The merger voltage (I) is rising on the electric grid, but it remains to be seen which will win out: current (E) policy or resistance (R) to it. The textbook formula, I equals E over R, won't help you find the answer to this question.
In its March 1995 Notice of Proposed Rulemaking (NOPR),1 the Federal Energy Regulatory Commission (FERC) accepted two basic assumptions: 1) that in the wholesale market, electric generation is fully competitive (at least for new, as-yet unbuilt plants), and 2) that an open-access regime for electric transmission will eliminate unfair competition in retail markets. But there is room for doubt, since the second theory rests by necessity on the first, which itself has yet to be proven.
That is because the "wholesale market" that many, including Professor Joskow at MIT2 have been looking at, is only the market for the sale of bulk-power supply by one utility bulk-power supplier to another, or by a nonutility generator (NUG) to a bulk supplier. Few have thought to consider the purchase options available to a retail distributor of power not also in the bulk-power supply business, even though this market is by far the largest part of the wholesale supply of electric power.
Consider the distributor that is generation-dependent. This
distributor requires a product suitable for the retail market (em a product that we can define here as firm requirements power ("RQ" power).3 But on one hand, those adjacent bulk-power suppliers that are able to sell RQ power at wholesale that such a retailer might use may fear retaliation by their neighbor (em especially if it's a very big neighbor. On the other hand, NUGs that can't wait to compete may not be able to sell RQ power usable by the retailer.4 A retail distributor without generation can only use RQ power. A municipality served at retail by a vertically integrated supplier can municipalize only if it can call on a source of RQ power that is priced competitively. Other types of power won't do.
In the Louisiana market (a market I have studied) some 85 percent of all kilowatt hours are supplied to distribution systems that can use only RQ power. The nationwide figure is likely comparable. Given these facts, it is important to look at horizontal concentrations of market power in the RQ market to ensure that the only result of a merger is not to enhance the ability to engage in unfair competition in the newly "competitive" market.5
It now appears that Commissioner Massey may have the same concern.
That doubts are rising, even before the NOPR is finalized, can be seen in recent discussions by Commissioners Massey, Hoecker, and others, suggesting a new standard for approval of mergers at the FERC. On October 12, 1995, in a speech presented at the Edison Electric Institute Fall Legal Conference, Commissioner Massey raised red flags several times on the dangers of too much concentration of generating assets:
"One solution [to generation concentration] . . . might be to ensure that consolidation of the grid assets does in fact produce a competitive gain through more efficient access. . . . In the Open-Access NOPR, for example, some commentators argue forcefully that the FERC should move beyond mere functional unbundling and should require operational unbundling.
* * *
Does this idea also have merit in the merger context, as a way to ensure that the transmission market power of the
larger, vertically integrated is fully mitigated, or as a remedy for possibly anticompetitive generation concentrations?
* * *
But what if vertical disaggregation creates incentives for horizontal consolidation, at least in [generation], the industry's most costly segment? Will we end up with more or less competition? Will we really be better off? Is the answer to allow vertical disaggregation but discourage horizontal consolidation of supply?"
This concern, of course, is nothing new. Regulators, judges, and academics have for years debated over: 1) the optimum size of generating units, 2) the optimum size of integrated systems of generating units, and 3) the optimum size of the companies that own such units and systems. The early debates culminated in passage of the Public Utility Act of 1935, which included the provisions we now identify as the Federal Power Act and the Public Utility Holding Company Act (PUHCA).
Even now, when the tide appears to be turning toward repeal of PUHCA, there is much to be learned from the judicial and legislative history that underlies enactment and major interpretations of the '35 Act (em from FDR to NOPR.
Concentrated Generation: Early Concerns
The 1935 Act reflected a concern over size for the sake of size. The Act's framers saw their work as essential to achieving economies and efficiencies. In their time, they were familiar with academic work by noted thinkers such as Gifford Pinchot and Herbert Hoover, writing on "Superpower" and "Giantpower," which showed that substantial scale economies could be derived from large-scale, multiplant firms that could be achieved without degrading reliability. But framers of the '35 Act were concerned that bigger would not be better unless the evils of concentration were to be offset by economies that could not be achieved in any other way.
