Open Transmission Access: Is There Trouble in Paradise?

A straightforward question deserves a straightforward answer. Yes. Having said that, however, it is important to qualify the simple answer, and the question itself is more complicated than might appear. If the real question is "have we gotten there yet?" it is important to define "there." It is also important to define "we."

How Do We Define Paradise?

Open Access transmission is but a means to an end — a market goal — rather than an end in itself. So let us define the goal. The goal is usually stated in terms of the development of a perfect market in generation products, which would produce the optimized pricing for the polity — both the ultimate consumers in the market, and those who must invest in new generation. Perhaps an addendum to this goal would be worthwhile to consider as well; a market that does not gyrate wildly and without rationality. Completely irrational market pricing behavior, like markets that can be manipulated by one or a few participants, leads neither to the necessary investment to build new generation nor to acceptance of the legitimacy of such markets by the electorate.

While many of the Greek deities who lived on Mount Olympus were considered to be volatile and not always rational in their behavior, there is currently little demand for the services of those deities. Most people now think of paradise as somewhat more rational and ordered. A substitute indicator of whether we have gotten "there" would be whether FERC can safely stand back and watch the market operate without significant complaints to it that the market is being manipulated. By either test, it is pretty clear that we have not even arrived at Paradise Junction, much less Paradise itself. This paper will examine a number of problems, many of them involving sacred cows for one or more of the stakeholder groups.

Who Are "We?"

The subsidiary definitional question is as to "we." There is little doubt that those large industrial concerns with a great deal of buying power should be able to improve the economics of their power supply in almost any of the potential restructuring schemes that are being discussed. But that would occur in any event, and certainly if there is a removal of regulation over prices, since it is generally accepted that, in an industry with declining long-run marginal costs, prices for industrial customers with options for self-generation will not recover full rolled-in historical costs. The ultimate issue here is whether the smaller customers, without a practical option for self-generation, will get a fair shake. After all, the polity as a whole determines what economic system will be utilized, and if there are no benefits for smaller customers — the residential and small commercial classes — it is unlikely that an experiment in market operation will long survive. "Trickle down" economics has rarely been politically sustainable in democratic countries.

Troubles? What Troubles?

It is certainly true that the actions taken by the FERC thus far, in its Orders 888/888-A/888-B/888-C/889/889-A/889-B, have not gotten us to the goal of a perfect market for generation products. Some would point out that this goal may be somewhat unrealistic, noting that it is also true that FERC has not solved the Y2K problem for the world, or brought about a lasting peace in the Balkans, and so what? While perfection is an always-elusive goal, however, FERC is not given the task of solving the latter problems, while it is given the task of solving the first. The questions we can examine are how far it has come, and how far it has yet to go. In one view we are seeing an amazing balancing act. FERC believes that it sees the way the world will look in the "end-state" and believes as well that it must get the industry to that end state in which price regulation of generation product is only an historical memory. Almost all of the other "regulated industries" that many of us grew up with have been deregulated, and there is a substantial volume of economic literature that purports to prove that this has been good for the economy as a whole. FERC does not quite know how to get there; either what regulatory steps must be taken, or what authority it has available to utilize to get to the necessary result. Moreover, it may be years before FERC finds out for sure what authority it does not have, or what authority Congress may give it. It may, therefore, be thought of as operating on the basis akin to a variation of the old story of the Emperor's New Clothes. The variation is that it is able to hold out the promise of wonderful new clothes (or at least business opportunities) to a substantial portion of the stakeholders in the industry, who in fact believe in the promise, and who will therefore go along with the effort. It is sort of an exercise in mass faith healing worthy of an article by H.L. Mencken.

• What are Characteristics of the End-State Market? How do we Know When we Have Arrived There? Let me start by attempting to be fair. FERC is certainly faced with a daunting task. It believes that it is charged with (or should be — there are differences in the implications of the phrasing) bringing the electric generation portion of the industry in this country through a paradigm shift: from a monopoly franchise situation regulated on a cost plus return basis into a situation in which there is no direct price regulation of generation product other than an efficiently operating market. This paradigm shift occurs as a result of the changing economics of the industry, and reflects an attempt to deal with these changes in a manner that economists tell us should be more efficient. To get to that point almost everyone seems to concede that there will have to be:
• • very large and efficient regional markets for the trading of generation products,
  • transmission grids that operate throughout each market with pricing which reflects only real incremental costs for each transaction and which recover fixed costs on an annual cost basis,
  • sufficient generation owners and diversity among them so that there can be some reasonable assurance of lack of monopoly power or lack of collusion in pricing,
  • limited transmission constraints,
  • a way of upgrading transmission grids to get rid of constraints that create large diseconomies, and
  • some means of getting the price signals correct so that additional generation is built when needed. As a practical matter, there may well be other necessary elements. We will know that we have gotten to wherever it is that we need to go when people stop making justified complaints to FERC.

