Press Release
Full Report (pdf format)

EXECUTIVE SUMMARY

During 1998 electricity markets on the West Coast and in the Midwest experienced repeated episodes when prices increased at least ten fold and as much as 300 fold. Hundreds of millions of dollars changed hands in a matter of days. Bankruptcies of over a quarter of a billion dollars resulted. One estimate places utility losses during the air conditioning season alone at over half a billion dollars. In spite of these unsettling events, by mid-1999 restructuring had been enacted in states in which more than half the electricity in the U.S. is consumed.

This paper draws lessons from these chaotic events by embedding the "facts" in a well-known analytic framework - the structure-conduct-performance paradigm (See Exhibit ES-1). As the title of the paper suggests, we conclude that the market problems of 1998 indicate that much more vigorous consumer protection is necessary, if the restructured electricity market is to benefit all consumers. These are not accidents or aberrations; they are exactly the behaviors one would expect to occur when rational economic actors take advantage of market imperfections and institutional weaknesses. The issue is no longer whether to restructure; it is now how to implement restructuring in a manner that prevents these abuses.

A. THE UNCONTROLLABLES DO NOT EXPLAIN THE PRICE SPIKES

Technology prone to outages in the supply of a commodity that is impossible to store creates significant potential for supply problems. Consumption is significantly influenced by weather. Pricing structures and capital equipment give little incentive or ability to alter demand in the short-term. These are the uncontrollables that some claim are the causes of the price spikes. However, the evidence we reviewed suggests that these factors were not sufficiently more powerful in 1998 than earlier years to account for the huge jump in the price of peak power.

Arguments that the weather and outages were unique factors underlying the price spikes are undermined by the fact that 1997 experienced similar factors with less drastic results. The difference in weather was not that great. "Accidents" affected a variety of technologies in a number of markets at different times. The problem is pervasive. Several major categories of baseload plant - nuclear, fossil, and hydro - have contributed to one or another of the unplanned outages. This was true of the summer price spikes in both California and the Midwest.

Weather and outages may tighten supplies within a given region; transmission constraints limit the ability of power to come from outside of the region to alleviate the local imbalance. There is no doubt that constraints on the transmission system played an important role in both the Midwest and West Coast price spikes. Once again, however, there does not appear to have been a great deal of difference in transmission capacity available between 1997 and 1998. Moreover, the transmission supply problem is pervasive and widespread. This long-term problem requires a response.

With prices hitting levels that were several orders of magnitude higher in 1998 than 1997, it is unlikely that weather or outages alone could account for the difference. In some respects 1998 was no worse than 1997. The physical difference was not sufficient to account for the financial differences.

B. A NEW MARKET CREATES DISORDER

Accidents do not just happen; controllable conditions and circumstances can make them more or less likely to occur or make their consequences more or less severe. There are ways in which new market institutions and transactions made the likelihood of accidents and their impact greater.

1. BREAKDOWN OF COORDINATION: The market reduced the ability of system managers to coordinate and run the transmission system. The problem stems both from complexity and from a lack of cooperation. Market participants do not have an incentive to cooperate. The number and complexity of transactions compounded the problem of system management. The number of traders increased over 50 fold; the quantity traded increased several hundred times. There were also complications of financial and ownership relationships between entities.

2. TRANSMISSION CONSTRAINTS: The introduction of competition into a capacity constrained system put a severe strain on already stressed facilities. In a competitive market, some entities gain an interest in hoarding this asset. The rules for allocating the scarce transmission resource during times of stress were far from optimum. As a result, markets may have appeared more constrained to buyers than they were in actual physical terms. With a mix of planned and market driven behaviors interacting with genuine concerns about physical shortages, the actual state of the available physical system is difficult to perceive.

3. INFORMATION INADEQUACIES: A new market with a multitude of new complex transactions and physical constraints on how to deliver the services placed a premium on information. Unfortunately, such information was not available. There is no centralized, reliable source of information. Information is much more difficult to gather for system aggregators. Moreover, the brokers who were the sources of information have interests that would be served by distortion of information.

4. DEFAULTING ON OBLIGATIONS: Given the tense conditions in the market, it is not surprising that defaults played an important role in some of the price spikes of 1998. Although a relatively small amount of power was at issue, confidence in the ability of firms to meet commitments was shaken when important traders could not deliver. The problem was compounded by the numbers of transactions and the structure of deals. Utilities invoked contract clauses that sent additional customers scrambling for replacement power. Even when they bought power, they could not be sure what price they would be charged or whether they would be able to receive the power.

