All three may apply, especially if regulators go wrong and let ISOs make the business decisions.

Electricity transmission is a real business. With more than $50 billion of net plant, another $3 billion annually in capital expenditures and yearly operating income that could reach $5 billion per year under normal circumstances, the power grid is roughly twice the size of the natural gas pipeline industry. One would never know that from current events, however. Utility management treats transmission as an inconvenient stepchild. Regulators view it as a wayward delinquent in need of institutionalization. Investors ignore it altogether.

In truth, transmission affects more than reliability and quality of service. It affects profits. When grid operators set a price and decide which transactions will take place, they influence the value and location of power stations. We're not talking about arcane, opaque, academic, engineering exercises. We're talking about real money.

Nevertheless, the current transmission system was not built to meet the needs of a competitive market nor has it matched pace with shifts in population or the growth in demand for electricity (see Table 1). Environmental restrictions and the limited rewards offered by the regulatory process have contributed to minimal investment in this sector. Current policy seems to assume that transmission investment shows no sensitivity to price incentives. %n1%n New technologies could reduce some transmission bottlenecks, but we have to convince transmission owners to make the investments. %n2%n Without new investment, the scarcity of transmission (em and its pricing at time of scarcity (em could create unpleasant problems.

To solve this impasse many suggest an independent system operator to manage the transmission grid. However, a power market made dependent on an ISO invites a host of questions:

• How to price transmission?

• How to create an incentive for new grid construction?

• How to ration grid access to address congestion?

• How to curtail grid use in an efficient and fair manner?

• How to remove the ISO from making commercially sensitive decisions?

• How to ensure efficiency without discriminating unfairly against buyers or sellers?

Investors might wonder why no one is paying attention to these and other technical questions. In fact, someone is paying attention. It is the people who truly understand these issues who are hard at work, figuring out how to tailor the rules for their own benefit. Meanwhile, restructuring arrives. The structure is set. The ISO forms and takes control over corporate assets. Curtailment procedures and other rules of the game are set that will affect the profits of individual transactions.

Hello, stockholders. Is anybody out there?

Issues: ISOs, Pricing, Congestion

In the brave new world of electric competition, customers will pay for power delivered, which includes the bulk price of power plus payment for delivery and services needed to keep the system running. Most people focus on the bulk power price, which the new market sets by one of the three methods:

1. Mandatory ISO. Power producers submit bids. The ISO chooses the winners by price.

2. Bilateral Market. Buyers and sellers make deals with each other. The ISO provides the requested transmission.

3. Voluntary Market. A "voluntary" ISO accepts both winning bids (as in method 1) and bilateral transactions (as in method 2), and schedules their delivery.

The United Kingdom's electricity system uses a mandatory ISO. Many American ISO proposals take their lead from the British. They will select generation units for operation by means of a Dutch auction, starting with the lowest bid, then move up the price ladder until they fill their requirements (see Table 2).

Under the auction system, the ISO pays all winning bidders the price required by the last (marginal) supplier. Presumably, as long as the bids relate to marginal costs, the difference between market clearing price and bid represents return on capital, which is necessary to attract new investment.

During the transition to competition, however, a bidding procedure presents problems. Power plants operating under rate-base regulation collect no more than cost. Thus, only non-utility generators benefit from the margin over cost (and the utilities miss out on still another opportunity to mitigate stranded costs).

Remember, though, that the purpose of a market is to make the most efficient use of resources (which power pools now do). The new market, too, should encourage efficient operation of facilities. If power auction bids reflect marginal costs of generators, then that market also approaches efficiency. In the bilateral option, in which buyers and sellers negotiate privately and do not provide price information, the ISO cannot take responsibility for the efficiency of the market. It simply implements transactions.

The ISOs would mix three types of transactions:

1. Bid to provide energy, submitted by the producer to the ISO.

2. Bilateral transaction that requires ISO implementation.

3. A supply bid, as in type one, above, except the marketer has hedged against the transaction and has failed to deliver.

Under present rules, utilityowned facilities participate in type 1 transactions, collecting operating costs only. Independent producers participate in all three types, collecting full market prices. The ISO has sufficient information to assure system operating efficiency only in the case of type 1. In type 3, the marketer will replace power committed at one location in the U.S. system with power injected in another area. In such an instance, the ISO must examine the implications on the available transmission capability. As of now, we have seen little study of how the "voluntary" ISO will work, how it affects market efficiency and whether it might function as a dominant market participant that acts as a price giver rather than taker. If market power is an issue, then that would argue for separation of the energy market from the transmission system support (ISO) function.