Impediments to Perfection

• The Process of Transition To get from here to "there," there are economists who argue that it would be economically rational to simply abolish franchises for the sale of electricity (as opposed to franchises for the operation of electric wires) across the country. House Majority Whip Tom DeLay has a bill that would do just that, and others come close. However, there are a few problems with that approach. These include:
• • Recalcitrant management. All segments of the industry (IOUs, Cooperatives, Municipals, and PMAs) have troglodytes among their top management who would have to be dragged, kicking and screaming, out of the 19th Century, as well as visionaries who think they see how to benefit their owners in a truly competitive market.
  • Recalcitrant regulators. Regulatory agencies also include troglodytes, as well as bureaucrats who are more interested in protecting their own regulatory turf than in getting a system that will work.
  • Stranded costs. All segments of the industry have participants with generation which is believed to be non-competitive in a truly competitive market, and thus have costs that are "stranded," i.e., without assurance that all of the sunk costs will be paid off in the New World Order. Some of these stranded costs are sufficiently large as to threaten bankruptcy for the owners of the facilities, especially where the owners are financed largely on a debt, rather than equity, basis.
  • Uncertainty as to existing authority to go further. FERC apparently believes that its authority to compel some changes is limited, so that it must cajole as well as require change.
  • Uncertainty as to existing authority to go as far as FERC has thus far gone. We are all waiting to see what happens in the series of Order 888, et al. appeals. FERC is apparently trying to hold down serious challenges to its jurisdiction by promising sweeteners to those in the industry that might challenge it.
  • Uncertainty as to what Congress will do. We are all waiting to see what Congress will do eventually to allocate power and decide, in its infinite wisdom, what the structure of the future may be for the electric industry. Some observers suggest that, in the absence of consensus, most in Congress "haven't a clue," and will not act until forced to do so by some outside calamity. Thus we may assume that the DeLay bill will not likely acquire the necessary political support for passage in the near future.

With these problems in mind, the FERC has apparently chosen an "intermediate" approach, in which it promises assurance of recovery of at least some of the claimed stranded costs where there is a nexus to its own actions in opening transmission and the stranding of the costs, and much of the industry, on the whole, has bought into that solution. The real dollars are at stake in the state restructuring proceedings, however, and most state regulators have simply given the industry an assurance of something close to 100% recovery of stranded costs. New Hampshire has not. This means, in practice, that the New Hampshire PUC has had to watch what purports to be a restructured world emerge in Massachusetts and Rhode Island, while its own restructuring is largely enjoined by a federal court, at the instance of Northeast Utilities. This is being used by the industry as an object lesson for regulators who will not play along. California has moved quite far toward a restructured industry, and, if it survives Proposition 9, will, simply because of the magnitude of the market, probably force the rest of the WSCC to adapt to it. It remains to be seen how the market in Pennsylvania, Massachusetts, and Rhode Island will play out. While the Nevada legislature has passed legislation decreeing deregulation, the importation restrictions make the chances for a competitive market pretty chancy. In each of these states, however, there have been major efforts by the major utilities to sell off generation so that the same entity that owns the wires does not control or own generation.

• What is a Market? What is Wrong With This Picture? Classical economic theory presumes that a market consists of a well-defined product (or service) as to which many buyers and many sellers competitively interact, thereby determining the optimal market price and quantity of the good sold. The model further posits that there exists a demand and supply curve which depicts the willingness of buyers to purchase and sellers to produce goods at various prices. Fundamental to a normally functioning market is that the demand curve obeys the law of demand — increases/decreases in the price of a good lead to decreases/increases in the quantity desired of that good— and that the supply curve obeys the law of supply — increases/decreases in the price of a good lead to increases/decreases in the desire of suppliers to provide more of that good. However, because of the various characteristics of the electrical market, the law of demand and supply may not work and, as a result, the market may not act as an effective regulator. In the classical model, which is pretty close to the New York Stock Exchange model, the market is transparent, that is, both buyers and sellers always know the price at which the last group of goods sold, and the important characteristics of the trade that are out of the ordinary. Sellers can choose not to sell below some price, and buyers can choose not to buy above some price. There are no constraints on the amounts of goods that can be traded in the market, which is essentially indifferent to the volume of trades; substantial variation in prices will usually attract more trades as buyers and sellers adjust their holdings to their needs at the new prices. If no trade takes place, then the goods in question simply remain in the hands of the party that would have been the seller, and the buyer retains the money. Both goods and money may be thought of as being stored. The reality of the electric market does not always correspond to the theory. For example, the idea of most markets is that there is significant elasticity of demand, a measure of the responsiveness of buyer's desire for a product to changes in the price of that product. But:
• • The law of demand (and thus real elasticity effects) depends upon some buyers being willing to reduce their use of the goods in question when the price rises, and that in turn depends upon knowledge of what the price is at any particular time.
  • Thus far, at least, there is no electric market with real time pricing for end users so that they could effectively modify their usage based upon knowledge of the price involved at the time. Thus the effective short-run demand is almost perfectly inelastic, as a result of imperfect information.
  • Thus far at least, there is no way to store significant amounts of electricity. Thus the law of supply will not always function properly and the short-run supply curve is also much more inelastic than in most traditional industries.
  • There are many more constraints as to transmission of electricity than appear in the classical market model.

