W.S. Read, W.K. Newman, I.J. Perez-Arriaga, H. Rudnick, M.R. Gent, AJ. Roman

The Plenary Session of the 1999 IEEE PES Summer Meeting focused on Reliability in the New Market Structure. The session was moderated by Wallace S. Read, and the presentations were as follows:

This two-part article summarizes the Plenary Session. Part 2 includes a summary of the presentation by A.J. Roman. Part 1, which was published in the December 1999 issue of IEEE Power Engineering Review, included summaries of the presentations given by W.S. Read, W.K. Newman, I.J. Perez-Arriaga, H. Rudnick, and M.R. Gent.

Since I began practicing law more than 25 years ago, I have been through restructuring for competition in several industries: airlines, railways, telecommunications, gas, and now electricity. Each restructuring has presented its own unique challenges. In electricity restructuring, one of the most interesting and, perhaps, most subtle problems is that of reliability.

If those drafting electricity restructuring legislation and regulations fully understand the reliability implications of what they are doing, they can assign responsibility for reliability to someone, and give that someone the legal tools and financial resources necessary to do the job. At the same time, legislators and regulators should be cautious that their efforts do not to create disincentives against using new developments in market-based techniques for assessing the level of reliability desired by various kinds of customers, and providing those levels of reliability.

Whenever the supply cannot meet demand for any reason, the system in that jurisdiction is no longer reliable. Such unreliability should, perhaps, be called a "deficiency of supply" rather than merely an outage, because the cause could include not only transmission or local distribution system failures but also blackouts or brownouts from insufficient generation capacity.

Whose legal duty should it be to ensure that whatever the demand for electricity at any given moment, the supply is there to meet it at the level of service quality for which customers are willing to pay? There are two possible answers: the marketplace or someone in a position of authority with the power to make a binding decision.

We seem, in North America, to have adopted a variation of the second approach with the North American Electric Reliability Council (NERC), the Canadian Electricity Association (CEA), and their various regional branches or affiliates: voluntary organizations with no legal authority to make a binding decision, but whose authority is accepted by consensus. There appear to be two reasons for this acceptance. First, the techniques and usage of market mechanisms to determine reliability levels is not yet widespread. Nor is it clear, given that transmission and distribution are seen as natural monopolies, that distributed generation through microturbines or fuel cells can compete effectively with most existing utility cost structures in the near term. Second, if voluntary efforts at controlling at least some kinds of reliability problems across large multijurisdiction power pools are not effective, governments could appoint regulatory bodies to carry out the same task, but with greater emphasis on the command-control model and less on consensus.

Where the decision has been left entirely to the marketplace before adequate competition had developed to mitigate market power, we have seen market failure. In the early years of the UK deregulation, as well as in Brazil, reliability decreased to clearly unacceptable levels while prices to customers increased. Most societies faced with deregulating electricity have been unwilling to allow the market alone to set reliability. There are at least two obvious reasons for this. First, electricity is an essential service. Second, it is not subject to effective competition from a number of competitors because it is provided, at least in part (transmission and distribution), by monopolies, and even the generation of electricity is usually subject to monopoly or considerable market power from a small number of large generators. Adequate competition in generation could easily take a decade or more to develop unless substantial divestiture is ordered and the new purchasers are prevented from merging to create new concentrations of market power. In transmission and distribution, the only effective competition possible is from distributed generation that is sufficiently inexpensive as to make transmission or distribution uneconomic.

For these reasons, it is part of the electricity policy of most countries, and certainly of all Canadian provinces, that the public should be protected against substantial unreliability by some form of public control. In most provinces, the entire electricity system from generation through transmission and distribution is publicly owned. When the cost of, and the reliability issues associated with, distributed generation have been resolved (perhaps over the next decade), its introduction may well result in enormous stranded debt in transmission and distribution. Where this debt is owed by government-owned utilities, they will pressure the government to resist the new technology for as long as possible. Eventually, the real risk of uncompetitive electricity prices will force the governments to accept new technology and market forces.