Franklin Roosevelt, who pushed hard for the '35 Act, decried unnecessary bigness in electric power as "private socialism." Howard Trienens, formerly general counsel of AT&T (before the divestiture), summarized the view of the framers of the act in his aptly named 1949 law review article: "The Utilities Act as a Solution to Sherman Act Problems."6 Trienens also refers readers to several other views showing the temper of the times, such as: "The Curse of Bigness," by Justice Brandeis, representing the final report of the TNEC (Temporary National Economic Committee), and the famous dissent by Justice William O. Douglass in U.S. v. Columbia Steel Co., a case in which the government failed in efforts to enjoin U.S. Steel Corp. from purchasing the assets of the largest independent steel fabricator on the West Coast.7
In his message to the Congress recommending the passage of the 1935 Act, Roosevelt declared:
"It is time to make an effort to reverse that progress of concentration of power which has made most American citizens, once traditionally independent owners of their own business, hopelessly dependent for their daily bread on the power of a few, who, by devices such as holding companies have taken for themselves unwarranted economic power."
The same goal of utility policy Roosevelt outlined as the desired result of the '35 Act was at that time also a goal of antitrust policy. The much respected Judge Learned Hand put it best when he said in a famous antitrust opinion:
"It is possible because of its indirect social or moral effect, to prefer a system of small producers, each
dependent for his success upon his own skill and character, to one in which the great mass of those engaged must accept the direction of a few."8
Unit Size; Company Size
In the early 1930s the maximum size of a baseload generating unit for even the very largest of electric generating systems was smaller than 50 megawatts (Mw), but most knowledgeable engineers thought that unit-scale economies increased in individual Carnot cycle units to about 1,000 to 1,500 Mw. Today, given experience gleaned from the '70s and '80s, we understand that outages of such large units adversely affect their economy. Now, the consensus greatly reduces optimal baseload unit size (em to about 500 Mw for coal-fired steam turbines, or even down to 150-200 Mw for combined-cycle units, if a long-term gas contract is available at current gas prices, which, of course, is not now the case.9
But in 1935, those who had mastered the fundamental engineering constraints that drive the economics of the electric utility
industry appreciated the key but little-understood complex rela- tionship between 1) unit size, 2) system size, 3) reserves, and 4) reliability of firm power supply. They sought to develop a regulatory environment in which utility systems could achieve that ideal company size that would permit installation of baseload units larger than 50 Mw to a size perhaps even as large as 1,000 to 1,500 Mw, without the necessity of maintaining so large a percentage of reserves to maintain adequate reliability that would offset the scale economies of the larger units.
It is really surprising to consider that after the Great Depression, when electric utilities started again to install new electric generating units in the late '30s, the largest unit size that could be installed by even the largest utility companies or holding company systems was only 50 or 60 Mw. It wasn't until 1948 that some of the very largest private companies commenced installing fossil-fueled steam units of 100 Mw or more. Most of the larger private companies didn't reach this size until the middle or late '50s or early '60s.10
By the late 1970s and 1980s, the larger systems had grown so large as to have arrived at the point where they could install 1000-Mw units and maintain only 20 percent reserves or less for adequate reliability. When the FERC approved the merger of Gulf States Utilities with Entergy, each of the two merger partners was big enough to install 1,000-Mw base-load units or larger, and yet maintain only about 18 percent or less in reserves for backup. One reason for this ability to trim reserves was the tremendous growth in system load between 1935 and 1995, first because of the onset of World War II, and later from the long era of economic prosperity, which many credit to the Marshall Plan. Until 1973 annual load growth averaged 7 percent per year.
