Electricity and Its Regulation

Important characteristics of electricity limit the possibilities for markets. First, reserve power plants must always be operating to instantly replace generators or transmission lines that fail. Centralized control (usually by computers) is required to meet both predictable and unforeseen changes in regional conditions. Power cannot be economically stored, and area-wide blackouts occur if production either exceeds or falls short of demand for as little as a second. Second, duplication of facilities is inefficient because a single high-capacity line minimizes both capital cost per megawatt (MW) transferred and line losses due to resistance. A typical large utility (or group of them) is responsible for reliability and economical operation in its defined “control area.” Each control area is interconnected with neighboring ones to facilitate emergency support, coordinated operations, and power purchases and sales. Third, an injection of power flows through the entire network according to Ohm’s and Kirchoff’s laws. Unlike water or gas, it cannot be directed down a single path. If a Utah generator sells power to a Wyoming user, only a small fraction of it flows directly between them. Because the power from Utah flows everywhere, it can overload lines in California and force Californians to curtail their own beneficial transactions.

Scale economies and reliability concerns left electricity dominated by large, vertically integrated utilities; that is, utilities that generated, transmitted, and distributed power. Direct competition had vanished by the 1920s as municipal franchise grants left nearly every city with a single utility. Between 1907 and 1940, all states formed regulatory commissions whose authority replaced that of cities. The reasons for the change are unclear: utilities may have sought protection from opportunistic city governments, or from competition in general. “Cost of service” regulation sets retail rates to recover expenses and give a “fair” return on capital. Problems in allocating common costs, as well as politics, allow latitude in setting rates for different customers. State regulators generally also require utilities to serve all customers and to plan facility additions in anticipation of growth. The Federal Energy Regulatory Commission (FERC) oversees “wholesale” or “bulk” transactions that occur prior to state-jurisdictional retail sales. The Federal Power Act requires that wholesale prices (including transmission charges) be cost based, but in practice, FERC simply accepts prices set by markets that meet its standards for competition. FERC’s general policy has been to expand the role of markets and to decrease direct regulation subject to the law’s limits, regardless of which party controls the government.

Electricity’s ownership structure is complex. In 1998, America’s 239 corporate utilities made 74.9 percent of retail sales; its 2,029 governmental (mostly municipal) utilities made 15.0 percent; and 912 cooperatives made 8.6 percent.4 Most states exempt governmental utilities and cooperatives from regulation and allow them to set their own rates. Municipal utilities such as Los Angeles and San Antonio own generation and transmission, but most are small, transmission-dependent resellers of purchased power. Corporate utilities owned 66.1 percent of generating capacity in 1998, governmental utilities 10.7 percent, cooperatives 3.1 percent, and nonutility generators 11.9 percent. Federal hydroelectric facilities account for 8.2 percent of capacity.5 Their output is preferentially allocated by law to municipals and cooperatives at below-market rates.

From the 1940s through the 1960s, regulation seemed to function well enough that few questioned its efficacy or looked to market alternatives. The states prohibited retail competition in utilities’ assigned territories, but demand grew and engineering advances consistently lowered production costs of new generators. Even if utilities were operating inefficiently, consumers saw falling rates, and regulators usually allowed investors attractive returns. The industry grew more capital intensive, culminating in the 1960s with a move to nuclear generators that many expected to produce power too cheap to meter.

By the 1970s, the good times were over. Oil prices rose with OPEC, and price controls had produced shortages of natural gas. Nuclear plants experienced massive cost overruns, aggravated by the 1979 Three Mile Island accident, and consumer interests pressured regulators to disallow utilities from recovering these costs with higher rates. Technological progress in coal-fired generation slowed, and utilities became major targets of a growing environmental lobby. Environmental regulation prohibited or delayed some new plants and raised the costs of existing ones. Some regulators required demand-reduction programs in lieu of new generation, and others imposed newly politicized planning processes on their utilities.

At about this time, regulators, consumers, and utilities began reconsidering markets. Systems with high-cost, unbuilt, or delayed generators might be able to buy power more cheaply than they could produce it. New transmission and control technologies allowed reliable flows over distances of one thousand miles. Interutility exchanges grew faster than retail sales in every year between 1980 and 1998.6 Some are under long-term contracts, while others are daily or hourly “spot” trades. They may be for energy (flowing power) or capacity (rights to a generator’s output). They can be firm (reliably backed up), interruptible, or in the form of options. Contracts for transmission service, known as “wheeling,” must match delivery and receipt obligations. Beyond energy flows, since the 1970s the greater risks of unstable prices and regulatory uncertainty led to increases in generation and transmission undertaken by utility consortia.