This means that prices are set in this market by a combination of suppliers who cannot store their product bidding into a group of purchasers with almost totally inelastic demand. A corollary of this is that there will be much greater variation in pricing during the course of any day than would be the case if buyers could rationally adjust their use of the product in response to real-time pricing. The market behavior of the electricity market is not very much like the market for pork bellies, and markets that are regulated by the Commodities Futures Trading Commission look tame by comparison. The California ISO maintains market statistics for all of the commodities traded in California that are much superior to any other source of data known to the author at this time. An arbitrary week shows many of these characteristics: These characteristics (albeit in more or less extreme form) appear in all of the hourly markets that seem to be operating, and may be assumed to be characteristic of a bid-based, market clearing price form of market. Economists tell us that this is likely to be the most efficient form of market for electricity as a good. With those characteristics in mind, let us look at some of the market imperfections that are now clear, as well as some of the clear structural problems that prevent us from getting to a satisfactory "there."

• Indicia of Known Market Imperfections There are probably more indicia of market imperfections than one could conceivably list in a paper of this scope. Some of the more obvious include the exit of some of the major suppliers from the residential market in New England and in California and the exit of some of the major players and a lot of the smaller ones from the marketing business in general, following the shock of the Midwest Price Spike. But these are symptoms, and it may be possible to get to more basic issues that have caused the symptoms.
• • Price spikes Price spikes (or "excessive price spikes") are a symptom of a problem in the infrastructure, but do not, in and of themselves, necessarily prove the existence of any specific problem other than the obvious political one: the inherent concern of the voting population about markets being manipulated is certainly raised by their existence. These price spikes have occurred this past summer in the Midwest (apparently originating around the ECAR region, but spreading throughout much of the Eastern Interconnection, excluding only New York and New England and adjacent Canada) and in California (although primarily focused on the ancillary services markets). In both areas the price went above $7,500 per MW per hour. Taking the California market as an example, there is more than a 50,000 MW load, and the market clearing price is in principal being paid for all power sold through the PX. If we assume that the ordinary average market clearing price is around $35 per MWh, and that all participants are taking power from the power exchange without hedging, it is not unreasonable to calculate the extra cost that would be borne by the users in the state from such a price spike as $7,500/MWh x 50,000 MWh, or $375 million per hour. Even in California terms, that is a lot of money. At that rate, one could pay off, say, a $300/kW gas turbine machine in 40 hours of operation. Let us say that this is an indication of a problem. FERC has reported that the Midwest price spike (June 25-26, 1998) did not arise from manipulation of the market, and was nothing more than an imperfection in the learning curve. In the Midwest, there is no one whose responsibility it is to "watch the store" on an assigned basis, and thus apparently FERC feels that it can treat the imperfections of the market as if it were a semi-disinterested observer. The California market, however, is another animal entirely. The California ISO, as a condition of its taking over the market structure, is supposed to monitor the market for signs that it is not working and that manipulation is taking place. The California market started off with only the day-ahead market (the PX) being structured on a market clearing price basis, but the anticipation was that all services would wind up being bid on that basis. After some months of experience with that market, the various generation owners asked FERC for permission to go to bidding and market-based rates for certain ancillary services on the hour-ahead market and in the market for ancillary services. That permission was granted on July 10, 1998. On July 13, 1998 (as corrected on July 16, 1998), however, the ISO had returned to FERC to report that the market for these new services was not working correctly, and to request approval for something it had already put into place, a price cap of $500/MWh (or MW per hour, in the case of capacity related payments). FERC did in fact authorize the ISO to establish price caps at a level it determined appropriate, and the ISO has now administratively reduced the price cap to $250/MWh, while attempting to find ways to make the market perform adequately without such a cap. The market price data, therefore, tend to show the "flat top" curves peculiar to price-capped markets on a lot of occasions. Everyone involved agrees that something is not working properly, but there is no complete consensus on what is not working.
  • Forward Planning; Is There Any? If Not, Should There Be? In all of the analysis of these issues, most of the function of forward planning for generation adequacy is assumed to be taken over by the market in the New World. But in all of these markets, the primary commodity being traded is energy, and ancillary services, always including at least spinning and non-spinning reserves. Some of the markets also have a capacity market (e.g., NEPOOL). At least a part of the thinking for such a capacity market is that this will allow a requirement for sharing of capacity obligations for the region by requiring that there be contracts for capacity by all sellers of power to cover not only the load but the installed reserve obligation. This requirement is supposed not only to assure that there are real machines in place to provide power (see also the discussion of tagging) but also to provide early warning of the need to construct new capacity. In New England, there is at least one participant (Enron) that has proposed doing away with the capacity market because the prices have tended to be very low or zero. Is this just a function of the fact that there is thought to be adequate generation under contract or in place for the moment, or is it a function of some more basic point that the market does not need this forward indicator? Without this sort of market, is there any assurance whatever that there will be adequate generation built?
