West-Indisch Huis, Amsterdam,
October 23, 2003
A Summary[1]
In this
ENCORE workshop two draft papers were discussed:
·
‘Designer Markets: The Art of
Restructuring’ by Roger Sherman (Department of Economics, University of
Houston)
·
‘Evolutionary Regulation: From
CPI-X towards contestability’ by Per Agrell (IAG School of Management and Core,
Université Catholique de Louvain) and Peter Bogetoft (Department of Economics,
Royal and Veterinary and Agricultural University, Denmark).
Here we
summarize the presentations and the discussion. The draft papers will be
revised by the authors and will be available on the website of ENCORE.
Designer Markets, inducing motives and providing essential facilities
in telecommunications and electricity
by Roger Sherman
The
electricity sector and the telecommunications sector were historically operated
as integrated monopolies. The sectors were then restructured with the aim of
enhancing efficiency by introducing competition. Sherman distinguishes two
sources of market incentives: (1) the “carrot” of profit and (2) the “stick” of
competitors. Competitors might achieve lower costs and win customers with lower
prices.
Industries
that came to be regulated often had facilities essential for the service that
were controlled by a single owner, largely to achieve economies of scale.
Restructuring usually introduces the carrot and stick by granting competitors
access to those once-monopolized essential facilities. The rest of the presentation
was devoted to the discussion of restructuring of telecommunications and
electricity.
In this
section Sherman paid attention to the regulation of Local Exchange Networks,
Interconnecting networks and access pricing.
Local
networks provide basic residential and commercial telephone services. In the US
local networks traditionally face rate of return regulation. Since prices were
based on the own cost of the firm, firms had little incentive to improve their
efficiency. Furthermore, firms were over investing, due to a generous rate of
return.
Price
caps as a regulation method induce both the carrot and the stick and therefore
give better incentives to the regulated firms. Nowadays price cap regulation in
the telecom industry is widely applied in the US: 70 percent of all states have
implemented some sort of price cap regulation. Sherman referred to a study of
Ai and Sappington[2] who
evaluated the effects of introducing price cap regulation as positive, since
operational costs decreased, prices became lower and firms invested more in new
technologies.
Pressure
for efficiency can also come from entry in the local network service, where
some elements are seen as essential facilities to be opened to competitors.
Access is also provided to interconnect networks so long distance companies can
complete their calls.
Connecting
separate telecommunication networks is subject to network externalities since
each caller connected to a network can reach more parties of the interconnected
network. In this case one network needs access to another network to complete
phone calls originating from its own network. In this respect the networks are
complements to each other. If the networks are interconnected, the access to
the other networks could also imply that the networks are substitutes for each
other: they can compete to provide services for all connected customers.
When two
networks operate as complementary networks a possibility of double
marginalization arises, which means that the price for consumers is higher than
a single integrated firm would have set because the two networks will each
mark-up their cost. If the networks operate as substitute networks they compete
for customers. Although this competition will tend to induce more competitive
pricing to win new customers, it might at the same time inspire firms to set
higher access fees for their network to raise the cost of the competing firm.
Foreseeing
this difficulty while seeking competition, the Telecommunications Act of 1996
called for incumbent networks to offer elements of their facilities for lease
to entrants. Having incumbents and entrants as direct competitors makes
efficient access pricing very important.
There
are two main theories of access pricing: the efficient component pricing rule
(ECPR) and Ramsey pricing. While economists generally know the latter term, the
former may need explanation. The ECPR calls for an access price that equals the
provider’s direct cost of arranging access, plus any opportunity cost of
providing that access. This pricing rule gives the incumbent the incentive to
grant access to new entrants and it incentivizes
investment of the entrants in their own facilities if their cost is
lower than the cost of the incumbent.
Because,
both ECPR and Ramsey pricing are hard to implement in practice, access-pricing
rules as set by the Federal Communications Commission (FCC) are different.
