Chapter 1-1

Why Deregulate?

The propensity to truck, barter, and exchange one thing for another
.. .is common to all men.

Adam Smith
The Wealth of Nations
1776

A N THE BEGINNING THERE WAS COMPETITION—BRUTAL AND INEFFI-

CIENT. Between 1887 and 1893, twenty-four central station power companies
were established within Chicago alone. With overlapping distribution lines, competi-
tion for customers was fierce and costs were high. In 1898, the same year he was
elected president of the National Electric Light Association, Samuel Insull solved
these problems by acquiring a monopoly over all central-station production in
Chicago. In his historic presidential address to NELA, Insull explained not only
why the electricity business was a "natural monopoly" but why it should be regu-
lated and why this regulation should be at the state level, not the local level. Insull
argued that
exclusive franchises should be coupled with the conditions of
public control, requiring all charges for services fixed by public
bodies to be based on cost plus a reasonable profit.
These ideas shocked his fellow utility executives but led fairly directly to regulatory
laws passed by New York and Wisconsin in 1907 establishing the first two state
utility commissions. Reformers of the Progressive era also lent support to regulation
although they were about equally supportive of municipal power companies.1 Their
intention, to hold down monopoly profits, was at odds with Insult's desire to keep
profits above the competitive level, but both sides agreed that competition was
inefficient and that providing electricity was a natural monopoly.2
1. See Platt (1991) for information on central station companies and Samuel Insull. The quotation (p. 86)
is from a contemporary account of Insull's address. Platt describes early competition as follows: "The
Chicago experience of rate wars, distributor duplication, and torn-up streets presented an alternative that
was attractive to virtually no one." For state commissions and progressives, see Rudolph & Ridley (1986).
2. Smith (1995) relies on Gregg Jarrell to conclude "regulation was a response to the utilities' desire to
protect profits, not a consumerist response to monopoly pricing." But Knittel (1999) tests causation by
utilities and consumers and finds no significant correlation between profit change and regulation after
correcting Jarrell's endogeneity problem. This result would be expected from an analysis of profit when
two equal forces have opposite motivations with respect to its level.
CHAPTER 1-1 Why Deregulate ? 7

On the scale of an isolated city, provision of electricity is a natural monopoly

and requires regulation or municipal ownership, but as transmission technology
developed, it brought new possibilities for trade and competition. The earliest
electric companies, for instance Brush's company which lighted New York's
Broadway in 1880, integrated generation with distribution, and, in fact, sold light,
not electricity. Edison initially did the same, installing the light bulbs in the homes
he lit and charging by the number of bulbs installed. Westinghouse introduced high-
voltage transmission using alternating current (AC) technology to the United States
in 1886, and by 1892 Southern California Edison was operating a 10-kV transmis-
sion line 28 miles in length. This, too, was an integrated part of a full-service utility.
Integrated utilities remained natural monopolies for many years while expansion
of the high-voltage transmission network continued, mainly for purposes of reliabil-
ity. Eventually the entire Eastern United States and Eastern Canada were united
in a single synchronized AC power system. By operating at extremely high voltages,
this system is able to move power over great distances with very little loss, often
less than three percent in a thousand miles.
Regulated, vertically integrated utilities were well established by the time the
transmission system made substantial long-distance trade possible. As a conse-
quence, trade was slow to develop, but the existence of the grid made the de-
integration of the electric industry a possibility.3 Generation could now be split
off to form a separate competitive market, while the remaining parts of the utilities
remained behind as regulated monopolies. By 1990, encouraged by a general trend
toward deregulation, the de-integration trend in electric markets was underway
in a number of countries. Today, more than a dozen semi-deregulated electricity
markets are operating in at least ten countries, with several operating in the United
States.
In spite of this apparent success, many fundamental problems remain. After
ten years of operation, the British market has declared itself a failure and replaced
all of its market rules. In a single year, the California market managed to cost its
customers more than ten years of hoped-for savings. Alberta (Canada) is worried
over the results of its recent auction of generation rights that brought in much less
revenue than planned. New York saw prices spike to over $6,000/MWh in 2000,
and the New England ISO had to close its installed capacity market due to extreme
problems with market power. But initial problems do not prove deregulation is
doomed; some markets are functioning well. A closer look at fundamental argu-

