Perspective

Two Cato analysts suggest a return to the past-vertical integration, but now with no state regulators.

The defeat of the energy bill in the Senate last year has thrown electricity restructuring back on its heels. There clearly is no consensus among politicians or academics regarding how this industry ought to be organized or how it might best be regulated. Finding our way out of this morass requires a reconsideration of how we got to this dismal point in our regulatory journey. Doing so suggests a surprising series of conclusions about what has gone wrong and where to go from here.

The Case for Restructuring

What, in theory, are we supposed to get out of restructuring? From an economics standpoint, there were two major problems associated with the old system.

First, because investment in capital received a guaranteed return, total generation investment was excessive and skewed toward capital-intensive facilities. Couple that with the one-time enthusiasm for nuclear power (once thought of as a progressive energy source that would be "too cheap to meter") and the growing hostility to coal-fired generation, and it's no surprise that some states moved strongly toward nuclear-with costs generally much higher than antici-pated.¹ Introducing market forces into the utility industry would eliminate the bias for capital-intensive projects by introducing uncertainty about returns.

Second, prices for electricity did not serve their usual role of signaling to consumers the marginal costs of additional consumption. Instead, they served solely as a device to recover costs. Thus, electricity prices were wrong all the time. They were too low on peak and too high off peak. Market forces, it was hoped, would introduce marginal-cost pricing and as a result reduce peak demand, increase off-peak demand,² and reduce the needless political fighting (most notably, the eternal fight over more supply versus less demand) that inevitably arises in electricity markets because of the absence of prices as a signaling device.

Average voters and the politicians that listen to them, however, do not associate markets in electricity with positive outcomes. Accordingly, anyone who believes market forces ought to play a larger role in electricity must argue convincingly that:

• The California meltdown and the Northeast blackout were not the result of market forces;
• The low costs of the states still under the old regulatory regime are not the result of regulation; and
• Efficiency improvements not possible in the regulated would take place in a truly deregulated world.

The California Story

During 2000 and 2001, a large supply reduction in hydropower and weather-related demand increases (a hot summer and very cold winter) increased electricity and natural gas prices in California. Those price increases were exacerbated by NO_x air pollution regulations in the Los Angeles basin, some design features of the California auction bidding system, and retail price controls.

The price controls were particularly harmful; they encouraged generators to price high because there would be no reduction in demand as a consequence of their pricing behavior. Moreover, retail price controls prevented utilities from passing on their higher costs to consumers, which caused the utilities to suffer a financial meltdown. Generators, in turn, increased prices to account for the possibility of not being paid. From November 2000, the California story is a financial meltdown story: Wholesale prices had a large credit-risk component.³

The central lesson from California is not that market forces have been tried and have failed, but rather that partial deregulation (wholesale deregulation combined with rigid retail prices) is an extremely dangerous institutional design.

The Blackout Story

From our perspective, the Aug. 14, 2003, blackout illustrates the difficulty of managing externalities on the grid. Markets were not responsible for the blackout, but the shift during the last 30 years from a world consisting of balkanized vertically integrated utilities to independent power producers and vertically disintegrated power service providers has increased the number of players whose behavior must be coordinated to maintain satisfactory operation of the North American transmission system.

While blackouts also occurred in the old regime, it's certainly true that, with greater interconnectivity between service territories, the chances of cascading blackouts might be enhanced by the introduction of restructuring as it is currently conceived.

About Those Low-Cost States

The low-costs states are low-cost largely because they did not change very much from the 1965 . Some never abandoned the use of coal in the production of electricity, while others had continued access to cheap hydropower. None of the states aggressively implemented long-term, fixed-price, independent power contracts.

Moreover, traditional rate regulation appears to benefit consumers through the use of weighted average pricing for electricity. Why this appears so-and why it is but a mirage- requires a quick review of some economic fundamentals.

In a free market, the market prices of commodities are determined by the most expensive source of supply necessary to meet demand. As a consequence, inframarginal sources of supply with lower costs receive economic rents.

In an unregulated electricity market, then, marginal sources of electricity-such as daytime peaking units-would need to earn at least a normal return. This implies that those facilities with lower costs whose supply is limited (such as old coal-fired units exempt from plant-specific emission controls under the 1970 and 1977 Clean Air Act amendments) and hydropower facilities (whose supply can't be expanded) would receive rents in an unregulated market.

Rate regulation by the states, however, currently suppresses those rents. Consumers are charged a weighted average of generator costs rather than the market price, which would be at least the marginal cost of the most costly unit necessary to meet demand.