  • Insurance/Pooling/Risk Sharing Pooling and reserve sharing are an important part of the economics of the electrical world with which we have lived for the last 30 years. It took years of dispute and of litigation at the old FPC (now FERC), winding up at the Supreme Court in Gainesville Util. Dep't. v. Florida Power Corp., 402 U.S. 515 (1971), before we got those rights extended to smaller systems. Even then, we had to fight to get into pools. See, e.g., New England Power Pool Agreement, 56 F.P.C. 1582 (NEPOOL I), reh'g denied, 56 F.P.C. 2862 (1976) (NEPOOL II), aff'd sub nom, Municipalities of Groton v. F.E.R.C., 587 F.2d 1296 (1978); Mid-Continent Area Power Pool Agreement, 58 F.P.C. 2622, reh'g denied, 59 F.P.C. 1651 (1977), aff'd sub nom, Central Iowa Power Coop. v. F.E.R.C., 606 F.2d 1156 (D.C. Cir. 1979). With reserve sharing and pooling, we essentially obtained energy at cost for a period of time in return for providing the capacity necessary to provide the same energy to others when needed. This also permitted all parties to lower their reserve margins to comparable levels, rather than allowing only the largest parties to optimize their reserve levels through loss of load probability measures, while requiring smaller ones to maintain "largest unit out" reserves or higher. This development, in and of itself, did much to permit smaller public and other entities to compete on a fair basis with the large IOUs. The concepts of pooling and reserve sharing translate in the New World to insurance and risk sharing. But in the New World, the assumption seems to be that the market will provide, and that no pooling or reserve sharing obligations are necessary. This is not necessarily so. The basic concept in most of the market designs seems to be that each entity responsible for load will contract for the energy necessary to meet that load, and will bear the risk that the load may not match that anticipated and the risk that some of the energy contracted for does not make it into the grid, for whatever reason. For most hours of the year, a market-clearing price may well solve the problem of risk sharing in an adequate manner. But if one's generation unit goes down, or one's supplier fails to comply with its contractual obligation at a time when the price spike occurs, the consequences can be severe. As a consequence of the Midwest price spike, for example, a few marketers went out of business, and very substantial losses were taken by other participants in the process. Several substantial entities announced that they were quitting the electric market business as a consequence, and there are suits pending against at least one municipal entity, Springfield Illinois, that may have the potential to bankrupt the city. In a situation in which a supplier defaults, there is an argument that it was the buyer's obligation to have assured itself of the financial capability of the supplier to perform, and that it is thus the buyer's fault if it is stuck with damages of this sort. But if that argument is carried to its extreme, the result of the argument would be a market in which only very rich marketers or sellers could operate. As a practical matter, forced outages happen at unpredictable times, almost as a matter of definition. When extreme situations occur, the desirable result would be to have all generators available to operate to help to meet the load. If the result of having brought on generation to help the situation is that an outage will expose the "good participant" to extreme losses, there is less likelihood that participants will in fact contribute to the good of the market as a whole, since they will reserve more of their own generation to meet potential outages of their own. This forces participants back toward the old "largest unit out" economics, effectively raising the cost substantially for smaller players, hardly the result that should be desired for an efficient market. One approach toward solving the problem would be to take all of the costs that are incurred by the ISO in matching load to generation in excess of the day-ahead market clearing price and fold them into an "uplift cost" applied across the board to all loads, just as the ancillary services costs are rolled into an uplift cost in California. But that assumes that all suppliers are equally reliable, and that there is no advantage in gaming your estimates of load and generation in the day-ahead market. Neither of these things is true. Another potential approach would be to require that all participants have available to them enough real and identifiable generation resources to meet their own load plus reserves adequate for the pool as a whole, and that there be some form of certification program so that there is an assurance that the units are real and maintained. In return for that contribution, the costs of replacement for forced outages — at least for some period of time — could be included into an uplift cost, as described above. But that would require an agreement as to the period of time that is covered and also as to the standards of maintenance. That has the risk of potentially leading to a least common denominator approach for maintenance. The issue is how to structure a risk sharing pool in a manner that encourages participants to maintain their own units adequately and also to take reasonable precautions as to the financial capabilities of their contractual counterparties. There is at least one theory that we should leave all owners of generators to prosper or fail depending on how they maintain their own generation; that seems rather Draconian. What is more, such an approach tends to reward those who can better internalize the risk of a single unit dropping out of service (those with many units) in a way that will drastically tilt the playing field in a manner that makes it more expensive for the smaller players.
  • Expertise If the electric market is becoming essentially a commodity market, albeit an extreme one, hedging becomes the essential technique in dealing with risks. Hedging capabilities and the knowledge of the market necessary to be able to use them are not in high supply thus far. Given the anecdotal evidence as to the behavior of participants in the Midwest price spike fiasco, it is clear that there are many more people who believed themselves to be experts in the area than could justify the claim of expertise after the fact. Perhaps this is merely a "transition problem." But how many systems are to be lost, how much can the consuming public be done in, before the system as a whole is rejected as unsuitable to a democratic society? There are two potential ways to deal with this problem. At one extreme, there could be, say, six equally sized marketers that cover the entire United States all of whom can compete to sell power to any load. Those marketers would be large enough to internalize the risk in the difference between what they sold for and what they bought for, and would, at least in principle, quickly