There has never been an "Alberta Hydro" comparable to Hydro Quebec or Ontario Hydro. Most of Alberta's electricity system (both generation and transmission), with the exception of the parts owned by a few large, municipalities, has been in private hands. Three large generation companies control most of the generation in the province, the largest company controlling some 60 percent of capacity. These same companies own (but no longer operate) most of the transmission. Thus, significant market power has been an issue, with continued regulation the response.

In Ontario today, even after Ontario Hydro (the provincial government's integrated monopoly) was legally disbanded on 1 April 1999, one of its successors, the Ontario Hydro Services Corporation (OHSC), of which the province is the sole shareholder, owns and operates the transmission grid. The other major successor, Ontario Power Generation Incorporated (OPGI), continues to own approximately 90 percent of the generation capacity. In virtually every municipality in Ontario, the distribution system is owned by the municipality.

Both Alberta and Ontario have started, but neither has completed the process of opening their markets to wholesale and retail competition. Even after the markets are legally open for competition, it will probably take several years before there are enough competitors with sufficient market volume to provide effective competition. Nevertheless, both provinces have given up the centralized control over reliability that existed under vertical integration, and control over reliability has become more diffused. There is a risk that effective control over reliability may fall between the cracks. By the time we realize that this has happened, it may be a very lengthy and costly process to correct the situation.

In the Canadian provinces in which the government owns the utility, and the utility plans and controls the reliability for the entire province, there is no issue about who is responsible for reliability. But the subject of this panel is "Reliability in the New Market Structure." This is a reference, in the U.S. context, to jurisdictions that are already competitive, and, in the Canadian context, to the provinces that have decided to open their markets to competition. The only provinces actively pursuing a new market structure are Alberta and Ontario; hence the remainder of this paper will focus on these two provinces.

Alberta: The transmission system is owned by the generation utilities, but is operated by ESBI, an in- dependent operator. The distribution of power is primarily by utilities owned by municipalities, e.g., Calgary and Edmonton. The Alberta Energy and Utilities Board has regulatory authority over rates and some aspects of service quality, but no direct control over the capital expenditures and operating procedures of the players. Otherwise, it could have intervened to prevent the current problems of insufficient generation and lack of intertie capacity to import power when necessary. Each of the several players has responsibility for some aspects of reliability. No one has either the responsibility or the legal authority to control the reliability of the entire system.

The generators have no economic incentive to allow their plants to become less reliable, as this would diminish the value of their output. On the other hand, at the present range of prices, given the uncertainty of the duration and end state of the province's restructuring efforts, there has been insufficient economic incentive to add generating capacity, despite supply constraints so severe that the system is often operating at or near full capacity. Indeed, there have been blackouts, brownouts, and curtailment on a few occasions, strictly due to insufficient generation and intertie capacity. And it may get worse before it gets better.

Ontario: Ontario's electricity restructuring is also well under way. Wholesale and retail markets are scheduled to be open to competition some time during 2000.

Ontario Hydro ceased to exist as a legal entity on 1 April 1999. It was broken up into several entities, the relevant ones for reliability purposes being Ontario Power Generation Inc. (OPGI, which took over all the generating stations), Ontario Hydro Services Co. (OHSC, which has the transmission and the rural distribution system), and the Independent Market Operator (IMO, which is responsible for operating the competitive spot market when it is created, for grid monitoring and dispatch). The 250+ municipal distribution utilities will be required to restructure their operations into separately incorporated wires businesses and other nonmonopoly businesses, such as retail supply and generation. The Ontario Energy Board (OEB) has the authority to regulate, by licensing OPGI, OHSC, and all other generators, transmitters, and distributors of electricity. This regulatory authority covers entry and exit from the default supply of small customers, generation and transmission rates, and service quality.

As in Alberta, no one authority in Ontario can control reliability directly and completely. No one seems very excited about this presently, as Ontario has not had major reliability problems in recent memory.

The licensing technique of regulation recently adopted in Ontario leaves much to be desired from a reliability regulation standpoint. That is because the linkages between licensing and reliability are very loose and indirect. Of course, no licensee will overtly defy its regulator by refusing to obey a lawful order. But what are the effective sanctions for subtle and gradual noncompliance? Foot-dragging and discovering numerous difficult technical and financial problems with compliance have a long history in regulation.