Thus, in a free market, the proportion of electricity produced by coal or hydropower would not affect prices if neither is the marginal source of power. But in regulated electricity markets, cheap inframarginal power does lower electricity prices to consumers because prices are weighted averages of producer costs rather than marginal costs of the most expensive producer. Thus, regulation would seem to play a role in the low costs of the states that maintain the old regulatory regime.

To most, the main consequence of weighted-average prices (that is, lower peak prices) seems like a good thing. But weighted average prices induce increased demand that must be met by state regulatory systems. The cost of that extra supply is paid for through excessive off-peak prices to cover the capital costs of generators that are underutilized at all times except peak.

In a free market, peak users would pay much higher costs. But, in turn, they would reduce peak demand and the variance between peak and off-peak use would be lower, resulting in less total capital investment and fuller utilization of the capital stock invested in generation. The owners of inframarginal assets, like old hotels on the ocean in July, would receive rents only during peak demand periods, and those rents would pay for the capital costs. In off-peak periods, generators, like hotels on the ocean in December, would price very close to marginal cost.⁴

Efficiency Gains

In regulated markets, economists usually can easily demonstrate that little or no efficiency rationale ever existed for regulation, and that efficiency gains suppressed by regulation would occur if deregulation occurred. But electricity regulation is more complicated. The old regime described above would appear to benefit consumers whose use is much more on peak than the average user (to the detriment, of course, of those consumers whose use is more off-peak than the average user, and to the overall efficiency of the system).

And unlike most other markets, electricity markets have characteristics that are difficult to manage through property rights and contracts. Regulation therefore has at least the possibility of a plausible rationale. For example, the alternating current (AC) grid is a commons. That is, the physical reality of the grid does not coincide with current private property rights or the 50-state regulatory schemes that govern the grid. Because power added by any generator on an AC transmission system follows all paths, favoring those with least resistance rather than the shortest distance between generator and customer, bilateral contracts between any willing seller and buyer of electricity affect all other buyers and sellers within each interconnected system in ways that are not captured by prices. The proper way to manage those externalities is the subject of great dispute.

Also, transmission additions confer benefits across all generators and consumers on the grid and thus have public-good characteristics.⁵ But developing property rights and prices that internalize those characteristics is difficult.

Traditionally, the commons problem was addressed through monopoly-franchise vertical integration. Trade between vertically integrated utilities was never very large and was governed by barter arrangements rather than markets. And where trade was extensive, voluntary arrangements such as the PJM arose to manage the flows across separately owned transmission systems through contract. Thus, historically, the commons characteristics of the grid did not create large externality issues.

But since the Energy Policy Act of 1992 and FERC orders 888 and 889-which facilitated the development of widespread trading on the grid, particularly by non-vertically integrated merchant generators-the mismatch between the physical reality of the grid and its current governance structure have become an important problem.

Solving the Public-Good Problem

What are the possible solutions to the public-good nature of the transmission system? The most commonly discussed possibility is aggressive regulation by FERC through regional transmission organizations and standard market designs to eliminate the discrepancy between the commons nature of the transmission system and the current fragmented system that governs it.

The problem with this solution is that it employs lots of lawyers to create a half market that leaves retail, state-level regulation intact. It also confuses rather than clarifies incentives in the governance of transmission by separating ownership from control.⁶

This problem manifests itself most clearly when it come to the subject of transmission investment. Given the inability of investors to capture the full benefits of their investment, won't investment in transmission capacity always prove sub-optimal?

Nobel laureate economist Vernon Smith believes the problem is surmountable.⁷ He argues that new transmission is a "club good" that facilitates the ability of generators to get their product to market. Consortia of generators could fund new investment and, in turn, get rights to inject or take power from the system in proportion to their financial contributions. If existing generators lose money because the new transmission investment alters their ability to sell power, so be it.

MIT economist Paul Joskow, however, is skeptical:

"Transmission investment decisions do not immediately strike me as being ideally suited to relying entirely on the invisible hand. Transmission investments are lumpy, characterized by economies of scale and can have physical impacts throughout the network. The combination of imperfectly defined property rights, economies of scale and long-lived sunk costs for transmission investments, and imperfect competition in the supply of generating services can lead to either under- or over-investment at particular points on the network if we rely entirely on market forces."⁸

A second possibility is drawn from petroleum economics: a unitization contract. Petroleum producers, after all, faced a problem analogous to the commons nature of the AC transmission system because surface property rights often did not coincide with the geological characteristics of petroleum reservoirs. This created incentives to drill and pump fast before other surface owners did the same, because no one represents the interests of the entire petroleum-rich geologic formation.