3. See Joskow (2000b, 16) for a similar view of the unimportance of "the demise of natural monopoly
characteristics at the generation level" and the importance of the expansion of the grid, and Ruff (1999)
for an alternative view of the role of the grid and transmission pricing.
PART 1 Power Market Fundamentals

ments for and against deregulation may help explain why such mixed results might
be expected.
Chapter Summary 1-1: Improvements in transmission, rather than changes
in generation technology, have removed the natural monopoly character of the
wholesale power market in most locations. This makes possible the replacement
of regulated generation monopolies with deregulated wholesale power markets.
In principle these can be more efficient than the old-style regulation. In practice,
California has proven bad deregulation to be worse than mediocre regulation, and
England has demonstrated that mediocre deregulation can bring cost-saving
efficiencies to a badly regulated generation monopoly.
In the short run, power-market problems tend to be more dramatic than the
benefits. The problems are primarily the result of two demand-side flaws: the almost
complete failure of customers to respond to relevant price fluctuations, and the
customer's ability to take powerfromthe grid without a contract. As fundamental
as these are, it is possible to design a workable market around them, but it does
require design as well as extensive and clever regulation. Recent U.S. history has
shown that there are three impediments to such progress: politics, special interests,
and overconfidence. The last is largely due to a dramatic underestimation of the
problem.
Section 1: Conditions for Deregulation. Deregulation requires the market
not be a strong natural monopoly. One view holds that small efficient gas turbines
have overturned the natural monopoly of large coal plants. Yet today's competitive
suppliers are far larger than any coal plant, so if the size of a large coal plant were
problematic for competition, today's markets would be uncompetitive.
Section 2: Problems with Regulation. Regulation can provide strong cost-
minimizing incentives and can hold prices down, but it must trade off one against
the other. Competition can do both at once. In practice, regulators hold prices down
near long-run average costs but leave cost-minimizing incentives too weak. The
result is high costs and high prices.
Section 3: The Benefits of Wholesale Competition. Competition provides
full strength cost-minimizing incentives and, at the same time, forces average prices
down toward their minimum. It may also encourage efficient retail prices.
Section 4: The Benefits of Real-time Pricing. Competition may induce real-
time pricing, which will reduce consumption during periods of peak demand. This
will reduce the need for installed capacity and, if extensively adopted, should
provide a net savings of about 2% of retail price. Although this could be achieved
 CHAPTER 1-1 Why Deregulate ? 9

easily under regulation, competition will provide some additional incentives, but
their consequences are still unclear.
Section 5: Problems with Deregulating Electricity. Contemporary electricity
markets have inadequate metering. Consequently it does not make sense for load
to respond to price fluctuations, and bilateral contracts cannot be physically
enforced in real time. As a result demand can and sometimes does exceed supply,
and competitive pricing is impossible at crucial times. Theseflawsresult in high
prices that must be limited, and they provide ideal conditions for the exercise of
market power. Electricity markets are also extremely complex and prone to prob-
lems with local market power due to the inadequacies of the transmission system.

1-1.1 CONDITIONS FOR DEREGULATION

Scale economies make it possible for natural monopolies to produce their output
more cheaply than a competitive market would. A 1-MW power plant is not very
efficient, and there is no way to produce power cheaply on this small a scale. A
10-MW power plant can always do better. Efficiency continues to increase signifi-
cantly to about the 100-MW level but ever more slowly beyond this level. It used
to increase to about 800 MW, and it was once assumed that nuclear plants would
be the most economical and their most efficient size would be even greater. If these
economies of scale continued, the cheapest way to provide California with power
would be to build a 25,000-MW power plant and a few smaller ones to handle load
fluctuations. But a large single power plant could not support competition. A
competitive market necessarily utilizes smaller plants and would therefore have
higher production costs. Consumers would have to pay more if a natural monopoly
is forced to operate as a competitive industry with small-scale plants.
Efficiency gains from the operation of multiple plants are another possible
source of natural monopoly. Even if very large plants are not more efficient, large
generating companies may be. A large company can hire specialists and share parts
and repair crews. If multiplant efficiencies continue to large enough scales, a
competitive market would again be less efficient than a monopolist.4
If a monopolist can produce power at significantly lower cost than the best
competitive market, then deregulation makes little sense. The lack of a natural
monopoly is a prerequisite to successful deregulation, or at least, the condition of
natural monopoly should hold only weakly.

Did Cheap Gas Turbines End the Natural Monopoly?