discriminate between reliable suppliers and unreliable suppliers, so that there would be a market mechanism to encourage reliability of operation without threatening instant bankruptcy for mishaps that could not be avoided. See discussion in Constraints, below, for one of the reasons why this proposal cannot work properly in the current environment. Unless we are going to head in the direction of a very limited number of marketers participating in the market, a second way to deal with this problem will have to be found. At the extreme, such a method could be some form of a widespread and uniform training session for marketers for every player in the business. Since that does not seem likely, a fallback solution would be for some form of mutual aid or insurance program, based on certain common standards of risk behavior permitted. In one sense, it would be logical for national organizations such as NRECA and APPA to sponsor such insurance arrangements for their members. But a common insurance program could only work within the context of a single market structure. Thus far, both FERC and the state PUCs (through NARUC) have adopted the Mao Zedong view that one should "let one thousand flowers bloom." If this continues to be the case, it is entirely possible that the result will be similar to Mao's Cultural Revolution: random but widespread disasters for the participants. In more direct terms, there may no longer be APPA and NRECA members. FERC has stated that the Great Midwest price spike is unlikely to recur. The reasons for this conclusion are somewhat opaque, but rest in part upon the assumption that the participants have learned something from the experience. That is always a hopeful thought, but there is an element of smoke and mirrors in the conclusion as well. It is certainly not obvious to all observers that the conclusion is warranted. As noted above, the California ISO has reached a somewhat different conclusion as to the price excursions in that state.
  • Tagging: Why Are We Always "It?" Differential Tagging And LLR/TLR As an historical matter, power flows (outside some tight pools) occurred, to the extent that generation and load outside the control area were involved, on something of a "black box" basis. A control area operator could have power lines endangered by thermal overloading, but have no idea what was causing it, and thus very little ability to control it. A system of intuition grew up, in which the better control area operators could usually tell whose transactions led to their lines being at risk or damaged, but events such as Ontario Hydro selling large amounts of power to New York or AEP selling large blocks of power to TVA demonstrated some rather surprising constraints where there should not have been constraints in theory. Some of the control area operators in NERC were given the task of developing a way in which it would be possible to keep track of whose transactions were whose in the New World, so that a security coordinator could preserve the system when needed. What they came up with was "tagging." Transactions between control areas were to be "tagged" so that a generator increment in one control area was to be "tagged" to a generator decrement in another control area. All of this sounds very rational. Unfortunately, the rationality disappears with scrutiny. First, of course, the idea that increments and decrements can be tied together only works if we are talking about what used to be considered economy energy transactions. The value of tagging assumes that there are things that can be done about overloads and constraints. There are usually alternatives when we are talking about economy energy, but not so frequently where we are talking about transactions relied upon to meet load. Thus a transmission system which is perfectly adequate to meet basic load from resources within a region may not be so well adapted to a structure where generation resources are being bid or dispatched into a much larger market on a least cost basis. This is the problem addressed under the heading of Constraints below. More significantly, however, if only transactions between control areas are tagged, the information is missing entirely with respect to generation being used to meet native load, and constraints are caused by combinations of generation meeting load, without regard to whether either the generation or the load are within or without the control areas. In short, NERC has created the function of security coordinators, and then emasculated them by withholding the key information from them. That would be bad enough, but in fact, it has developed that a great many transactions among control areas are not tagged either. As a practical matter, the longer term arrangements upon which people like us depend are tagged; the shorter term transactions, which are those that should in fact be cut as non firm, are not. Northern States Power has attempted to bypass the review of Order 888 et al. by appealing a different order, requiring it to adhere to the curtailment provisions of its open access tariff to the Eighth Circuit. Northern States Power Co. v. FERC, No. 98-3000. That case is moving forward on an expedited briefing schedule, and the basic proposition urged by NSP is that FERC is exceeding its authority with respect to native load, by requiring that curtailment take place on a comparable basis. So far Dairyland and the Attorney General of Minnesota have supported the NSP side. This case should be watched as a potential "end-run" around the 888 litigation.
  • High Cost Infrastructure A market structure should cost something, but the savings to the public at large should outweigh the cost. As a general proposition, that seems to be unassailable. The California ISO/PX structure cost something more than $500 million to set up. Even in California, that is significant. Of course, there is an argument that the pre-existing computer "back room" infrastructure was totally inadequate to the needs of a market in which, in principle, every citizen of the state has the right to a bilateral contract with any supplier he or she chooses, so that a good part of the new infrastructure would have been required with or without the ISO/PX structure, but the California investment number is nonetheless being used in several state proceedings as a reason why no ISO should be required. Whatever the reason for the high cost, the transaction costs of shifting from the price regulation structure to the New World are clearly not inconsequential.