If the regulator considers the level of reliability in a system unacceptable, and if it can determine the causes and solutions, not always an easy task when multiple sources and operators can all contribute to reliability or unreliability. What then? It is inconceivable that the OEB would revoke the licenses of OPGI or OHSC. That would leave everyone in the province without electricity. Likewise, revoking the license of a municipal utility is impossible. It cannot fine or imprison the executives of the companies involved. It cannot order them to spend more money as such, or to construct a particular facility. It can reduce their allowed rate of return, but so what? These publicly owned entities exist to provide a service, usually at cost, not to maximize profit for shareholders. OPGI, OHSC, and the municipally owned local distribution companies have little to fear from the OEB' s potential reduction in their allowed rate of return if they are accustomed to this being close to zero, and may not wish to increase it.

The only way a regulatory agency can control reliability in electricity systems is to have at least one, or both, of two powers. First, the power to order, directly, what it regards as appropriate facilities' improvements or investments in upgrading (the command-control model of regulation). Alternatively, it must have the power, indirectly, to reduce the shareholders' income stream by an amount that is greater than it would cost to improve reliability satisfactorily (the economic incentive model). As noted above, the latter model is particularly weak when applied to publicly owned entities (whether or not incorporated), which are not operated to maximize income for shareholders.

No Canadian jurisdiction, and certainly not Alberta or Ontario, have given such powerful tools to their electricity regulators. Indeed, the objective of the restructuring has been presented as "deregulation", not the creation of a powerful new regulator. There will be less, rather than more, central control over reliability.

In Ontario, the PBR mechanism proposed by the OEB may well have a negative impact on reliability if it achieves its intended objectives. The proposed PBR will allow a utility to increase its permitted rate of return on equity in direct proportion to the percentage of annual cost savings it chooses to achieve. This annual feature is important. Reliability requires a long time horizon. Rewarding annual cost savings focuses attention on the short term.

A relatively inefficient, "fat" utility could put itself on a cost cutting "diet" by reducing staff available for maintenance, expenditures on maintenance and upgrading, etc., and have almost no immediate impact on reliability if it made the reductions appropriately. A relatively efficient, "lean" utility, however, has little or no "fat" to lose. The performance of the manager who has created a lean utility, therefore, will appear to be far less successful under PBR than that of a manager who had created a fat utility. If the municipality has any interest in eventually selling the utility (as many would, at the right price), then the higher permitted rate of return will attract a higher selling price, other things being equal.

Under PBR, especially as some municipal electric utilities are privatized, the utility management (and perhaps all employees) will probably be given monetary incentives such as bonuses and stock pptions, to cut costs and produce higher returns. With such incentives, managers will prosper in direct proportion to how much cost cutting they can generate. There will be strong incentives to cut comers on redundancy to increase reliability, to stretch equipment replacement and maintenance schedules, and to reduce staff, perhaps even beyond the point of eliminating fat, all the way to anorexia. This is especially so if negative reliability results do not show up for several years, by which time the manager will have milked the system and retired as a wealthy hero, leaving the next manager to resolve the problems created by years of under-spending. In essence, what may happen is that the proposed PBR scheme may mistakenly reward as productivity increases what are really inappropriate deferrals in investment or maintenance expenses.

But surely the OEB can stop this; or can it? Even if engineering measures eventually show a reduction in reliability, this does not necessarily mean that the OEB should or will do anything about it. First, there is the argument that a lower than historical average level of reliability is acceptable, or even desirable, because reliability has been higher than necessary, and too costly. Second, it is not clear what the OEB can do even if it considers the level of reliability to have fallen to an unacceptable level.

Given its commitment to PBR, the OEB cannot fail to honor its part of the PBR bargain: cut your costs and we will allow you higher returns. While this bargain is implicitly conditional upon providing satisfactory service, returns have not been tied to any particular level of reliability. It would be manifestly unfair for the OEB suddenly, after the implementation of the extensive cost cutting it has encouraged, retroactively to set a reliability standard and to penalize anyone for not being in compliance with it.

In fairness, however, the OEB only took on regulatory authority over municipal utilities the successors to Ontario Hydro when the latter ceased to exist as a legal entity, on 1 April 1999. As the OEB gains experience with regulating electricity, it may eventually adopt reliability standards and find some direct, effective manner of enforcing them.