Another reason for lowered reserve requirements evolved from a changing regulatory environment. In the middle '60s, under Section 202(b) of the Federal Power Act, the Federal Power Commission required large utilities to "share reserves" with small systems.11 In the early '70s under Section 105c of the Atomic Energy Act, the Nuclear Regulatory Commission ordered many large systems that were applying for nuclear licenses not only to wheel power and share reserves, but also to coordinate the development of base-load units with small systems.12 The 1964 National Power Survey encouraged voluntary power pooling between and among middle-sized companies to carry out the same arrangements. The compulsory institutional arrangements permitted smaller companies to obtain the benefits of large-scale, baseload units without offsetting reserve penalties. Middle-sized companies would voluntarily engage in the same activities with other companies of about the same size since they would more equally share in the incremental cost savings and
reliability benefits of those new arrangements.
Today, therefore, with a regulatory environment encouraging and compelling institutional arrangements by which much smaller and middle-sized companies can install optimal unit sizes, is there any longer a reason to suffer the curse of bigness (em and its concomitant market power to engage in unfair competition (em in order to obtain scale economies?
Mergers and Markets
These days, the U.S. Department of Justice gives short shrift to the social and political policies favoring small-scale enterprise. The Department of Justice now considers the only goal of the antitrust laws to be economic efficiency in production and distribution. And FERC policy in decided cases has virtually dispensed with the idea that too much concentration in generation may disrupt competition in wholesale markets.
In 1994, in Kansas City Power & Light Co., the FERC approved
market-based pricing for bulk-power sales, presuming that fair competition in long-run generation markets was inevitable: "[F]or sales from new (unbuilt) generating capacity there is no need . . . to focus on whether the seller has market power in generation."13 And, when it comes to mergers, the FERC asks the opponents of
a merger to show a detriment, rather than requiring its proponent to show that the merger
was vital to achieving scale economies.14
What then, must competitors prove to show that market
concentration in generation disrupts competition? What then, is a merger "consistent with the public interest"?
Antitrust lawyers generally agree that exercise of approval authority over mergers represents one of the two best opportunities to curb market power in electric generation.15 They also agree that opportunities to obtain scale economies should be taken into account.16 How should these seemingly conflicting goals be reconciled?
The most recent standard the FERC has used under section 203 of the Federal Power Act was applied in the GSU-Entergy merger case.17 That standard was, in my humble view, far from what was intended by the framers of the 1935 Public Utility Act. This recent standard does not require a showing of benefit to the public from the merger, only that it is "consistent with the public interest."18 This standard places a burden on the opponents of merger to show a detriment to the public interest, whereas the premise of the 1935 Act was that "bigger is badder," unless offset by added scale economies available only from horizontal integration of electric utility systems.19
In fact, in the first cases it tried under section 203, the Federal Power Commission (predecessor to the FERC) actually applied a standard that required proof of scale economies from system integration, but was reversed in the courts by the Ninth Circuit.20 It does not appear from the decision that the Ninth Circuit was apprized of the relationship of unit size, system size, reserves, and reliability that was the raison d'etre for the '35 Act's basic concept of justifiable scale (em that federal policies should permit the formation and continued existence only of those holding companies whose operating properties were integrated through their own transmission, because only such companies could achieve scale economies.
Nevertheless, hope springs eternal. In late November, perhaps mindful of the warnings from Commissioner Massey, the FERC showed indications of moving back to a standard that would require merger proponents to prove scale economies from system integration. In a preliminary hearing in the proposed Altus merger between The Washington Water Power Co. and Sierra Pacific Power Co., the FERC declined to issue approval and instead granted motions to intervene and consolidate the proceedings for hearing and decision. In its draft opinion, the FERC noted that the merger appeared to achieve no fuel savings related to integrated operations and central dispatch. It added, in fact, that because of their geographic separation, the two companies "will be operated as two separate electric control areas and will not be jointly dispatched as a single integrated system."21
The Antitrust Analysis
The dangers of market concentration in wholesale power demand closer attention by regulators. At the federal level, the FERC could start applying sound antitrust policy by investigating those engineering and economic factors it is uniquely qualified to determine: optimal baseload unit size, the amount of reliability required for commercially acceptable firm power (measured in terms of the "loss of load" or "loss of capacity" probability methods), and how low a reserve percentage must be maintained to be price competitive. With these parameters and an estimate of annual load growth, it could determine the optimal system size absent institutional arrangements such as "reserve sharing" or the "coordinated development of base-load units." Then it could downsize optimal company size to take into account the availability of those institutional arrangements. The affidavit of Mr. Sylvan Richard, which was submitted to the commission by the Louisiana Energy & Power Authority in support of its Reply Comments in the Mega-NOPR rulemaking (FERC Docket No. RM95-8, RM94-7) illustrates the first part of this computation.