A unitization contract is a set of payoffs to all existing surface owners that induces them to give up their production autonomy. It is a genuine improvement if operation of the reservoir by one operator produces so much more revenue in present value that it compensates all existing owners to give up their rights and still leaves a surplus.⁹

Many tough questions, however, remain. Is there a set of payoffs to all existing players in electricity transmission (including state regulatory regimes and incumbent utilities) that would induce them to go away and turn over operation of their systems to a welfare-maximizing operator in return for a contractually determined share of the increased profits? What plan would the welfare-optimizing operator implement? Is the plan achievable through private action, or are transaction costs prohibitively high? And, if they are high, is coercion by FERC likely to achieve the same outcome?

First, how large are the unexploited efficiency gains in the existing system that a unified operator could achieve? Hale et al. provide one example of the possibilities in the East. They demonstrate that several small transmission investments reduce peak power prices in the summer across several states. It would appear that the gains to consumers far exceed the costs of the investments, and yet the links have not been built because no one represents the beneficiaries across numerous state and utility boundaries.

The price discrepancies between states suggest another source of gains to trade that are not exploited in the currently balkanized state-based system. For example, in 2002 Kentucky's electricity cost 4.3 cents/kWh while New York's cost 11.3 cents.¹⁰ But efficiency gains exist if and only if the low price in Kentucky, for example, reflects marginal costs and not weighted-average costs. If Kentucky were connected to New York via new extensive transmission connections, for instance, the expanded Kentucky output would probably have costs greater than 4.3 cents/kWh because the main source of low prices is cheap inframarginal coal generation, and supply can't be expanded because it is the result of old sources (under the Clean Air Act) with supply that cannot be expanded by definition.

If gas is the marginal source of electricity everywhere and coal is inframarginal, then prices would not vary across states in an unregulated market because the price of gas-fired output would set the market price everywhere. If correct, this implies that gains to trade not occurring under the current balkanized system are much smaller than many believe. Accordingly, the fight between the old regime and a restructured regime (that is, the case for a transmission-intense versus balkanized system) is a fight about wealth rather than efficiency.

Back to the Future?

A third possibility, heretofore dismissed by market-oriented analysts, suggests itself: to go back in time to vertical integration, no merchant generators, and very limited trade-that is, back to the old regime. This arrangement, of course, is subject to the bias toward capital intensity and the problems associated with weighted-average pricing, but it at least avoids all the problems associated with mandatory open access.

It might seem odd for the authors, two Cato policy analysts, to entertain a re-embrace of New Deal-style regulation, but the case for such a policy is surprisingly strong even from a free-market perspective.

First, it's not at all clear to us that the economic problems of the old regime were greater than the economic problems involved in transforming the transmission grid from what was essentially a series of private roadways into one large public commons.

That's particularly true given that the problems associated with regulation are less today than they were 30 or 40 years ago because incentive-based or price-cap regulation has replaced traditional rate-of-return regulation.¹¹ "While such regulation introduces an undeniable bias toward capital-intensive facilities, the profitability of such facilities today is far less attractive compared with natural gas-fired facilities than it was in the 1960s and 1970s. In short, we have a hard time believing that, with a return to the old regime, nuclear power plants would once again get built.

Moreover, the inefficiencies associated with public utility commission oversight of the industry were overstated for the simple reason that the ability of such commissions to effectively constrain utility operations or pricing behavior was overstated.¹²

Second, it is not at all clear to us that the economic gains at the end of the restructuring rainbow are worth anywhere near the political and regulatory effort being expended to secure them. As noted above, there are few gains to trade available through interconnectivity. Regardless, one way to get the advantages of the old regime while still allowing more electricity trade between service territories is to promote the more extensive use of direct current (DC) transmission links between AC systems that have one owner and thus no externalities. DC links end the commons problem.¹³

One reason we have a far-flung AC system is because it was a low-cost system under guaranteed return regulation. Smaller AC systems with DC connections between them cost more, but such a design reduces externalities and management requirements. The problem to getting there from here is cost. George Loehr, a member of the New York State Reliability Council, estimates that it would cost $7 billion to $8 billion to break up the Eastern Interconnection into 10 smaller interconnections linked by DC lines.

The other pot of economic gold-the introduction of real-time pricing-does not, in theory, and has not in fact, followed from the restructuring experiment. We might just as easily secure those gains by mandating real-time pricing regimes on local, vertically integrated, publicly regulated utilities.

Getting From Here to There

To execute such a transition, Congress would be best advised to simply rule that state regulation of the electricity business is an unconstitutional interference with interstate commerce-a precedent established when Congress pre-empted state trucking regulation. Congress would then remove any legal barriers to vertical reintegration of the industry and any requirement that grid owners open their wires to parties under regulated terms and conditions. Service territories, however, would no longer be protected, and barriers to entry would be eliminated.