There have not yet been a large number of lawsuits in Canada over reliability issues. Nevertheless, it is useful to ask the question, "Who do I sue if the lights go out?" It helps to identify who may he legally responsible for what, and to whom. It also helps to understand the nature of potential liabilities against which it may be prudent to obtain insurance or other protection.

Nobody would likely bother to sue for a short interruption that caused limited damage. A serious outage caused by no clear "act of God" or sabotage would be a different situation. This is more likely to result in a lawsuit, particularly in the era of liberalized class action rules in Quebec, Ontario, and British Columbia.

As it may be difficult to determine in advance who is legally liable, plaintiffs' lawyers would normally sue everyone who could be. The action could be brought as a class action, to maximize the number of individuals who might be compensated, and also, to maximize the potential fees for the plaintiffs' lawyers.

There would be two central, related legal issues:

Contract Issue: Most utilities do not use written contracts with the majority of their customers. Nor is it likely that a monopoly transmission or distribution service would be permitted by its regulator to impose one-sided contracts transferring all or most risks of supply problems to customers. In the absence of a written contract between the customer and its several possible suppliers, it would be left to the court to decide the terms of the unwritten contract. This exercise in retroactive contract writing by the court is necessarily somewhat hypothetical and speculative.

The court's decision would be based on the conduct and the reasonable expectations of the parties. The plaintiffs would argue that because electricity is an essential service, the parties to the contract must have understood that reliability of service was very important. It was reasonably foreseeable to both of the parties that, in the event of a serious interruption, there would be serious damage. For the residential customer, this could mean frozen water pipes and other property damage in winter, and spoiled food in the summer. The business customer would, foreseeably, incur loss of income due to the inability of employees to work because lights were out, computer systems were useless, elevators were not working, and so on. The suppliers of electricity, it would be argued, have a contractual duty to plan and build the electricity supply system to meet the demand at all times, with service interruptions only'in extreme and unpredictable circumstances, and then, to be able to correct these outages very quickly.

Let us assume, for the sake of argument, that the cause of the service interruption was within the control of someone in the supply chain. It would be necessary for the plaintiffs to establish that this interruption constituted a breach of contract. On the one hand, it would probably be untenable for the plaintiffs to argue that their suppliers have a contractual duty to guarantee that they will continue to deliver electricity without interruption, regardless of the circumstances. On the other hand, it would be untenable for the defendants to argue that, despite the fact that electricity is an essential service, their only legal obligation is to do "their best" (as distinguished from providing an objectively reasonable level of service) when the issue is whether "their best" is good enough. A balance would have to be found between these two extreme positions. In seeking an appropriate balance, the court would be inclined to consider what level of reliability a customer could reasonably expect from a modem electricity system of the size and location of the defendants.

The defendants would respond that if they are held liable, liability should be limited to the electricity not delivered, and should not extend to consequential damages such as damage to homes or loss of industrial production. This is contrary to the general rule that damages should include the reasonably foreseeable injury to the plaintiffs. The defendants would have to argue that the general rule should not apply, because this is an exceptional case. The relatively small amounts paid for electricity should not render the supplier liable for virtually unlimited amounts of consequential damages. If a contract were being negotiated today, and if the cost of the consequential damages sought was quantified, the resultant price of electricity would be so high as to be unaffordable.

The plaintiffs would call expert evidence from engineers and others who would testify that reasonable managers could, at reasonable cost, have prevented the interruption, or at least greatly reduced its duration. (We know that would be their testimony, because, if it were not, the plaintiffs' lawyers would not call them as their witnesses, but would find other witnesses.) The plaintiffs' lawyers would then argue that the defendants' conduct amounted to a breach of contract, because it is an implied term of the contract that the level of reliability provided by the defendants will be what reasonable managers can provide at a reasonable cost. The defendants would call expert evidence in reply, to the effect that they had followed generally accepted procedures within the industry, and had, over the years, provided a normal level of reliability.