A company whose utility system has no access to reserve sharing or to the coordinated development of baseload units and wants to compete in price and reliability must, in today's market, install 500-Mw, coal-fired steam turbines as its baseload units. Since such a system must maintain installed reserves at least equal to its single largest unit, it is easy to calculate that the smallest system able to install 500-Mw units, (where reserves are only 20 percent of its capacity) is one with a load of 2,500 Mw. But this would hold true only for a static system (em one not growing. All systems do experience growth in load.
When a company with a
2500-Mw annual peak load installs a new 500-Mw unit, it will not be able to use all the additional capacity for several years unless its load growth is 20 percent per year. Twenty-percent load growth is far greater than load growth has been since the oil crisis of 1973. (Even greater than it was before 1973 when load growth averaged 7 percent). Load growth of 5 percent per year on a 2,500-Mw system would require only an additional 125 Mw of new capacity per year, and would take 4 years to reduce reserves back down to 20 percent after adding a new 500-Mw unit. But a system of 10,000 Mw growing at 5 percent could use all the capacity of a new 500-Mw unit in a single year. No excess reserves need be maintained.
Knowing the optimal size baseload generating unit, the system size and its load growth will enable the FERC to determine the size of a company that even without access to reserve sharing and coordinated development of
baseload units could install
competitive-sized baseload units and maintain competitive reliability with reserves at a competitively low level. Then the FERC could factor in whether the institutional arrangements of reserve sharing and coordinated development were available, and reduce the optimal company size if and to the extent they were available.
Applying this standard would also facilitate the FERC's goal of fostering honestly industrial competition in wholesale electric bulk-power markets and avoid the predatory competition that otherwise would run rampant where some systems were far larger than needed to attain scale economies. For example, the Entergy merger created a system with $5 to $6 billion in annual revenues and some $25 billion in investment, a system far greater than needed to obtain scale economies. The major claimed economies of the merger were those from central economic dispatch of the combined Entergy-GSU system, which is a relatively simple kind of coordination readily achievable by contract between two entirely independent systems.
Roosevelt and others who fought fiercely for the 1935 Utility Act would be surprised to find that it (em drafted on the premise that systems larger than needed for scale economies should be dissolved (em is now interpreted to place the burden on the small-scale competitor to establish that proposition. What was mandated in 1935 was a rule that requires those who would concentrate
assets in only a few hands to prove that such increased company size is vital to achieving scale economies.
Measuring Market Power
Utility bulk-power suppliers sell to two classes of customers: other utility bulk-power suppliers and those only in the retail distribution business.
Bulk-power suppliers will normally operate a "control area" by means of a "control center" where dispatchers, ensure 'round the clock that generation equals load.22 When one source of power becomes unavailable, they swiftly become aware of it from the meters and other "remote monitoring" devices at the center and can promptly switch to another source using "remote supervisory" equipment.
On the other hand, a system that is engaged only in retail distribution operates no such facilities or personnel. For these companies, the first indication of a problem will come when an employee notices that his system has gone black. Those in charge of the system may be home in bed when the outage occurs. All they can do is call their bulk-power supplier. If they are a partial requirements customer, they can start up their own local generation to carry at least some of the load center's more vital loads, such as hospitals, water pumping for fire hydrants, police stations, and the like, until the primary source of power again becomes available. If the outage occurs offpeak, the distributor may be able to carry the entire load, but only after it has mobilized the necessary resources to take action. This step may take two or three hours. If it does, the primary source will likely be back in service by then.