  • Costs Being Charged to Transmission Cynics, and those who have criticized the regulatory regime in general would predict that costs that had previously been charged to generation expenses or to generation fixed costs would be charged to any other category that might permit an assurance of their recovery on transmission service. One of the things that may show those cynics to have been correct is the fact that we are seeing this prediction coming true. Stranded costs of generation and "regulatory assets," of course, are being charged to those who must utilize transmission or distribution services. We knew that that was going to happen, whether it made economic sense or not, because some version of it made political sense. But the more disturbing thing, for those of us who thought that there was merit in regulation, is that every other cost that anyone can argue could or should have any connection whatever with transmission or distribution is being taken out of generation accounts (where it would be subject to competition) and tucked into the T&D accounts, where it will be assured of recovery. Not only does this tend to inflate the cost of transmission and distribution, but it also unfairly subsidizes the generation component vis á vis competitors who must build their generators from scratch and without transmission investments to inflate.
• Structural Problems There are a number of structural problems without a solution to date. As noted, the "faith healing" approach to these problems is to pray, and to hope that the participants will find it in their interest to cooperate in the interest of higher profits later on, while simultaneously working with a portion of the industry to change the laws to provide the necessary authority. There are several areas in which there simply is no clear answer so far to the known problems; indeed, some of the stakeholders are the problem in at least one problem area.
• • Constraints. How is pricing of constraints handled? LBMP? Roll-in? Others? As a general proposition, there were relatively few constraints in the electric transmission system when operating solely in a wholesale market, with limited transactions among participants. There was (and is), of course, the very limited constraint between the Eastern and Western Interconnections. There is the even more limited constraint between ERCOT and the Eastern Interconnection in the form of the SPP. There is an almost impervious constraint between the bulk of Mexico and ERCOT. And there is the constraint between Québec and the remainder of the Eastern Interconnection (although Québec has now been declared officially to be a part of the Eastern Interconnection). All of these are clearly constraints, since they are DC ties, or ties that have to be physically rearranged so that there is no AC connection, and there is no other way for power to flow around the constraint. All of the theorists assumed from the beginning that there would be limited markets across these constraints. But there are more constraints than most of us thought of when retail choice begins. Within California, for example, there are the "active constraints" on "path 15" (between Northern and Southern California) and in the Humboldt region, which are expected to, and do, overload when there is an effort to dispatch the cheapest resources bid into the system for a significant number of hours during the year. As to those "active constraints," the prices on either side are allowed to diverge when the lines are overloaded. And there are also some recognized but "inactive" constraints, which include the San Francisco area, where there is clearly inadequate transmission to meet the loads of the San Francisco area without running generation internal to San Francisco for a much larger number of hours in a year. Then there are other areas, including San Diego, in which there are supposedly no constraints, but in which there still is a requirement to run generation internal to the area to maintain load. In any other place, these would be considered to be "load pockets" and therefore constrained, but in California they are considered to be something else. Between California and the rest of the country, there are not infrequent periods when the California/Pacific Northwest path (sometimes referred to as the California Oregon Intertie, or "COI") to the California-Oregon Border ("COB") loads to its maximum, and also when the California/Desert Southwest paths ("Palo Verde") fills. If the ISO were larger in scope, these would become internal constraints as well, as would the path between California and Northern Nevada. But there are other constraint issues that occur and which need to be dealt with. For example, if A, B and C have built generating plants on a transmission line, and if the operation of all three of those plants will fill the line, what happens if D comes along and connects another generating plant to the same line? All four cannot run at once. In a bid-based, market clearing price model, D may well have a lower operating cost, and thus be able to bid a lower price than A, B or C, in which case the PX mechanism will select the lowest cost set of bids to meet its needs, and whichever of A, B or C has the higher cost will functionally be landlocked, without ability to deliver its energy whenever D bids. This will occur even if the highest cost of the four is below the market clearing price of the PX. It is one thing to say that changing from a regulatory structure to a free market structure in generation product pricing may not be actionable, since the owner of generation still has an opportunity to employ its assets productively to earn in either form, simply at different times. It is another to change a structure where the owner of generation that has been able to assure a way to market may be forced into a situation in which the way to market is foreclosed without compensation. The ultimate question is who is responsible for payment for the upgrade to the transmission system that would be required for A, B, C and D all to be free to market their electricity. Under the pre-New World order, the cost would be imposed primarily upon D, if not spread across all users of the system. Under some versions of the New World order, the cost would be imposed entirely upon whichever of A, B or C has the higher marginal cost of operation. This change in reasonable, investment-backed expectations as a result of restructuring would arguably be a "taking" under the Fifth and Fourteenth Amendments to the United States Constitution. How is this sort of constraint handled? Obviously, an ISO cannot permit a single generator to set the price even for load within the load pocket, when there is no competition. Thus the usual answer is that some generation will be characterized as "must run" and paid something approximating its cost of service in order to be available for operation at the call of the ISO when it would be required to run for reliability purposes. This "Reliability Must-Run" class of units is taken out of the bidding structure, thus effectively reducing the number of bidders into the market, and becomes much larger when constraints become more pronounced. There is also a class of "Regulatory Must-Take" units, that frequently seems to include such things as nuclear units whose revenues are tied to the frequency of operation, QFs, and the like. The removal of these units from the bidding pool and the removal of the load served by the output of those units from the load to be served by remaining bidders has the effect of reducing the effectiveness of the market as well. While this problem area is not a direct result of transmission constraints, the fact that many owners of generation are vying to be characterized as "Must [anything]" does have a perceptible effect on the operation of the market itself.