If the court accepts the higher standard of conduct proposed by the plaintiffs' experts, then it will be found that the plaintiffs were justified in assuming that they would receive this level of service as a term of the contract, and the defendants were in breach of the contract. If the court accepts the lower standard proposed by the defendants' experts, then the plaintiffs were not justified in assuming that they would receive a level of service any higher than that proposed by the defendants, and the defendants were not in breach of the contract. A ruling in favor of the plaintiffs may be tantamount to a finding that common industry procedures and standards will be found inadequate when weighed against the injury done to plaintiffs by a significant service interruption. Such a ruling would require an increase in reliability standards throughout the industry.

In summary, the contract requires suppliers to use the investment policies, management methods, and equipment necessary to keep supply in balance with demand, to be able to serve all the load all the time, subject to expressed customer willingness to pay, absent reasonably unforeseeable and unavoidable circumstances. Y2K noncompliance, for example, would be difficult to defend, because this problem has been common knowledge for years. The failure to begin to work on it in time, or to devote sufficient resources to it, would arguably be inconsistent with meeting contractual obligations to provide an essential service.

Some utilities will be able to raise as a defense that their regulator has approved a set of service conditions (setting out the terms of the contract between the utility and the customer) that limit the utility's liability. The purpose of such provisions is to protect the utility from claims for consequential damages. If the regulator has the clear legal authority to limit liability in this manner and if it has done so in accordance with its legal authority, in all likelihood the courts will uphold it. But the regulator only regulates the contractual relationship between the supplier and the consumer. This does not include negligence claims. A regulator is not a court of law and would not, normally, have the jurisdiction to replace the authority of the courts in negligence claims.

Negligence Issue: It is possible to sue for breach of contract only when there is a contract between the parties. A consumer could sue the local distribution company or retailer for breach of contract, but not the wholesaler, because there would be no contractual relationship with the wholesaler. It is possible, in some cases, even in the absence of a contract, to sue for negligence. That is because a successful claim in negligence requires only a showing that the defendant owed a duty of care towards the plaintiff, and that the defendant failed to live up to this duty. The claim in negligence, therefore, could be applicable to anyone the court finds as having a legal duty of care, including, perhaps, the generator, the electricity vendor, the transmitter, and the distributor, depending upon the circumstances.

The scope of liability in negligence is not unlimited. In Anglo-Canadian law, there is a distinction drawn between pure economic loss (for which the defendants would not be held liable unless contractually committed to accept such liability) and injury or damage. While this distinction is somewhat arbitrary, all such lines are inherently somewhat arbitrary, yet a line must be drawn somewhere, as liability cannot be infinite to an infinite number of parties. Thus, recovery may be denied for the business loss resulting from a plant being closed for two days, but granted for the damage to the plant if the pipes froze and ruptured because the furnace could not operate without electricity.

In negligence cases, the first legal issue the court usually deals with is whether a particular defendant owes any duty of care to the plaintiffs. Generally speaking, a supplier, particularly of an essential service, owes a duty to take reasonable care not to cause injury to a person who is necessarily reliant upon him, and who has, in fact, relied upon him. Such a duty has commonly been imposed upon public utilities and professionals, such as doctors, nurses, lawyers, engineers, and others in similar situations. I have not done the research to determine whether there are any precedents for finding that such a duty exists for suppliers of electricity in Canada, but there does not seem to be any reason in principle why that should not be the case.

Assuming that it is found that any of the defendants do owe a duty of care to the plaintiffs, the next question is whether that duty was breached. Obviously, not all deficiencies of supply are the result of negligence. The court must distinguish between deficiencies of supply resulting from negligence and those which arose without negligence. The test would be: what was the appropriate standard of conduct for a reasonable person in the position of the defendant generator, transmitter, distributor, or retailer? What would a reasonable person do to avoid such supply deficiencies and, if they occurred, to minimize their duration? These questions would be answered with the assistance of extensive expert evidence. Experts would be asked to testify as to what could have been done to avoid the problem. Was the problem relatively unknown to experts in the industry, or was it well known? Was it avoidable at reasonable cost, or would it make electricity unaffordable to have avoided it by known methods? Have others had the same problems and resolved them the same way, or have others, through greater prudence, had fewer such problems?