While a bulk-power supplier
can use the firm, full-, or partial-requirements power sold to a retail distributor, the distributor cannot use the various kinds
of nonfirm power sold to a
bulk-power supplier by other bulk-power suppliers or NUGs.
These facts led the FERC in 1979 to classify electric product markets into two categories: 1) the power exchange or coordination services market, and 2) the firm, full-, or partial-requirements market.23
These two markets ordinarily would reveal possible adverse impacts from a merger, except that if only bulk-power supply or NUG interests are involved, one would examine only the power exchange market and its various submarkets.24 On the other hand, if a retail distributor complained, then the firm requirements market (full or partial) should also be scrutinized for adverse impacts on competition. This market would become a "relevant market" in deciding the controversy.
The geographic market also plays an important role in identifying adverse merger effects, especially the distance involved in determining whom it is practical to turn to for an alternate power supply. Reliability of supply is a function of distance. Power exchange transactions can take place over long distances because a bulk-power supplier experiencing difficulty in receiving power from one source and operating a control center can easily switch from that source to another. The retail distributor needs complete assurance of supply and will want the facilities of his supplier to be located within the same control area in which load is located, or very close by.
Most retail distribution markets are captive markets. The FERC has recognized that these markets must be included in market share,25 for the reasons stated by Judge Learned Hand in the famous Alcoa decision.26 Since most markets for the purchase of full-requirements power are captive markets, this recognition of captive markets should have an important effect on measuring concentration of market power.
In a gasoline market, would you lower your price if you knew another very large seller would retaliate with an even greater price cut? Maybe, or maybe not. Studying competitive activity could demonstrate whether two putative sellers in a market are actually competitors or part of an oligopoly. Will one electric utility invade the service area of another or, even more aggressively, make an offer to serve an existing RQ power customer of that neighbor at a price that comes in well under its neighbor's price? One must look to see if it ever has.
The Tools to Act
With a way to ensure that companies can reach optimal scale, and a way to measure concentration in all of the relevant power markets, the FERC has the tools it needs to set standards for mergers consistent both with the original intent of the 1935 Utility Act and also with antitrust policy and the policy embodied in the Commission's Mega-NOPR. t
Attorney Wallace Edward Brand is a sole practitioner in Washington, DC, where he represents small electric systems. From 1963 to 1970 he worked as a trial attorney at the Federal Power Commission, where he represented the FPC in the merger of Commonwealth Edison Co. and Central Illinois Electric & Gas Co. At the U.S. Justice Dept., Antitrust Division, in the early '70s, he tried the precedent cases on nuclear applicants' obligations under the Atomic Energy Act to wheel power, share reserves, and engage in developmental coordination. Mr. Brand is grateful for the assistance of Milton J. Grossman, Esq., former assistant section chief for electric utility regulation at the Antitrust Division in preparing this article.
1. Promoting Wholesale Competition Through Open-Access Non-Discriminatory Trans. Servs. by Pub. Utils., Dkt. RM95-8-000, and Recovery of Stranded Costs by Pub. Utils. and Transmitting Utils., Dkt. RM94-7-001, March 29, 1995, 70 FERC (pp 61,357.
2. See, affidavit of Professor Joskow, attached as Appendix A to the Edison Electric Institute's Mega-NOPR Comments, and those of Professor Shepherd and Mr. Richard, attached to the Initial & Reply Comments of the Louisiana Energy & Power Authority.
3. RQ power is firm full-requirements or partial-requirements power. The energy associated with that power is designated "RQ" energy in FERC Form 1.
4. No NUG has as yet supplied RQ power to a retail electric distributor not currently in the bulk-power supply business. At present it is only a hypothesis of economists yet to be proved, that commercially acceptable RQ power can be marketed by NUGs.
