowledge at the = design=20 rather than operational level. See IPPs, Power Marketers Vote No On PJM=20 Transmission Proposal, Energy Daily, Aug. 22, 1996.=20

n126. See, e.g., Robert Cooter & Thomas Ulen, Law and = Economics 492=20 (1988); and George Priest, The Common Law Process and the Selection of = Efficient=20 Rules, 6 J. Legal Stud. 65 (1977).

n127. Benjamin Klein, Robert = G.=20 Crawford, & Armen A. Alchian, Vertical Integration, Appropriable = Rents, and=20 the Competitive Contracting Process, 21 J. L. & Econ. 297 (1978); = Oliver=20 Williamson, Economic Organization 101, 197 (1986).

n128. William = J.=20 Baumol & Janusz A. Ordover, Use of Antitrust to Subvert Competition, = 28 J.=20 L. & Econ. 247 (1985).

n129. In State of Illinois ex. rel. = Burris v.=20 Panhandle Eastern Pipeline Company, 935 F.2d 1469 (7th Cir. 1991), = plaintiff was=20 denied the right to purchase gas at market prices in lieu of supplies = that it=20 had contracted for with defendant at the high prices that prevailed in = the early=20 1980s. In City of Chanute et al v. Williams Natural Gas Company, 955 = F.2d 641=20 (10th Cir. 1992), the court found that defendant had a business = justification=20 for closing its pipeline to interim open access, and = that an=20 offer to sell gas at prices approved by the FERC constituted reasonable=20 access.

n130. See Michaels & De Vany, supra = note=20 14, at 320.

n131. Borough of Lansdale v. Philadelphia Electric = Co., 692=20 F.2d 307 (3rd Cir. 1982).

n132. Town of Concord, Mass. v. Boston = Edison=20 Co., 915 F.2d 17 (1st Cir. 1990).

n133. Cities of Anaheim et al = v.=20 Southern California Edison Company, 955 F.2d 1373 (9th Cir. 1992). This = case=20 (and City of Vernon infra note 134) litigated alleged incidents of = transmission=20 discrimination described in Order No. 888, App. C.

n134. City of = Vernon=20 v. Southern California Edison Company, 955 F.2d 1361, 1367 (9th Cir. = 1992).=20

	Home =20 S= ources =20 How=20 Do I? Site=20 Map What's = New Help =20