The Public Utility Regulatory Policies Act (PURPA) of 1978 opened wholesale markets to nonutilities. Prior to PURPA, utilities could refuse to interconnect or purchase from nonutility generators at will. Part of the Carter administration’s conservation policy, PURPA would encourage industrial generation from waste heat (“cogeneration”) by requiring utilities to purchase it at the “avoided cost” of building and operating their own plants. The consensus in the late 1970s was that fossil fuels would remain expensive, particularly relative to the average cost of utility owned generation fleets. This was thought to make self-supply with a fossil-fuel-burning generator uneconomic for many industrial users, the exceptions being those—thought to be few—that had high heat requirements and little choice but to get them from fossil fuels. Instead, oil prices collapsed over the 1980s, and decontrol of natural gas both ended its shortage and produced two decades of low prices. New gas-fired generators under 100 MW capacity became as cheap to operate as coal-fired plants ten times their size, and they had lower costs of environmental compliance. Experience with cogeneration led to larger nonutility plants whose output was cheaper for utilities to purchase than to generate themselves. Since 1992, utility purchases from nonutilities have grown more than twice as fast as retail sales.7

The Energy Policy Act of 1992 (EPAct) removed a final obstacle to generator competition by allowing FERC to order transmission owners to carry power for other wholesale parties. Nonutility generators now had access to any willing counterparty in the region, and a new industry of power marketers that handled less than 8 million mwh in 1995 traded more than 1.5 billion in 1999.8 Transmission access allowed municipal utilities to become independent and build the power supplies they wanted. Retail customers, however, still lacked choices. They could enjoy the benefits of competition only if state regulators allowed them to leave existing utilities.

In 1994, the California Public Utilities Commission (CPUC) became the first to investigate choice for retail customers. With power costs 50 percent above the national average, even the CPUC’s research staff blamed overregulation and proposed greater reliance on markets. Competitive suppliers and many retail users welcomed the proposal, but California’s three large corporate utilities understandably resisted. They claimed that competitive prices would not allow them to recover about $20 billion in above-market PURPA contracts and unamortized nuclear costs. The CPUC agreed that utilities had some claim to these “stranded costs,” which they had incurred in the expectation that a regulated monopoly would remain.

In 1996, California’s legislature unanimously approved Assembly Bill 1890, a comprehensive but internally inconsistent compromise that authorized an Independent System Operator (ISO) to ensure nondiscriminatory use of transmission that utilities still owned but might use to stifle competition. Utilities had to divest most of their in-state gas-fired generation and were to purchase all power from markets operated by the new California Power Exchange (PX) and the ISO. They could transact for deliveries no more than a day ahead, and other risk-management activities were prohibited. Assembly Bill 1890 froze or discounted retail prices until 2002. The utilities had until then to recover most stranded costs in the difference between fluctuating wholesale prices and government-frozen retail rates. Customers choosing nonutility service were billed for their shares of stranded costs.

For two years after the markets opened in April 1998, supply and demand kept prices low and allowed utilities to recover substantial stranded costs. This ended in the summer of 2000, however. Poor snows left the Northwest with little of the surplus hydropower it normally sent south. Also, natural gas supplies became limited and expensive; because gas-fired generators tended to be the highest-marginal-cost producers, and because market prices were equal to the marginal cost of the highest-cost producer, electricity became quite expensive. Another contributing factor was the increase in the price of pollution permits by several hundred percent. California’s long resistance to constructing new power plants was one factor that kept prices from their usual decline when summer 2000 ended. Allegations that generators exercised market power and traders manipulated PX and ISO rules remain in controversy and continue to be litigated. Utilities faced insolvency as rising power costs met frozen retail prices. Short-term energy prices were high everywhere in the West, but only California required its utilities to use only short-term markets and not to pass higher prices on to customers. Dependence on short-term markets, as well as the technical nature of the rules, encouraged suppliers and utilities alike to game the system.

As California’s situation deteriorated in 2000, FERC reluctantly capped short-term prices. One of its two largest utilities lost creditworthiness, and the other went bankrupt, as did the PX. State government took over power purchasing in January 2001 and, by midyear, had entered long-term contracts for most of what utilities could not produce in the plants they still owned. As new generation came online in mid-2001 and hydroelectric conditions improved, market prices fell to their precrisis levels. Again, Californians are locked into uneconomical contracts whose costs must be allocated. The state’s utilities are attempting to revert to their former monopoly roles, and other interests are attempting to again involve the state in their resource planning.

California’s reforms were founded on a gamble that low energy prices would allow utilities to recover their stranded costs while rates were frozen. Customers in some states that reformed more rationally are reaping substantial benefits. All large northeastern states now allow choice, as do Illinois, Ohio, Michigan, and Texas. Most northeastern states avoided California’s risks by adding fixed stranded cost payoffs to everyone’s bills and allowing customers access to markets that existed for years prior to reform.

Customer choice varies with market conditions. In October 2004, 78 percent of the power consumed by industrial users in New Jersey was competitively supplied, by about thirty nonutility providers.9 Utilities have ceased supplying nearly half of the power consumed by industrial users in New York and Massachusetts.10 Small customers can also benefit—25 percent of Pittsburgh’s residential users have new suppliers.11 New York state already has thirty-three competitive nonutility sellers vying for customers.12 We are still a long way from fully competitive markets (and California is retrogressing), but even the limited competition available now is producing substantial benefits.13