We don't expect this to happen in the foreseeable future. But we do expect that the case for restructuring as it is currently conceived will continue to come under increasing political and economic stress. Market-oriented analysts should beware: The case forward is a lot murkier than generally is acknowledged.

Endnotes

1. It is not clear whether nuclear power was imprudent from the start or was made excessively costly by federal safety regulation (in the form of the Nuclear Regulatory Commission). In 1975, Resources for the Future projected that the total costs of nuclear plants in 1985-88 would be cheaper than the total costs of equivalent coal plants. See William Spangar Peirce, (Westport, Ct.: Praeger Publishers, 1996), pp. 216-217. A set of costly nuclear plants came on line during the early 1980s, and electricity rates rose 60 percent from 1978 to 1982. See Caleb Solomon, "As Competition Roils Electric Utilities, They Look to New Mexico," , May 9, 1994, p. A1. By 1990 nuclear plants had total costs that were about double those for coal plants (Peirce, p. 216). Not all nuclear plants are more expensive than coal-fired plants. Peirce reports (pp. 217-218) that the least expensive nuclear plants have total costs lower than the cheapest coal plants but that at every other point in their respective distributions, nuclear plants are more expensive.
2. In 1996, if implemented, full utilization of conventional steam-electric "baseload" facilities off peak would have resulted in a 25.5 percent increase in power production and a similar percentage decrease in price. See Michael T. Maloney, Robert E. McCormick, and Raymond D. Sauer, "Customer Choice, Consumer Value: An Analysis of Retail Competition in America's Electric Industry," Washington, D. C.: Citizens For A Sound Economy Foundation, 1996, p. 32.
3. For a more thorough discussion of the crisis, see Jerry Taylor and Peter Van Doren, "California's Electricity Crisis: What's Going On, Who's to Blame, and What to Do," Policy Analysis 406, Cato Institute, July 3, 2001.
4. Tim Brennan, "Mismeasuring Electricity Market Power," 26 (issue 1, 2003): 60-65.
5. Douglas Hale, Thomas Overbye, and Thomas Leckey, "Competition Requires Transmission Capacity: The Case of the U.S. Northeast," 23 (issue 2, 2000): 40-45. The authors use optimal power flow analysis to demonstrate that small additions to the grid in Northeast would lower prices for consumers across several states.
6. Robert J. Michaels, "Can Nonprofit Transmission Be Independent?" 23 (issue 3, 2000): 61-66.
7. Hung-po Chao and Hillard Huntington, editors, (Boston, Mass.: Kluwer Academic Publishers, 1998), Chapter 7.
8. Chapter 1 in Chao and Huntington, p. 24.
9. Gary D. Libecap and James L. Smith, "Regulatory Remedies to the Common Pool: The Limits to Oil Field Unitization," 22 (issue 1, 2001): 1-26.
10. U.S. Department of Energy, Energy Information Administration, Electric Power Annual 2002, figure 7.4, p. 43.
11. Shimon Awerbuch, Leonard Hyman, and Andrew Vesey, , (Vienna, Virginia: Public Utilities Reports Inc., 1999).
12. See, for instance, George Stigler & Claire Friedland, "What Can Regulators Regulate?" 5, October 1962, pp. 1-16; Harold Demsetz, "Why Regulate Utilities?" 11, 1968, pp. 55-65; Richard Posner, "Natural Monopoly & Its Regulation," , February 1969, pp. 548-643; Thomas Gale Moore, "The Effectiveness of Regulation of Electric Utility Prices," 36:4, April 1970, pp. 365-375; Michael Crew & Paul Kleindorfer, "Governance Costs of Rate-of-Return Regulation," (ZgS) 141, 1985, pp. 104-123; and Michael Denning & Walter Mead, "New Evidence on Benefits and Costs of Public Utility Rate Regulation," , James Plummer & Susan Troppmann, eds. (Palo Alto: QED Research, Inc., 1990), pp. 21-40.
13. Lori A. Burkhart, "Blackouts? Never Again (But . . . .)," (Oct. 1, 2003) p. 30.

---

 

Articles found on this page are available to subscribers only. For more information about obtaining a username and password, please call our Customer Service Department at 1-800-368-5001.

"Transmission investment decisions do not immediately strike me as being ideally suited to relying entirely on the invisible hand. Transmission investments are lumpy, characterized by economies of scale and can have physical impacts throughout the network. The combination of imperfectly defined property rights, economies of scale and long-lived sunk costs for transmission investments, and imperfect competition in the supply of generating services can lead to either under- or over-investment at particular points on the network if we rely entirely on market forces."⁸