Plaintiffs' lawyers will tend to select their expert witnesses from among those who will propose exemplary standards: professors of electrical engineering known for their perfectionism, consultants, and engineers from utilities with the highest possible reliability standards. Defendants' lawyers will choose expert witnesses who will testify that the reasonable standard should be around the middle of the range. The court will have a tendency to "split the difference" and pick a compromise around the midpoint between the witnesses. Thus, the court is likely to choose a position that is well above average, and probably near the bottom of the first quartile. That is why the "reasonable and prudent person" test tends to result in standards that are ahead of the industry, to raise the average standard of conduct towards a higher level than is actually the statistical norm. The surprising implication of this tendency is that there is a substantial likelihood that the court will find the average person's conduct to be negligent.

Once a court forms its opinion as to what a reasonable and prudent electricity supplier should have done under similar circumstances, the court must then decide whether the defendants' conduct met or fell short of that standard. The defendants normally argue that, if they were required to use more costly methods of preserving reliability, these costs would have to be passed on to their customers, making the cost of electricity excessive, or even unaffordable for some customers. Depending upon the evidence, such arguments may be accepted or rejected by the court. The cost of electricity to the customer is equal to the cost to use electricity plus the cost of quality of service not received (when there is a supply deficiency). If one adds these two costs, it is entirely possible that the total cost to customers is so high that it makes the level of reliability that the utility used unaffordable for the utility. In that case, there is a real possibility that the suppliers will be held negligent.

As a general proposition, the best way to limit liability is by contractual provisions under which the other party agrees to accept the risks specified in the contract. This feature of unregulated competitive markets is not found at all, or is severely restricted by regulation where suppliers have either monopolies or market power. The best defense for a regulated transmission or distribution utility would be that it complied at all times with the reliability standards set by its regulator. When regulators set standards, these have the force of law over the regulatee, and therefore, the utility complies with the applicable law.

To the plaintiff's lawyer, this would appear as a utility trying to hide behind a regulator. The plaintiffs would try to remove the shield of the regulator by attacking the adequacy of the regulator' s standards. Although regulators are normally protected by statute from lawsuits for negligence, that only means that the regulator cannot be sued. It does not mean that the regulator's standards cannot be found inadequate in a suit between the plaintiffs and their electricity suppliers.

The defense of following regulated standards can only be raised if such standards have actually been set. It is quite possible that the regulator monitors reliability, but has not set specific standards of reliability. Or, if such standards are used, the regulator may not have developed them itself, but may have simply adopted standards used by the utilities it regulates. If the regulator did not address its mind squarely to the reliability issue and make a decision to impose particular standards, which the defendant utility has met, the defense of regulatory compliance may be given little weight. It is important, therefore, for utilities concerned about liability to try to ensure that their regulators conduct hearings or other procedures, and then set reliability standards. The regulator should also periodically review and revise these standards if necessary. Standards recommended or accepted by NERC or the CEA are likely to be given greater weight than standards from a single jurisdiction.

Modem tort cases use a mixture of concepts. Sometimes they are rationalized by the courts with the traditional tort vocabulary of duty of care and standard of care. More recently, however, courts have tended to be more candid in explaining that they are also substantially influenced by the notion that the loss should be borne by whoever is in the best position to bear it.

Interesting borderline cases arise in situations of physical injury caused indirectly by supply deficiencies. For example, it is foreseeable that, if the lights go out, so do the traffic lights. If someone is injured in a traffic accident as a result, is the utility liable? Similarly, if the lights go out, so might police communications systems or hospital lights. Are utilities liable for physical injury arising from these circumstances? Perhaps, but there are two good arguments against imposing liability. The first is that essential services such as police and hospitals have their own duty to acquire special backup systems to ensure that they obtain the extraordinary reliability they need. The second is that no liability of any sort should be imposed for consequential damages that are so remote from the original cause, because the duty of care does not extend that far. How such borderline cases will be decided remains to be seen, but much depends upon the circumstances, the trier of fact, and the strength of the defendant's evidence. Utilities would do well to protect themselves from potential claims of negligence related to insufficient reliability.