5. INEXPERIENCE: Facing a chaotic and uncertain situation, bidders drove the price of power up. These extreme reactions to the defaults can be attributed in part to inexperience on the part of those who were caught short. Individuals may learn, but they remain vulnerable if they need supplies. Unfortunately, those seeking to manipulate markets may learn too. It is an open question which learning process will proceed most rapidly.

6. THE UTILITY OBLIGATION TO SERVE CREATES SEVERE VULNERABILITY IN A RESTUCTURED MARKET: One of the key factors that drove prices up was the need of utilities to ensure physical availability of supplies. For all the focus on market efficiency, the ultimate test of electricity service is keeping the lights on and some entities still have the obligation to ensure that they do. Consumers have generally supported this fundamental principle of utility service because electricity service is just too important to be unreliable. However, in an unfettered market for supply there are adverse consequences of this behavior. It is difficult for utilities to exercise restraint, as supplies become tight. Restructuring may require much more attention to interruptible rates to facilitate the response to tight markets. Interruptible customers must be prepared to actually be interrupted. New incentives call into question whether utilities will live up to the non-price terms of their interruptible tariffs, given the high price they can fetch for released power or be avoided for purchased power. Interruptible rates based on a regulated system that did not contemplate frequent interruptions may be inadequate. Rewards for releasing power need to reflect the higher prices being paid at peak. Given the greater frequency and higher prices occurring in the marketplace, new rules on who is cut back and who is not and how customers are compensated are needed.

C. MARKETS FAILED ON THE SUPPLY-SIDE

The above factors are "innocent," if not benign, "inept," if not illegal - a constellation of events, obligations and frailties that tightened the marketplace and drove participants to pay high prices without a purposeful intent to do so. If these were all the factors that created the price spikes there would be still considerable room for public policy intervention to reduce tensions, but there is more to the story.

There is also considerable evidence that structural flaws in the marketplace led to breakdowns in conduct - purposeful actions taken to increase the prices paid to market players who would thereby increase their profits. Self-interested behaviors that sought to exploit and reinforce the frenzied behavior in the market were not prevented by market structures or regulatory institutions.

1. EXERCISE OF MARKET POWER: Market power can be exercised in generation and transmission markets because they are thin. With little spare supply available at certain times and few competitors, restriction of supply is feasible. The analysis of bidding behavior indicates that market power was being exercised. Suppliers with market power watch the price rise, well above the level of costs, but do not sell because they are confident that there are not enough other producers who can enter the market. Transmission capacity was taken off the market with the declaration of emergencies. A practice of swapping electricity allowed owners of transmission bottlenecks to raise the price of power that was allowed to flow. Since the same entities that control the bottlenecks also own generation whose price is increased, manipulation of transmission can be highly profitable.

2. MANIPULATION OF TRANSACTIONS: The increase in the number of transactions was compounded by the nature of many of the transactions. Daisy chains passed power through a long line of sequential owners without ever physically being delivered, except by the last owner. This adds no new supply to the market. At least some of the transactions on which the market was built were fabrications - deals in which the buyer and seller were one and the same. This institutional structure was clearly implicated in the price run-up when financial transaction increased apparent demand. In tight markets traders' financial problems add to those of entities bidding for power. Entities with need for physical power compete with entities with financial needs for power, but the underlying physical supply and demand have not changed. Utility entry into the market through affiliates raises separate questions of conflicts of interests. Favoring affiliates with access to transmission is only one potential manner in which incumbents can control the market to their advantage. Industrial customers complained that incumbent utilities had diverted their own supplies to maximize profits.

The number of suppliers and their ability to bring product to market must be sufficient to deliver workably competitive markets. Highly concentrated markets with bottleneck facilities that lack open access rules make the market prone to the exercise of market power. Market institutions should be developed before, not after trading begins so that conduct is transparent and disciplined by market forces. Undeveloped information and trading mechanisms are prone to manipulation. When abuse occurs under such circumstances, it is no accident; it is the result of bad public policy choices or poor policy implementation.

A. BASIC CONDITIONS MUST BE ADEQUATE TO SUPPORT COMPETITION

Policy makers have an obligation to ensure that the basic conditions are adequate to support competition before they unleash market forces. It is simply irresponsible to create markets that suffer from significant problems like inadequate capacity at the outset.

1. EVALUATING CAPACITY BEFORE MARKETS OPEN: Policymakers should assess and take responsibility for the supply/demand balance as they enter the restructuring process. If conditions are tight, they must address the problem of how to expand capacity or provide consumers with options that protect them from price spikes that are likely in such circumstances.