Then comes the matter of congestion. Transactions could produce line flows exceeding operating limits. The ISO would have to curtail some or all transactions to maintain system reliability. How does the system price its services? And what does equal access mean in such circumstances?

The simplest proposal, which the Federal Energy Regulatory Commission appears to favor, grants equal access to all. It would collect revenues from users based on megawatts of power injected into the system and use postagestamp or contract-path pricing, or both. But what defines equal access when the ISO has to ration system use? Does the ISO have sufficient information to ration transactions on an economic basis, or perhaps by physical criteria, such as the impact of each individual transaction on line loading and constraints? Without defining these criteria, the term "equal access" will have no meaning.

Proposals: Bidding, Rationing, Hedging

At least three proposals now on the table would suggest how to run the transmission system and ration its use, including financial instruments to hedge price swings. The proposals come from three academic institutions: Harvard University (William Hogan), the University of California at Berkeley (Felix Wu, et. al.), and the Massachusetts Institute of Technology (Marija Ilic, et. al.). While all three are technical, and all will keep the lights on, the plans could produce different monetary results for individual market participants.

Harvard Plan. In the Harvard plan, the ISO plays an active role in determining transmission prices under constrained circumstances. The ISO economically and functionally bundles energy and transmission. The competitive market participant learns the price of transmission (the difference in bulk power prices between generation node and the consumption node) after the fact. The marketer, however, can hedge prices through transmission-congestion contracts. The plan gives the ISO the power to make decisions (including allocation of charges and of financial rights) that affect the profits of market participants.

As a hedging tool, the Harvard plan calls for the use of transmission-congestion contracts (which gain value as congestion increases) to average the costs of congestion over time. This approach would cause users of transmission to value the service, because they have to decide, each time, whether they garner greater profit from using the transmission system or selling their contract to use it. As a drawback, however, these contracts may not give any incentive to the owner of the transmission system to expand the capacity of the network.

Berkeley Plan. The Berkeley plan proposes a three-step system. First, buyers and sellers negotiate in an open energy market. Then, the ISO examines the proposed transactions and either allows all transactions to take place at agreed prices (no constraints) or allows only some of the transactions to go through (system constraints). Finally, the ISO tells disappointed parties how to make trades without exceeding system limits. With that information in hand, they can buy power from certain generators, with the understanding that they will resell some of that power to other generators to stay within system constraints.

In effect, the ISO's curtailment procedure determines how much power is priced in the uncongested (first) market, and in the congested (second) market. Obviously, the purchaser wants to obtain power from the former market. Studies conducted at MIT indicate that the market of choice significantly affects the profit of individual players. %n3%n As a financial instrument for hedging, the Berkeley plan calls for a joint forward market covering rights to energy and transmission, thereby incorporating the price of transmission congestion into the trading decision. A sub-market for transmission access could develop, though, which allows participants to trade rights (em subject, of course, to physical constraints on the system.

MIT Plan. The MIT plan envisions a multitiered structure that maximizes the use of market forces and minimizes the ISO's control over transaction profitability. First, participants in the energy market (bulk power) make their deals, without revealing the transaction price. Then, they ask the ISO to implement the transactions. Based on the information furnished, the ISO quotes a system support charge for the transaction. (Dealing with congestion is part of the system support services needed to allow the transaction to take place.)

The participants use the ISO's quote to decide whether to go through with the transaction as is. If the deal is modified, then the ISO modifies its quote, and the process continues until both parties are satisfied. The MIT plan gives participants the price of the transaction before the deal is complete. It also takes into account how the timing of the transaction affects its value, and lets system users make decisions without the need for ISO-determined controls.

This process may appear complex. However, it is similar to techniques used by securities traders, who somehow manage to execute seeming complex orders with blinding speed. In fact, the MIT plan informs participants of the price of the transaction ahead of the trade. It also takes into account the manner in which the timing of the deal affects its value. It would allow system users to make decisions without any need for controls determined by the ISO.

For price hedging, the MIT plan operates differently. In this case, the MIT plan conceives a two-tiered process that provides transmission system supports, approves use of market resources, and charges participants for their use of the system. Users receive technical and economic "feedback." %n4%n Thus, the plan does not require financial instruments to operate fairly. Nobody has a "right" to use the system. "Equal access," properly defined, provides the ticket to entry. Of course, market participants can hedge prices. But these participants, along with insurance firms, would be the ones to develop those financial instruments. That job would not fall to the ISO.