FCC
bases access prices on ‘ideal’ networks, i.e. the long run incremental cost for
each service using the best available technology. Since prices are not based on
the own cost of the firm, this gives the firm an inventive to be efficient.
Since prices based on the optimal technology access prices are low, demand for
access will develop.
Unfortunately,
these access prices destroy the incentives for the incumbent firm to invest.
For successful investment, the profit will be shared between the incumbent and
the entrants. If the investment is a failure, the incumbent will have to bear
the losses.
Since
prices are based on the optimal technology, entrants will not invest in own
networks. Sherman does not expect that the present system will last for a long
time, due to the poor investment incentives. Sherman believes that the
discussion should concentrate on the question about which elements should be
available for (unbundled) access.
An
important feature of electricity is that it cannot be stored, which means that
if demand exceeds the supply of electricity or the supply of network capacity
the system will collapse. Normally prices would coordinate supply and demand.
Due to rate of return regulation firms operating metering devices in the US are
poorly developed, so that peak load pricing is virtually absent. Therefore
demand will not respond to scarcity of supply.
Before
restructuring coordination was provided by vertically integrated monopolies.
After restructuring new institutions should provide that coordination.
First
Sherman briefly discussed the restructuring of the electricity sector. The
former monopoly is split into generators, transmission networks, transmission
coordinators, distribution companies and retail marketing companies.
In 1992
the Energy Policy Act granted independent generators access to the transmission
grids. Generating companies are open to ‘the stick of competition’. Competition
is also possible at the retail level, although competition at the retail level
is not always introduced.
Since
the authority to design the market structure is not federal, but at the state
level, competition is likely to be found in ‘high-cost’ states, while in low
cost states monopolies prevail.
An
important problem of competition in generation of electricity is the
possibility of market power. The Federal Energy Regulatory Commission (FERC) evaluates
the market power of generators. If generators have market power according to
FERC, their prices should be approved by FERC.
Initially,
concentration measures were used by the FERC to judge market power, and under
these standard, generating companies were seldom found to have market power.
But concentration measures are inappropriate to judge market power when
suppliers control separate units of capacity.
Borenstein c.s.[3] for example claims that market power could be blamed
for 59 percent of the increase in the wholesale electricity prices in
California during the crisis in the summer of 2000. Even without collusion the
physical withholding of generation capacity can be profitable and drive prices
up.
The
absence of long-term contracts is believed to increase the market power of the
generators. Recently the change to the New Electricity Trading Arrangement
(NETA) in England and Wales allowed for long term contracts. Electricity prices
in England and Wales have decreased since the adoption of NETA.
Sherman
calls another important problem in the electricity sector ‘the transmission
troubles’.
Transmission lines can only carry a certain amount of
electricity, or they will melt from excessive heat. So transmission is limited
to rated capacity levels. Congestion on the wires can then prevent transmission
into areas of high demand, called ‘load pockets’, and the resulting scarcity
can create local market power. Even without market power, when electricity
cannot be delivered into an area because of congestion, the price on the
importing side of the congestion can be high. The cost of electricity is higher
on the importing side because older and less efficient facilities must be used
there to meet demand when electricity from other locations cannot get past
congested lines.
Investment
signals from congestion charges constitute a ‘lumpy carrot’ that complicates
private enterprise investment decisions in transmission facilities. Congestion
may produce a large congestion charge on a line, but expanding the line eliminates the congestion and with it
the financial benefit of building the line. This problem of on-again, off-again
investment signals might suggest public ownership as a possible arrangement for
financing and controlling transmission investment. But some degree of private
ownership may be workable with clever arrangements that preserve efficiency
incentives.
According
to Sherman, retail competition is the US has not been a great success yet. He
explained that the introduction of the restructuring lowered prices, because
asset values did fall. To win the support of the incumbents for the
restructuring nearly all states paid the incumbents the stranded costs. Prices
since the restructuring reflected the efficient level, which made new entrance
difficult. The passage of time might reduce the problem caused by the
remuneration of stranded cost.