  • Lack of Adequate Siting Authority If transmission constraints are going to become more of a concern and more of a hindrance to the effective operation of the electric generation product markets in the country, the first question an outside observer would ask is "so what?" If it were easy to site and construct new transmission, this enhanced concern with transmission constraints would become of only marginal and short term concern. But siting and construction are not necessarily easy. Let us look at the siting problem. The siting of transmission lines has traditionally been thought of a state activity, usually delegated to the state PUC, although frequently involving other state agencies as well. This was a reasonable allocation of functions so long as transmission lines were considered to be essentially for the use of the vertically integrated IOU in serving its load. The tradeoffs there were direct and reasonably effective. If the PUC did not get the line in place within a reasonable period of time, some consumers would risk not being served, and there would be political penalties that would be applied to the PUC by the unhappy citizens. All of these political checks and balances have lessened or disappeared as transmission was constructed for regional and national purposes. State PUC members have come, in some instances, to view themselves as "gatekeepers," whose responsibility is to collect the "tolls" to transfer the economic benefit of the potential transactions in power to the state in which the transmission is located. To the extent that this has become harder with FERC pushing for a seamless transmission system, the rake-off has transferred to the permitting process for new lines. If the national interest is in creation of an adequate transmission system for the New World, then the state by state siting process arguably has become or is becoming dysfunctional. Almost all participants in the process recognize that this is a problem. Even NARUC has a position that envisions a set of regional multi-state siting processes, as an alternative to legislative pressure to let FERC do it. The DeLay — Markey bill, H.R. 4432, 105th Cong. (1998) simply proposed to switch this jurisdiction to the ISO for some purposes. The Bingaman bill, S. 1276, 105th Cong. (1998), proposed to switch this jurisdiction to the FERC, relying upon the joint board concept of Section 209 of the FPA.
  • Who Builds Transmission Upgrades? Who Pays for Them? Another of the basic questions is who, in the New World, will build and pay for transmission upgrades that are necessary for an adequate grid. Obviously, this is a key question, and one that impacts many participants in ways that are not necessarily consistent with their public posture on other issues. If we assume that there will not be full divestiture of transmission facility ownership to some independent regional entity, then some form of regional ISO or "ISO equivalent" will be doing the planning for the transmission upgrades that are necessary for the region. If there is still a utility owning the wires, however, the utility which originally owned the wires in the region will, if past experience is any indication, want to build and control the new facilities to be constructed. On the other hand, some utilities have declined to construct or plan new lines, because they are concerned that they will not have control, or alternatively that they cannot raise the capital necessary for the additions. Our side has usually taken the position that we will build new transmission lines if no one else will, but there are tax implications as well, at least if this concept is taken to the extreme. An essentially risk-free regulated rate of return is still something that will attract investors, so that someone should be willing to put in the money and the effort to construct the necessary new facilities. If the ISO has the right of eminent domain now typically held by the IOUs, and simply puts the right to build a line of specified characteristics out to bid, it seems likely that there will be an adequate number of responses to assure completion at reasonable cost and within a reasonable time frame.
  • Can FERC deal in a contract-based market? Will anyone else? FERC has, essentially since its predecessor's inception in 1920 (or 1935, if one wishes to focus on Parts two and three of the Federal Power Act), based its work on the concept that its primary job was to determine a just and reasonable rate, and that refunds are an adequate remedy. That is only true if what is being sold is a completely bundled product, and if there are no alternative suppliers. What happens in a market where prices are determined by the market for generation products and where a failure to provide transmission service that should have been provided will mean damages of perhaps hundreds of times what the transmission rate would have been? Obviously, FERC (or whatever other agency is determined to be the appropriate decisional authority) will face different issues, and very different approaches to the legal questions to be resolved. In a situation involving market abuse, for example, the SEC has sanctions that are applied directly to the miscreants rather than waiting for the market participants injured in the process to take action. The FERC has no legislative authority to act in a manner consistent with the SEC approach to market abuses. Similarly, the commercial common law of contracts and torts has evolved so that there are mechanisms that are available to the courts that, at least in principle, will protect a participant who has been wronged by another from the financial consequences of the dereliction. FERC has no such tools clearly available to it; while it has thus far been able to fashion tools to accommodate its most pressing needs, there is not unambiguous Congressional direction to use them. There is a counterpoint view to all of this. State and Federal courts are quite loathe to get involved in matters involving issues under the Federal Power Act, or even arguably under the FPA. While FERC itself has swung back and forth over the years as to whether it will encourage or discourage the use of dispute resolution mechanisms other than FERC itself, it is very clear that a high proportion of state trial court judges, and many Federal District Court judges, do not want even to come close to a case that smells of FERC jurisdiction. They fear the complication, and the interrelated nature of the issues. So there probably is an inherent tendency to favor having FERC, as the entity knowledgeable in the area, decide more issues rather than fewer.