So, which plan should the ISO adopt? Admittedly, a process that informs participants of costs ahead of time appears more businesslike. On the other hand, the ISO's decision to select any one plan might not matter in an unconstrained market, or in one with perfect information flows. In our case, we have neither.

Pricing transmission based on historical costs may encourage transmission use and discourage investment in improving the grid. Curtailment procedures that give the impression the market participant is trading in a free market, but puts its profits at the mercy of the ISO, represent a new type of risk for electricity supply firms.

Operations: Maintenance, Upgrades, Incentives

One might readily dismiss real-time operating issues as not involving business outcomes. After all, market participants will receive transmission congestion rights and can use them to hedge price swings. Yet, this is a market in formation, not one with a long record of trading between seasoned participants. The system, not yet in equilibrium, might not exhibit equality between prices and costs. The inevitable correction (em the one that brings prices and costs closer together (em could easily surprise traders, upsetting the rationale behind deals and undoing strategies founded on incorrect assumptions.

In practice, real systems operated by real people often depart from optimal solutions. For instance, as the transmission grid opens to generation outside its border, a utility will try to price for congestion in a way that keeps anyone from losing out. This "win-win" situation may be impossible to achieve, but people still will try. The fact is, opening the grid could affect the profitability of local generation. Transmission and generation under one roof could muddle congestion pricing, eliminating incentives to provide least-cost power to the consumer. To use generation effectively, enhancement of transmission must be preserved as an alternative to out-of-merit generation. In other words, if an upgrade to the grid offers the best way to cut costs for consumers, then somebody should have the incentive to make that investment.

Nevertheless, the design of incentives highlights the trade-off between simplicity and accuracy. In that regard, congestion-pricing methods should capture long-term rather than just short-term efficiencies. The network should not, for instance, build a new line the first time the system cannot meet a request. Analysis to identify inefficiencies and the most effective solutions should precede any such move.

As for the energy market itself, the ISO also must develop the means to provide incentives that encourage efficient maintenance scheduling. The British pool does not do this, but with a limited number of generators, often within view of each other, one control area and a surplus of capacity, it may not need to.

How, then, should the market choose between power plants and grid upgrades?

Perhaps, the network should base decisions on formal, computer-based methods approved by regulators. Use of such a method would eliminate the need for the ISO to make decisions that are likely to require conflict resolution specialists, as all of the "stakeholders" pitch for the solutions that benefit them the most.

All these ideas boil down to a simple maxim: The ISO should develop a method of congestion pricing that is most likely to induce long-term efficiency and meaningful enhancements of the grid. Regulators should require such programs, rather than settling for the lowest common denominator that keeps the lights on, or worse, vague promises of future efforts.

Making It Work: Efficiency, Feasibility, Fairness

In the past, power pools provided services and designated transactions to ensure the best system-wide operations, rather than enabling specific transactions. The ISO, under open access, will attempt to do the same, at least as far as reliability is concerned. Does the ISO have the responsibility to assure efficiency as well? Can it do so with the information available?

Everyone talks about equal access, but we have no technical definition for the term when it is not possible to implement transactions, that is, when the definition has value. The ISO could curtail all transactions equally, or only those that have a serious impact on the system. Bilateral transactions of a given magnitude may not have the same impact on the system as a set of transactions of the same size managed simultaneously by the ISO. Feasibility tests must take into account the uniqueness of each transaction to prevent discrimination against certain market participants.

We believe that the best way to assure fair treatment is to use an iterative auction for transmission system services, on a first-come first-served basis, with the market participants knowing the charge, in advance, as determined by system conditions. The ISO requires no price information about the transaction terms, and the market participant decides for itself how much it will use the system, and at what price. Imposed curtailments affect profitability, which puts the ISO into the business of making commercial decisions for users. As for the need for insurance against unexpected shifts in operating conditions, need for systems services, or pricing, that is a matter between the market participant and insurance companies.

Some industry groups propose to give rights to use the system to specific users at no cost. We view this proposal as discriminatory. The ISO should attempt to optimize social welfare, not allocate profits in a supposedly fair manner, especially when no definition of fairness exists. The ISO, moreover, should not actively curtail individual transactions, since doing so affects individual profits. The market participants, the users of the system, to maximize profits should evaluate the cost of system services, and decide what to do in light of the profit possible from the trade. They auction energy; they can use a similar process to buy system services. They can take insurance against the unexpected, or not do so, making the same type of risk management decision as in any other business.