Discussion
Jules Theeuwes (ENCORE) complimented Roger Sherman with his article, which
important contribution is a combination of theory and the application of the
theory in practice.
The
first question of Jules Theeuwes concerned the point made by Sherman that the
current guidelines for access pricing by FCC were not sustainable. According to
Theeuwes, regulated prices should be fair and should provide incentives for
innovation. Theeuwes asked Sherman to explain his view on the possibility to
set prices that would satisfy these restrictions.
Sherman
answered that the bottlenecks of the current pricing guidelines could be
decreased if the FCC would be more selective in rewarding network elements with
the status of essential facilities. Only those elements that are needed to
ensure competition should be selected as essential facilities. Entrants will
then choose to invest in their own facilities. A good market analysis by FCC
could therefore narrow the negative effects of the current regulation. In
addition, the regulation based on the ideal network ensures that only firms
that have taken perfect decisions would get reimbursed their cost. Sherman
expects the access prices to go up to ensure a sufficient level of investment
in the future.
Regarding
the electricity market, Jules Theeuwes said that in Europe the single European
Electricity Market is a dream. He asked the opinion of Sherman on this idea of
a single market, in particular: ‘are local markets good enough and how to
achieve a single market?’
Sherman
pointed out that a single market does not exist in the US. However, a
development towards market integration can be seen. The integration of the PJM
markets (PJM ensures the reliability of the largest centrally dispatched
control area in North America by coordinating the electricity flow in all or
parts of Delaware, Maryland, New Jersey, Ohio, Pennsylvania, Virginia, West
Virginia and the District of Columbia) is a good example of this development.
At this very moment, there are about 10 Regional Transmission Companies in the
US. Most of them serve several states, but there are also regional transmission
companies, for example, the one that serves 85 percent of the territory of the
state of Texas.
Masja Stefanski
(Ministry of Economic Affairs) asked Sherman to give his opinion about the
current discussion in the Netherlands about the privatization of the network
companies.
Sherman
answered that the ‘transmission troubles’ suggested that public ownership of at
least the transmission networks might be worthwhile to consider. On the other
hand the transmission troubles could be mitigated by smart regulation of
congestion. Sherman added that in the US there is a long time commitment to
private ownership. Private ownership in generation works very well.
In
addition, Martijn van Gemert (Dutch
Energy Regulator) asked whether smart regulation could solve the transmission
troubles.
Sherman
answered that there exist smart pricing regimes for congestion, but they are
complex and the investment incentive problem is still not solved. He added that
regulation models could not solve all problems. He referred, as an example, to
the heavy storms in Virginia where half of the electricity network was
destroyed. A price-cap system would make it very difficult to restore the
network.
Evolutionary Regulation: From
CPI-X towards contestability
by Peter Bogetoft
The
market mechanism is the guiding economic principle in Western society. A
competitive market ensures, by ‘the invisible hand’, that the right services
are produced at the right time and place (coordination), that parties have
individual incentives to make coordinated decisions (motivation) and that
coordination and motivation are provided at the lowest possible cost
(transaction cost). However, the market can fail due to, for example, the
existence of natural monopolies, externalities and public goods.
It is
generally acknowledged that activities operated by (regional) monopolies
require government intervention. Therefore, a regulatory body is typically
assigned to set price controls, which eliminates monopolistic profits and
encourages efficiency. However, due to asymmetric information about, for
example, costs and technology, the regulator faces both moral hazard and
adverse selection problems.
The
regulator has the objective to achieve the same outcome as in a competitive
market, which is to ensure coordination, motivation and minimal transaction
costs. However, the regulator knows that regulation can never achieve market
outcomes; at best the regulator can mimic a competitive market by carefully
using elicited information.