One popular argument for deregulation claims technical progress has recently
nullified the conditions for a natural monopoly in generation. This view assumes
generation had previously been a natural monopoly because the most efficient size

4. See Joskow and Schmalensee (1983, 54) for a discussion of firm-level economies of scale.
10 PART 1 Power Market Fundamentals

power plant was approaching 1000 MW, and new technologies have made 100-MW
plants almost as efficient.
If this argument were correct, then an 1000-MW supplier must in some sense
be a monopolist, and the market must need suppliers that have capacities smaller
than 1000 MW to be competitive. But in this case, deregulation must certainly have
failed in the United States because every market contains suppliers with capacities
exceeding 1000 MW. Yet no one who suggests small efficient plants are a necessary
condition for competition seems worried by the presence of huge suppliers in the
new markets.
The beliefs that the most efficient size power plant must be quite small, and
that competitive suppliers can own many such plants are contradictory. Most likely
the former is incorrect, at least in markets with peak loads of over 5000 MW.
Fortunately, vast transmission grids have made such large markets the norm. When
small efficient plants are necessary for competition, suppliers with total generating
capacity greater than the most efficient size plant should be prohibited. Greater
threats of natural monopoly conditions come from the economies of multiplant
companies and weaknesses in the power grid that effectively isolate "load pockets"
during peak load conditions.

1-1.2 PROBLEMS WITH REGULATION

The most common argument for deregulation is the inefficiency of regulation. There
can be no quarrel about its inefficiency, but it does not follow that deregulation
will be better. Deregulation is not equivalent to perfect competition which is well
known to be efficient. Electricity markets have their own inefficiencies that need
to be compared with the inefficiencies of regulation.
To date, such comparisons have been largely speculative. The most decisive
answers inevitably are based on the least information. One side claims regulation
is essential because electricity is a basic need. What about housing? That need is
even more basic, and housing is 99% deregulated. The other side claims competition
provides incentives to reduce costs, while regulation does not. What about the many
regulated utilities that have provided reliable power for many years for less than
60/kWh? Do they lack all cost-minimizing incentives?
One argument posits that when a regulated utility makes a bad investment,
ratepayers pick up the tab; but, when an unregulated supplier makes a bad invest-
ment, stockholders pick up the tab. This analysis is myopic. Particular losses fall
on the stockholders of particular companies, but the cost of capital takes into
account the probability of such mistakes, and every mistake increases the estimated
probability. Like all costs, the cost of capital is paid for by consumers. Not only
does the cost of capital average in the cost of mistakes, it also adds a risk premium.
To the extent stockholders of regulated utilities are sheltered, they demand less
of a risk premium than do stockholders of unregulated suppliers. With competition,
there may be fewer mistakes, but the mistakes will be paid for by consumers, and
a risk premium will be added.
CHAPTER 1-1 Why Deregulate ? 11

Regulation has two fundamental problems: (1) it cannot provide a strong

incentive to suppliers as cheaply as can a competitive market, and (2) regulatory
bodies themselves do not have proper incentives. Well-trained regulators could
provide much better regulation. But for government to provide competent regula-
tion, the political process would need to change. Thefirstproblem, that of incentives
provided by regulators, is more susceptible to analysis.

The Regulator's Dilemma

Truly competitive markets do two things at once; they provide full-powered
incentives (1) to hold price down to marginal cost, and (2) to minimize cost.
Regulation can do one or the other but not both. It must always make a trade-off
because suppliers always know the market better than the regulators.5
This trade-off is the core idea of modern regulatory theory. Perfect cost-of-
service (COS) regulation is at one extreme of the regulatory spectrum. It assures
that, no matter what, suppliers will recover all of their costs but no more. This
includes a normal rate of return on their investment. Perfect COS regulation holds
prices down to long-run costs but takes away all incentive to minimize cost.6 If
the suppliers make an innovation that saves a dollar of production costs, the
regulator takes it away and gives it to the customer.
At the other extreme is perfect price-cap regulation. It sets a cap on the supplier's
price according to some formula that takes account of inflation and technical
progress, and it never changes the formula. Now every dollar saved is kept by the
supplier, so its incentives are just as good as in a competitive market. But it's
difficult to pick a price-cap formula that can be fixed for twenty years at a time.
A perfect (very-long-term) price cap must always allow prices that are well above
long-run cost to avoid accidentally bankrupting suppliers. Consequently, prices
will be too high.
The reader unfamiliar with the theory of regulation may be tempted to invent
clever ways for the regulator to provide full cost-minimizing incentives while
holding prices down to cost, but all will fail. The inevitability of this trade-off has
been established repeatedly and with great rigor; however, the trade-off can be
improved by improving the regulator's effective knowledge.7 The main technique
for making the trade-off is to adjust the price cap more or less frequently. Constant
adjustment produces COS regulation while extremely infrequent adjustment
produces pure price-cap regulation. In between, incentives are moderately strong
and prices are moderately low. If the regulator has a fair amount of information,
this trade-off can be quite satisfactory, but it will never equal perfect competition.