  • I Thought the Consumer Was Supposed to Benefit From Restructuring? Are We Supposed to Take Comfort in the Fact That Generation is Being Sold at Multiples of Book Value? The idea behind the transition to the New World in this country is that the removal of price regulation and the transaction costs associated with administrative regulation of the price of generation products will free the market forces to operate and produce a cheaper and more abundant source of electric power for the populace at large. All of us who have been associated for a long time with regulation and regulators can understand that rationale. The vertically integrated utilities which found themselves faced with the potential of being cast into the market to compete initially were so concerned by this potential that the "Carl Sagan Theory of Stranded Costs" was developed, in which "billions and billions" of dollars of investment were assumed to be stranded by such a transition. There should have been at least some skepticism of the basic assumption that prices would inevitably come tumbling down. For there was historical precedent available: the idea of a transition to a market based economy in electricity was not one that was created in this country, nor one which was first tried out in this country. The privatization of the electric industry in England and Wales occurred under Prime Minister Thatcher, with the passage of the Electricity Act in 1989. The success of that venture led to the later privatizations in Australia, New Zealand, Chile, and in other countries in South America and in Europe. The hidden secret here is that the change in industrial structure that occurred in each of these privatizations was driven by another goal as well. The goal, however, was the maximization of profit to the government, which had previously owned the entire electric system in those countries. Thus the idea was to create a structure which would make it feasible for participants to bid more for electric plant than would otherwise be obtainable, rather than to drive down the costs of power for the populace. While it is conceivable that the price of generation units might go up at the same time that the price of power from those units is going down, an element of doubt should be recognized if this proposition is going to be taken seriously. There are, after all, only so many ways in which efficiencies can be discovered and applied. And the "magic of securitization" is essentially the same as one of us extending the mortgage on our houses. The monthly payment may be less, but the overall cost will be appreciably higher as interest is paid over a longer period. Thus getting ratepayers to commit to pay off the stranded costs of inefficient utilities through securitization could easily have been anticipated to lead to the sort of ratepayer revolt that is now occurring in California and Massachusetts, once ratepayers figured out what was happening. Putting the magic of securitization aside, the fact that a number of the sales of generation that have occurred have taken place at multiples of book value of the generation assets as held by the selling utility should cause some questions to be raised. Someone is assuming that there is going to be more money received by at least some generators than was in fact received by the previous generation of utility operators. This seems to be inconsistent with the idea that ratepayers will benefit across the board.
  • NERC/NAERO. Fiddling While Rome Burns? Another of the basic problems with the industry has been that reliability issues are supposedly being handled by NERC, a voluntary organization owned by the regional reliability councils, also voluntary organizations. This structure sort of worked when NERC was first established, after the great Northeast blackout, when it was better than nothing. But reliability issues are intertwined with economic issues now, and as the industry in general disaggregates itself for the coming of the New World the economic impact will become even greater. See the discussion above under Tagging, for example. To its credit, NERC established a blue ribbon panel, which concluded that the reliability structure of the industry needed to be revised, and recommended the development of a self-regulating organization, under the general supervision of the FERC, more or less along the model of the National Association of Security Dealers (NASD) to accomplish this task. Such a SRO would require legislation to have powers delegated to it from FERC. This is the sort of legislation that would pass quickly if all stakeholders were in agreement, and which has little or no chance of passing by itself if there is disagreement. There is disagreement, at the very least:
  • • between the Eastern Interconnection and the Western Interconnection
    • between Canadian stakeholders and US stakeholders,
    • between those who believe that NAERO should take the primary responsibility for the reliability rules for the nation and those who believe that NAERO should simply be a clearing house for the rules established by the regional reliability councils or their successors,
    • between the existing regional reliability councils, and those who believe that the law requires that if powers are to be delegated by a federal agency it will be necessary for the delegee to comply with certain structural obligations of fairness and due process, and
    • between those who believe that the addition of a reliability provision would make overall restructuring legislation move more rapidly, and those who believe that a "rifle shot" reliability bill would be better.

The real question can also be viewed as whether we continue with NERC as it is and the regional councils as they are, or we move to a much more open and accountable structure. The drafting process involved in all of this seems to be slipping into a morass, and it is not clear that anything is coming out of it that could be is passable as a stand-alone bill.

CONCLUSIONS I return to the point at which we started. We are not "there" yet, wherever "there" may be. Nor is it clear how close to "there" we are, since many of the issues that will have to be faced are still opaque. FERC is certainly trying to move us all along the path, to a destination that is seen through a glass, darkly. If the genius of the market works, it will be because some will have a better idea than others of where we are going, and will guess right, thus showing everyone else what needs to be done.