The concept of charging each transaction relative to the impact that it has on system reliability, which follows from the iterative process that we propose, has another consequence. It provides the system with a way to accumulate the necessary resources to enhance the system and maintain reliability as the market changes. %n5%n The economist marooned on the desert island with one can of tuna fish and no can opener solved his problem by assuming that he had a can opener. We note a similar tendency on the part of ISO organizers and regulators: Just assume that the resources needed to maintain reliability will be made available. Do not specify by whom. In a market-based plan, the users provide the resources.

With reserve margins at their lowest level in 30 years, a transmission system barely larger than it was a decade ago and the country about to launch what could be the most complicated deregulation ever, we would have expected fewer public policy pronouncements and more attention to the details of how to make the system work fairly and efficiently. However, from an investor or managerial standpoint, we would have expected more attention to the question of whether these seemingly technical matters affect profits. They do. People who understand the issues will figure out how to make the rules benefit them. Curtailment procedures affect the profits of individual transactions. The ISO takes control over corporate assets and nobody cares. Is anybody out there? t

Marija Ilic is a senior research scientist at MIT. Leonard S. Hyman is a senior industry advisor at Smith Barney Inc.

Table 1. Electricity Infrastructure and Demand in the U.S.

% Annual Growth

1975 1985 1995 1975-1985 1985-1995

Capacity (1,000 MW) 527.6 711.6 817.2 3.0 1.4

kWh Generated and Imported (Billions) 2014.2 2614.2 3450.9 2.6 2.8

Peak Load (1,000 MW) 356.8 460.5 620.5 2.6 3.0

Transmission Circuit Miles (1,000s) 513.9 607.1 672.2 1.7 1.0

Source: Edison Electric Institute

Table 2. Dutch Auction Bids Example (Requires Amount = 550 kWh)

Bid Amount Marginal Cost Price/ Available Price Revenue to Produce Profit cents from Amount Paid/ Received at Bid on per Bidder Accepted cents per by Bidder Price TransactionBidder kWh (kWh) by ISO kWh (cents) (cents) (cents)

A 0.5 100 100 2.0 200 50 150

B 0.6 50 50 2.0 100 30 70

C 1.0 200 200 2.0 400 200 200

D 1.5 150 150 2.0 300 225 75

E 2.0 60 50 2.0 100 100 0

F 3.0 100 0 (em (em (em (em

G 4.0 50 0 (em (em (em (em

550 1100 605 495

Note: ISO requires 550 kWh. Seven bid at marginal cost. The ISO accepts all the power bid from A to D and part of E's bid power. All bidders collect 2.0 cents per kWh. The accepted suppliers receive 1100 cents, which is 495 cents above marginal cost. That profit represents a return on investment.

1This policy resembles the Federal Power Commission's old and discredited attitude toward natural gas pricing. Gas, the agency believed, was discovered only as a by-product of oil exploration. Therefore, there was no reason to raise gas prices to induce greater supply.

2Richard E. Balzhiser, "Technological Transformation: What's the Bottom Line," presentation to Wall Street Utility Group, Jan. 18, 1996, New York City.

3Marija Ilic, Leonard Hyman, Eric Allen and Ziad Younes, "Transmission Scarcity: Who Pays?" Electricity Journal, July 1997.

4The ISO itself seeks bids from service providers to determine the lowest cost means of consummating the transaction. See M. Ilic, L. Hyman, E. Allen, R. Cordero and C-N. Yu, "Interconnected System Operations and Expansion Planning in a Changing Industry," in Shimon Awerbuch and Alistair Preston, eds., The Virtual Utility: Accounting, Technology and Competitive Aspects of the Emerging Industry (Boston: Kluwer, 1997).

5As an alternative, one could attempt to modify the existing regulatory system, putting emphasis on long-term incremental costs. That, at least, might help to attract capital to transmission. See Alfred F. Mistr, "Incremental-Cost Pricing: What Efficiency Requires," PUBLIC UTILITIES FORTNIGHTLY, Jan. 1, 1996, p. 33. See also, "Electric Transmission: Jury Still Out on Flow-Based Pricing," by Bruce W. Radford, PUBLIC UTILITIES FORTNIGHTLY, June 15, 1997, p. 41.

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.