The
regulator can use more or less ingenious and effective tools to achieve a
solution for the market failure of a natural monopoly. Bogetoft stressed that
theorists and practitioners only start to know about the dynamics of
regulation. The dynamics of regulation is especially important since regulated
firms face very long-run investment programs, which makes regulation a long-run
game. Behavior of regulated firms is a response to current regulation, past
regulation and anticipated regulation. We only start to learn about what drives
technological development, the time needed to learn and to adopt strategic
behavior and endogenous preferences. Fundamental questions are: Why are firms
inefficient? Do firms choose their inefficiency in a rational way? How is
inefficiency affected by regulation?
Bogetoft
distinguished four different types of solution.
- Cost recovery regimes[4]
In this
regulatory mechanism, the regulator sets prices based on the incurred costs of
the regulated agents. The incurred costs are often marked up by a ‘fair and
reasonable’ rate of return. Unless subject to good information regarding cost,
this approach results in skewed investment incentives, perverse efficiency
results and lack of managerial effort.
As
already mentioned in the previous presentation by Sherman, regulatory
authorities worldwide are gradually abandoning this approach.
- Fixed price regimes
Fixed
price regimes (also called price caps, revenue caps or RPI-X regimes) are
launched to solve the apparent problems of the cost recovery regimes. In these
so-called “high powered regimes” the regulator caps the price or revenue of
regulated firms for a certain pre-determined period (commonly set on five
years). Since the difference between price and incurred cost leads to a profit,
the regulated firm is incentivized to minimize its cost. However, the resetting
of prices at the start of the new pre-determined period with the hindsight of
realized cost savings limits the efficiency savings. Furthermore, fixed prices
regimes might lead to bankruptcy of good firms (when prices are set too low) or
to excessive rents for firms (when prices are set too high).
Empirical
research has shown that firms do play strategic games under price-cap regimes
in anticipation of future price reviews.
This
approach is used in different forms in various countries like Great Britain,
Norway, Chile and New Zealand.
-Yardstick regimes
The idea
behind yardstick competition is that prices charged by the firms are not
directly linked to the incurred own costs but depend on the performance of
other firms. For example, Shleifer (1985)[5]
proposes a simple scheme in which the regulator sets a price-cap for a firm
based on the average per-unit cost of other firms in the sector. Firms are
allowed to keep the difference between the price-cap and their realized cost.
It is easy to see that such schemes work quite similar to competitive markets.
Since all firms have the incentive to beat the average cost, the average cost
will be minimal. Yardstick competition can yield first-best solutions under the
condition that exogenous factors facing the regulated firms are correlated. In
practice this means that firms should be comparable; at least the relative
differences in the exogenous operation conditions have to be known.
When the
data have limited noise and the underlying cost structure and technology of the
firms are complex, multi-dimensional yardstick competition can solve problems
caused by differences in exogenous operating conditions. In his paper and in
the presentation Bogetoft showed some examples of DEA[6]-based
yardstick regimes.
When
regulators use yardstick regimes they should address the issues of scale and
scale adaptation (for example by mergers).
Examples
of applied yardstick competition include the regulation of water and sewerage
in England and Wales and the regulation of distribution networks in electricity
in the Netherlands.
- Contestability
A simple
method to elicit accurate cost information while assuring agents’ participation
is to arrange franchise auctions. The idea is to award the delivery rights and
obligations based on an auction among qualified bidders. The regime leads to
perfect information about the heterogeneity of the exogenous operating
conditions, since prices may vary across different franchises.
Due to problems of investment incentives and the transfer of assets (the
succession of franchisees), auctions are likely to be used sparingly in Europe
in the near future.
Bogetoft
shortly paid attention to the possibility to use DEA-based auctions to be able
to compare multidimensional bids.
A
regulatory regime can only be optimal with respect to a certain industry in a
certain market situation. Since investments are planned and implemented under
the consideration that assets might be used during a long period, regulatory
commitment is very important. On the other hand, both regulators and firms have
a desire for flexibility in the regulatory regime, since they both will learn
about changes in technology, economic tendencies and political preferences.