2. ENSURING ADEQUATE INFRASTRUCTURE: Policymakers should ask whether the infrastructure is adequate to support the competitive market model. If the transmission system is inadequate to support competition, it is irresponsible to introduce competition. Economies of coordination are still strong in the physical marketplace. Therefore, rules of the operation of the grid must be in place to ensure its reliability. To the extent that coordination can lower costs, through lowering reserve requirements, this function too should be preserved.

3. SUPPLY-SIDE PERFORMANCE PENALTIES: One of the areas where incentives have become a problem in the transition to competition is the motivation to prevent blackouts at all costs and then pass the costs through to ratepayers. The consequences of failing to keep capacity on line have increased, but there are no penalties. Changes in this approach may be necessary to ensure consumers are protected from undisciplined bidding behavior.

4. CONFRONTING THE INELASTICITY OF DEMAND: Consumers express a strong commitment to reliability and an aversion to price shocks. It may be difficult to accomplish both goals at historic price levels in unfettered commodity markets. Improved incentives to participate in peak load management programs should be examined. Making interruptible rates more effective for those who are interested, and facilitating aggregation or other forms of participation, may elicit more demand reduction. However, we do not believe that residential consumers want to see their prices tracking the commodity price of electricity or be forced to evaluate and implement complex hedging instruments. For firm residential and small business customers, it may be just as important to develop programs that let them enjoy stable prices without sending utilities plunging into markets to avoid blackouts. Proposals to build peaking reserves at stabilized prices become attractive, if markets are going to be extremely volatile.

B. MARKET STRUCTURE

1. ACCESS TO THE HIGHWAYS OF COMMERCE: As a general proposition, vertical divestiture is the only solution that eliminates the problems of affiliate transactions. Short of divestiture, a truly independent system operator must be established. This system operator should be given the authority to run the system solely for reliability and social efficiency (lowest total social cost) purposes. Divestiture or an ISO is generally seen as a sufficient response to concentrated markets. That is, if divestiture takes place, the assumption has generally been that the generation market will not be concentrated. Alternatively, if the transmission system is operated in an open manner, enough generators will be able to enter the market to prevent the abuse of local market power.

2. SUPPLY CONCENTRATION: Markets can become very thin very quickly for a variety of reasons, and prices can rise very rapidly. Market power can be exercised for short periods of time in specific markets and result in substantial sums being transferred. It may be useful to put a halt to increasing concentration through large mergers until we have a better idea of how market structures and institutions will function under the unique conditions of the electricity industry. It is also important to monitor closely the supply, bidding and pricing behavior of generation entities even in markets where divestiture and/or open access have taken place. The basic supply and demand conditions in electricity markets may be so severe that market structures traditionally defined as competitive will break down situationally.

3. TRADING: Trading institutions must also be more highly developed quickly. The overheating of the market in 1998 reflected a fundamental lack of institutions to convey information and ensure the soundness of transactions. Measures to ensure openness and confidence in transactions should be taken. Securities and commodity exchanges impose rules to protect the public and ensure an orderly market. Power exchanges should impose rules on traders that seek to ensure transparent pricing, control the flow of trading, impose memberships criteria, require registration of participants, and manage the types of trades including issues such as short selling, margin requirements, credit requirements and option rules. There is no reason that a physically important and difficult to manage commodity like electricity should not be subject to at least the rules that are routinely found in commodity and financial markets.

C. CONDUCT

Aggressive policies to discipline abuse of market power should be implemented. Any entity that engages in actions that tend to tighten electricity markets and then seeks to exploit that situation through sales at inflated prices should be presumed to have engaged in market manipulation (see Exhibit ES-2). They should bear the burden of proving that they are not guilty of profiteering and the penalty for market manipulation should be severe. The prohibition could apply to both merchant generators (and brokers) and incumbent utilities. Penalties must be sufficient to dissuade this activity. Financial fines are only a first step. Repeated offenses should lead to suspension of trading rights and ultimately banishment from markets as a trader or broker.

D. PERFORMANCE:

Having experienced repeated spikes policymakers should also implement a series of circuit breakers to prevent the sort of abuse that has occurred. These should remain in place until regulators can affirmatively conclude that market structures are functioning in a manner that is likely to prevent such abuse. The most obvious circuit breaker is a price ceiling or cap that simply does not allow trades to take place at prices above a certain level. This is generally considered the most extreme measure. Other circuit breakers can be utilized before a cap is imposed. For example, trading above a specified price could be suspended for a period (as is the practice with the stock market). Unfortunately, since the physical movement of electricity cannot be suspended, nor is that necessarily desirable, suspension of trading could be tricky. At some point, the FERC could declare that prices above certain levels are not deemed just and reasonable and therefore, market based rates are suspended. The transactions could be allowed to move forward, but the final price would be subject to adjudication.