5. Suppliers having better information than the regulators is at the root of the regulatory trade-off problem
which, though fundamental, is too complex to discuss here.
6. In reality, cost-of-service regulation does provide incentives in two ways: regulatory lag (see Joskow
2000b) which is discussed shortly, and the threat of disallowed costs.
7. For instance, yardstick regulation compares the performance of one regulated firm with other similar
firms, thus giving the regulator some of the benefit of the other firms' knowledge without requiring the
regulator to know details (see Tirole 1997).
12 PART 1 Power Market Fundamentals

Regulation in Practice
Competition can hold average prices down to long-run costs while putting full
strength pressure on cost minimization. At best, regulation does a decent job of
both but does neither quite as well as competition. But how does regulation work
in practice?
Regulation tends to err in the direction of driving prices down toward cost. In
fact, most regulators believe this is their entire job and would implement pure COS
regulation if they could. Fortunately, it's just too much bother to re-adjust rates
continuously, so the result is roughly a price cap that gets reset about every three
years. This inadvertent "regulatory lag" is a major factor in saving COS regulation
from providing no incentive at all. It provides some incentive for cost minimization,
but less than would be provided with an optimal trade-off. Even that is too little
by the standard of a competitive market. In practice, regulation has typically done
a passable job in the United States and could do much better if the effort spent
deregulating were spent improving regulation.

1-1.3 THE BENEFITS OF COMPETITIVE WHOLESALE MARKETS

Competition provides much stronger cost-minimizing incentives than typical "cost-
of-service" regulation and results in suppliers making many kinds of cost-saving
innovations more quickly. These include labor saving techniques, more efficient
repairs, cheaper construction costs on new plants, and wiser investment choices.
Distributed generation is an area in which innovation may be much quicker
under competition than under regulation; cogeneration is one example. Regulated
utilities found such projects extremely awkward at best, so avoided them. A
competitive market easily allows the flexibility that such projects require.
The other advantage of competition is its ability to holdprice down to marginal
cost. This is less of an advantage simply because traditional regulation has stressed
this side of the regulatory trade-off. Sometimes price minimization can still be a
significant advantage. Again cogeneration provides an example. Once regulators
decided to encourage it, they needed to price cogenerated power. A formula was
designed with the intention of mimicking a market price. Naturally, political forces
intervened and the result was long-term contracts signed at very high prices (Joskow
2000b). These gave strong, probably much too strong, incentives for cogeneration.
A competitive market can get both incentives and prices right at the same time.
While holding down prices, competition also provides incentives for more
accurate pricing. Because it imposes the real-time wholesale spot price on the
retailer's marginal purchases, wholesale competition should encourage real-time
pricing for retail customers. This can be done easily by a regulated retailer, but a
competitive retailer should have an added incentive to provide the option of real-
time retail pricing because that would reflect its costs.
 CHAPTER 1-1 Why Deregulate ? 13