Until now, we have been discussing the potential legal consequences of a claim that reliability was too low. It is also possible, of course, that litigation could be brought alleging that reliability was excessive. Lawyers representing residential customers may commence a class proceeding for damages, alleging that the reliability standards used by the utility are the result of unnecessary "gold plating," designed to favor a handful of large industrial customers only, causing the plaintiffs to have paid too much for their electricity. The lawyers for the plaintiffs would seek recovery of the amount of the alleged overpayment. If such a case went to trial, the court would have to determine what was a reasonable level of reliability to make available to the plaintiffs. If the court found the level of reliability to have been excessive, it could order that the excessive reliability be reduced to the level it considered reasonable.

Awarding restitution of amounts overpaid would be more difficult. That is because the overpayments were probably not retained by any of the defendants, but were balanced by underpayments from other customers. As it would be difficult, if not impossible, to recover all these underpayments from the other customers, the court might decide, absent any evidence of unjust enrichment on the part of the defendants, to leave the past damages where they lay.

I do not intend to suggest that there will, in fact, be any lawsuits of the types just discussed. Electrical engineers may well continue to work happily in legal obscurity, without ever having their reliability judgements questioned by anyone. On the other hand, as the new market structure becomes more competitive and power rates are unbundled, customers will begin to understand better what the different aspects of their service cost them. As this awareness develops, electricity may no longer be seen as simply a homogeneous product with a uniform level of reliability but as a range of products, with the price being influenced by the level of reliability of that product.

It may be in the interests of electricity suppliers in an era of growing competition to rebuild their systems, over time, to permit their various customer classes to purchase only the reliability they need. Then, anyone who wishes to purchase a higher-than-average level of reliability would pay more for it. This is fairer than requiring the smaller customers to pay for more reliability than they need or want, effectively forcing them to subsidize the larger customers.

Today, with reliability ultimately governed by regulators, potential defendants would be well advised to take these steps:

As Alberta and Ontario (and perhaps others) move to a new competitive market structure, responsibility for setting and enforcing reliability standards becomes more diffuse and less easily determined. Unless someone is given the legal responsibility for setting and enforcing reliability standards, not just at the bulk power system level but from generation to distribution, reliability may fall between the cracks.

The move to performance-based regulation complicates the issue further, because it may create incentives for a short run view of reliability. When productivity gains are equated with percentage of cost reduction, PBR may create the perverse incentive to defer reliability-related expenditures, perhaps unduly, to maximize short run profits. The Ontario Energy Board's proposed PBR system (and others, whose design is similar) is unable to detect or to deter such excessive cost cutting.

Looking at the issue from a broader perspective, however, it is unclear how much reliability is really needed or wanted by the "average customer" or by various customer classes. The techniques for allowing customer choice of levels of reliability for all customer classes are still relatively new, and have yet to be proven in widespread implementation. In the meantime, the degree of reliability chosen has been seen as a technical engineering decision, despite the fact that it may be responsible for a substantial amount of income redistribution through cross-subsidization. Reliability will increasingly be seen not merely an engineering issue, but also as an economic and social policy issue,

To attempt to determine who might be legally responsible for reliability under the new market structure, I have posed the question: "Who do I sue if the lights go out?" The answer to this is still rather speculative. Nevertheless, the courts might well hold suppliers of electricity legally responsible for reliability that is either unreasonably low (and, therefore, imposes undue costs on customers) or unreasonably high (and, therefore, also imposes undue costs on customers).

Andrew Roman has been a member of the Ontario Bar for 25 years and is a partner with Miller Thomson, Toronto, in its Energy Group. His electricity practice includes legal work for IPPs, utilities, and regulatory agencies. For example, he has been a legal consultant to the Macdonald Committee, the Ontario Ministry of Energy in preparing its white paper on electricity restructuring, and counsel to the Ontario Energy Board. He has also provided legal advice to energy clients in Alberta, as well as the Alberta Department of Energy. He has appeared before the National Energy Board, the Ontario Energy Board, and numerous other boards, and all levels of court, including the Supreme Court of Canada. He is the author of more than 80 published articles and a book. Effective Advocacy Before Administrative Tribunals. He has been a sessional lecturer at four law schools, and, in spring 1998, held the chair of Natural Resources Law at the University of Calgary, where he taught an advanced seminar on energy law.