5. Fair competition eliminates the inefficient. Unfair competition is aimed at a target company that could be a very efficient but small recent entrant to the market.
6. 44 Ill.L.Rev. 331 (July, 1949).
7. 334 U.s. 495, 534 (1948).
8. U.S. v. Aluminum Co. of America, 148 F.2d 416, 427 (2d Cir.1945). This opinion has the binding effect of the Supreme Court. Judge Hand was designated in a special Act of Congress to hear the case, which was to have gone to the Supreme Court but could not because too many of the Justices had recused themselves and the high court could not assemble a quorum.
9. There is some opinion that small fluidized-bed units can serve as based-load units, but the near bankruptcy of a small utility in Texas trying to get its fluidizedbed unit to operate properly suggests to others that the technology may not be mature.
10. Evidence of unit sizes in this early period is hard to come by. I have recently seen one source suggesting that the Brookland Edison Co. had installed 200-Mva General Electric units as early as 1934. See, Dawes, Electrical Engineering, vol. II, McGraw-Hill, New York (1934), p. 2, footnote 1.
11. See, Gainesville Utils. Dept. v. FPC, 402 U.S. 515 (1971).
12. Consumer Power Co., Midland Units 1 & 2, 6 NRC 892 (1977).
13. 67 FERC (pp 61,183, May 13, 1994. The FERC did, however, make an exception to the case where barriers to entry could be shown. Below we show such barriers to entry into the RQ market.
14. See, Midwest Power Systems, Inc., 71 FERC (pp 61,386, June 22, 1995 (Massey and Hoecker concurring), describing the FERC's current "public interest" standard for mergers, as set out back in 1996 in Commonwealth Edison Co., 36 FPC 927, Aff'd sub nom. Utility Users League v. FPC, 394 F.2d 16 (7th Cir.), cert. denied, 393 U.S. 953 (1968).
15. See, II Areeda & Turner, (sc 406.
16. Id., at (sc 408.
17. Docket No. EC92-21, ER92-806, 62 FERC (pp 61,073. (In this case the author represented the Louisiana Energy & Power Authority, an opponent of merger between GSU and Entergy).
18. 62 FERC (pp 61,073 at 61,370.
19. Since there was no open access in those days, integration required direct connection of all the generation and load properties by transmission controlled by their owner. Only in this way could be owner of the properties be assured that he could achieve scale economies available because of the relationship of unit size, combined system size, reserves, and reliability.
20. Pacific Power & Light v. FPC, 111 F.2d 1014 (9th Cir.1940).
21. Dkt. Nos. EC94-23-000, ER95-808-000, Nov. 29, 1995, draft order at 12.
22. Sometimes two or more bulk-power suppliers will share the use of a control center to maintain a common control area.
23. Florida Power & Light Co., Opin. No. 57, 8 FERC (pp 61,454 (1979).
24. There a host of different types of power in this market. These are not interchangeable inter se, but are normally sold together as schedules of an interchange agreement and therefore the combination reflects commercial realities. Market analysts usually classify this type of market as a "cluster market." See, e.g. U.S. v. Philadelphia National Bank, 374 U.S. 321, 356 (1963); U.s. v. Grinell Corp., 384 U.S. 563, 572 (1966). If any two or three of these submarkets are involved, and the RQ or full- and partial-requirements market is not involved, it would be perfectly correct to refer to each of them as markets rather than submarkets, as the FERC did in the Northeast Utility merger. Opin. No. 364, Northeast Utilities Service Co., 56 FERC (pp 61,269 at p. 62,001 (1991).
25. Florida Power & Light Co., supra, note 22, at p. 61,456.
26. U.S. v. Aluminum Co. of America, 148 F.2d 416, 424 (2d Cir.1945). See also, In Re Consumers Power Co. 6 N.R.C. 892, 1013 (1977).
Articles found on this page are available to Internet subscribers only. For more information about obtaining a username and password, please call our Customer Service Department at 1-800-368-5001.