1-1.4 THE BENEFITS OF REAL-TIME RATES

One view holds that the main benefit of competition will come from the demand
side of the market more than the supply side. The price spikes of the wholesale
market will be passed on to customers—at least for marginal consumption and will
cause customers to curb their demand when the price is highest and generation is
most costly. This will allow fewer generators to be built and will reduce the total
cost of providing power. A competitive market will pass this savings on to consum-
ers.
Arthur Wright of Brighton, England, invented the real-time meter which Samuel
Insull heard of while visiting his homeland in 1894. He sent Louis Ferguson to
Brighton to study its use. Insull soon replaced most of his meters with Wright real-
time meters and by 1898 was installing them with every new residential hookup
(Platt 1991). Although Ferguson invented the notion of charging according to the
time of peak demand, this was rarely used. Instead customers were charged for peak
demand (a demand charge) and total energy. This combination became known
as a Wright tariff. The purpose of such metering was to improve the system's load
factor (average load over peak load), and it did so. Central station load curves from
Chicago show that their load factor improved from 30.4% in 1898 to 41.7% eleven
years later. The Wright tariff is not real-time pricing but has a similar, though
generally weaker, effect on peak demand.8
The cost-saving effect of real-time pricing cannot be doubted, but how great
is it? Unfortunately it is much smaller today than it was in 1898, and that may
explain in part why residential customers no longer have real-time meters. Today's
load factors are typically near 60%, although, in a system like Alberta's where
regulators have imposed heavy demand charges, the "load factor" can approach
80%. Real-time pricing could do the same. It can never cause load to become
completely constant (100% load factor) because this would put an end to real-time
price fluctuations, and there would no longer be a reason for customers to shift
their consumption off peak. Thus load-shifting caused by real-time pricing is self-
limiting. Assume that the shift would proceed halfway if real-time pricing were
fully implemented. Changing the load factor from 60% to 80% gives about a 25%
reduction in needed generation capacity. This is dramatic, and perhaps a bit optimis-
tic, so for numeric convenience, assume the reduction is only 24%.
With an 80% load factor, most load would be baseload because high real-time
prices reduce peak load. Peaking generators cost roughly half of what an average
generator costs per installed megawatt. The reduction in generation fixed costs
should be in the neighborhood of 12%. Because fuel costs are about as large as
the fixed costs of generation, total wholesale power costs will be reduced only about
6%. (Real-time pricing shifts load to off-peak hours but its effect on total energy
consumption is minimal and could work in either direction.) Wholesale costs are
about % the cost of retail power, so retail costs are reduced by about 2.25%.

8. For a comparison of demand-charges and real-time rates and an explanation of why real-time rates are
crucial for reducing market power, see Borenstein (2001b).
14 PART 1 Power Market Fundamentals

Peak transmission use often occurs during shoulder hours, not peak hours, so
there is no easy proof of a transmission cost savings. But for simplicity assume
that this savings is % as great as the saving in generation. That brings total savings
to 3%.

Result 1-1.1 Savings from Real-Time Rates Would Be Small

Fully implemented real-time pricing would improve load factors roughly from
60% to 80%. This would reduce the cost of supply by approximately 2.25% of
retail costs mainly due to savings in peak-load capacity. Additional consumer
costs of accomplishing the load shift would likely more than cancel any transmis-
sion savings for a net savings of about 2% of retail costs.

This 3% reduction in supply cost is offset by a cost increase on the demand

side. Shifting load to off-peak hours is costly for customers as it requires the
purchase of smarter appliances and changes in consumption that they find undesir-
able. In fact customers will shift load up to the point where the marginal cost of
shifting is just as great as the marginal savings on their electricity bill. A rough
estimate puts the total cost to consumers at about l/3 of the savings and this does
not include the cost of the real-time metering and billing. The net savings from real-
time metering should be in the neighborhood of 2% of total cost of delivered
power.9
A second question about real-time pricing as an argument for deregulation is
why it cannot be done under regulation. Regulated systems have for years computed
real-time system marginal cost. Technically, it would be no problem for a regulated
utility to install real-time meters and charge customers marginal-cost prices.
California is now testing this possibility (Wolak 2001). It may be that regulators
simply lack the will to do this or the ability to carry out the details effectively. On
the other hand, it may take more will and cleverness to implement a competitive
wholesale market than to implement real-time pricing.

1-1.5 PROBLEMS WITH DEREGULATING ELECTRICITY

Electricity is a peculiar product. It is the only product that is consumed continuously
by essentially all customers. In fact it is consumed within a tenth of a second of
its production and less than a tenth of a second of power can be stored as electrical
energy in the system.10 These physical properties result in a product whose marginal
cost of production fluctuates rapidly and, thus, whose delivered cost also fluctuates
rapidly. No other product has a delivered cost that fluctuates nearly this rapidly.