Optimal cost and the level of service are unknown to firms and regulators, and
might change over time.
After
acknowledging the dynamic nature of regulation, the question arises whether it
is possible to develop a consistent approach toward changes in the regulatory
regimes.
The
basic idea behind accommodating a change in the regulatory regime is to allow
for a transition period, in which firms are allowed to choose amongst different
regulatory regimes.
In his
presentation, Bogetoft gave an example of the transition from a price cap regime
towards a yardstick regime.
Discussion
Martijn van Gemert
(Dutch Energy Regulator) shortly highlighted the Dutch development in
regulation from virtually no regulation towards the application of a yardstick
regime. Van Gemert emphasized the importance of reliable data for regulation.
There are only a few electricity distribution companies in the Netherlands. Van
Gemert asked Bogetoft to comment on the probability of collusion.
Bogetoft
answered that collusion is regarded to be a problem in yardstick competition. A
simple model could make it more difficult to cheat in the case of a few firms.
Another option is to use ‘optimal net configurations’ as a benchmark for
observed behavior.
Cees van Gent (Dutch
Competition Authority) made a comment that it would be difficult to maintain an
effective cartel on cost, since cost is more stochastic than prices.
Bogetoft
answered that to his knowledge there is no empirical proof of collusion in a
yardstick system.
Van Gemert
explained that under a system of DEA-based yardstick competition most firms
would be evaluated against the most efficient firm in the population. He asked
if agents would still want to participate in the regulation game, since an
average firm might expect a loss.
Bogetoft
answered that the regulation model could ensure that any firm would be able to
survive and earn its investments back. He pointed out that the number of
acceptable bankruptcies under regulation is merely a political choice.
Van Gemert
explained that in the Netherlands all regulated electricity networks are
assumed to have the same potential for productivity change. In the regulation
system in the Netherlands, all changes in correlated exogenous factors are
passed through the tariffs to the consumers. Van Gemert asked Bogetoft whether
this assumption that all firms have the same ability to change their
productivity could be tested.
Bogetoft
answered that testing these assumptions is difficult. He mentioned that both
consumers and firms might be able to rearrange the extent of risk sharing by
insurance contracts.
Jeroen Hinloopen
(ENCORE) mentioned that firms under yardstick competition have an incentive to
innovate. By licensing their innovation, firms have a monopoly on this
innovation.
Bogetoft
agreed and added that it is this monopoly that will encourage firms to
innovate, like in any competitive market.
Arno Meijer (Dutch
Telecom Regulator) mentioned that there are different methods of benchmarking,
for example, the World Bank seems to advocate Stochastic Frontier Analysis (SFA).
He asked Bogetoft if he preferred DEA or SFA.
Bogetoft
answered that his choice would depend on the situation. He prefers DEA in
situation where the underlying cost structure and technologies are complex and
the data are reliable.
December 2003, ENCORE /Misja Mikkers
[1] This is an ‘unauthorised’ summary in the sense that
we did not double-check with the speakers and discussants. The summary reflects
as objectively as possible what has been said at the workshop. But neither the
speakers nor the discussants and certainly not the institutions with which they
are affiliated are responsible for the contents of this summary. The contents
of this summary and especially its errors and omissions are the sole
responsibility of ENCORE.
[2] Ai, Chunrong and David E.M. Sappington (2002), The
Impact of State Incentive Regulation on the U.S. Telecommunications Industry, Journal
of Regulatory Economics, 22: 107-32.
[3] Borenstein, Severin, James B. Bushnell and Frank A.
Wolak (2002), Measuring market inefficiency in California’s Wholesale
Electricity Market, American Economic Review, December, 92, 1376-1405
[4] Including cost-of-service, cost-plus and
rate-of-return regulation.
[5] Shleifer, A. (1985), A Theory of Yardstick
Competition, Rand Journal of Economics, 16, p. 319 – 327.
[6] Data Envelopment Analysis