9. Significant savings in transmission and distribution should not be expected because transmission lines
are typically used most heavily during off-peak hours, and both have very large fixed-cost components that
will be unaffected by deregulation.
10. More kinetic energy is stored in rotating generators, much more potential energy can be stored by
pumping water up hill and vastly more energy is stored in local fuel supplies, but all of these stores of
energy must be converted to electrical energy by the process of generation before they can be delivered.
 CHAPTER 1-1 Why Deregulate ? 15

Two Demand-Side Flaws

Although real-time metering began in the late 1800s, it has been discontinued for
residential customers, and almost no industrial or commercial customers see real-
time prices. Consequently, almost no customers respond to the real-time fluctuations
in the delivered cost of power.
Even with this demand-side flaw, the market could operate in reasonably close
accord with economic principles if not for the second demand-side flaw, the ability
of a load to "take power from the grid without a prior contract with a generator"
(Ruff 1999, 28; FERC 2001b, 4). If bilateral contracts could be enforced by
physically cutting off customers who exceed their contracts, the market could
function almost in alignment with the theory of competitive markets. In no other
market is it impossible to physically enforce bilateral contracts on the time scale
of price fluctuations.

Demand-Side Flaw 1: Lack of Metering and Real-Time Billing

Demand-Side Flaw 2: Lack of Real-Time Control of Power Flow to Specific Customers

The first demand-side flaw causes a lack of demand responsiveness to price

or, technically, a lack of demand elasticity. The second demand-side flaw prevents
physical enforcement of bilateral contracts and results in the system operator being
the default supplier in real time.
Because demand responds only minimally to price, the supply and demand curve
may fail to intersect, a market flaw so severe it is not contemplated by any text on
economics.11 The system operator, as default supplier, is forced to set the price,
at least when supply fails to intersect demand. It can also improve the market by
setting price under slightly less dire circumstances. Presently, all power markets
operate like this and will continue to do so until very-short-run demand elasticity
is significantly improved.
These are not just theoretical problems. While the average cost of production
is about $35/MWh, and the maximum cost with new equipment is about $ 100/MWh,
prices in the $ 1,000 to $ 10,000 range have occurred in many markets. All four of
the U.S. markets have formal price caps, and the Midwest market has informal caps
set by system operators who refuse to buy required reserves when the price gets
too high. With the extreme demand inelasticity caused by the first flaw, scarcity
alone would produce high prices, but the flawed demand side coupled with scarcity
also produces ideal conditions for market power which pushes prices still higher
(Joskow 2001a).
Dramatic though these flaws are, it should still be possible to design a well-
functioning market. But it does require design! Deregulating power markets is called
"restructuring" in the United States because the resulting competitive markets have
more federal regulations than the regulated markets they replaced (Borenstein and
Bushnell 2000). In the long run, the demand side of the market should develop
enough price elasticity to clear the market at a finite price. There is a good chance

11. See Jaffe and Felder (1996), Kahn et al. (2001), and Green (1998).
16 PART 1 Power Market Fundamentals

that eventually price spikes will be low enough to need no caps. This change in
market structure should be encouraged from the start. A responsible deregulation
of electricity would first fix the demand-side flaws and then start the market—they
are cheaper to fix than the problems they have already caused.

Complexify and Local Market Power

There are two other fundamental problems with deregulating electricity: complexity
and local market power. A power system is a delicate, single machine that can
extend over millions of square miles. Every generator in the system must be
synchronized to within a hundredth of a second with every other generator in the
AC interconnection. Voltage must be maintained within a 5% limit at thousands
of separate locations. This must be accomplished on a shared facility, half of which
(the grid) must be operated for the common good and half of which (the generators)
are operated for hundreds of different private interests.
Complexity can be overcome by a sufficiently well designed set of market rules,
but the problem of local market power may need to be solved, at least for the
present, by interventionist means. So far, it has been. More than half of the genera-
tors in the California ISO were declared "must run," meaning sometime during
the year they were crucial to system operation and, therefore, had such extreme
market power there was no choice but to regulate their price. San Francisco and
New York among other cities are "load pockets;" they require more power than
they can import. As a consequence, the two generators in San Francisco would
have extreme market power during peak hours every day if they were not regulated.
These generators are required for their "real" power production, but most must-run
generators are required for their "reactive" power, a concept not well understood
by regulators (see Chapter 5-2).
The most difficult and costly problems with new electricity markets are mainly
matters of market structure as opposed to market architecture (see Chapters 1-7
and 1 -8). When this is understood, and demand-side flaws and the problems with
market power and transmission are squarely faced, adequate solutions will probably
be found. Then wholesale power markets should prove superior to regulated
monopoly generation.