Rapport à la Régie de l'énergie – Partie 2
PARTIE 2
RAPPORT D'EXPERTISE
DE PACIFIC ECONOMICS GROUP
Original : 2005-12-23
(En liasse)
Performance
Based Regulation
for Power Transmission
Performance
Based Regulation
for Power Transmission
December 2005
Mark Newton Lowry, Ph.D.
Partner
PACIFIC ECONOMICS GROUP
22 East Mifflin, Suite 302
Madison, Wisconsin USA 53705
608.257.1522 608.257.1540 Fax
Table of Contents
I.
Introduction ............................................................................................................. 1
II.
Performance Based Regulation ......................................................................... 2
2.
Rationale for PBR................................................................................................... 2
2.1 System Design Criteria............................................................................ 2
2.1.1 Efficiency........................................................................................ 2
2.1.2 Fairness ......................................................................................... 4
2.1.3 Conclusion ..................................................................................... 5
2.2 The Regulatory Challenge....................................................................... 5
2.2.1 Cost of Service Regulation............................................................. 5
2.2.2 The PBR Alternative....................................................................... 9
3.
Rate Caps ............................................................................................................... 13
3.1 Overview ............................................................................................... 13
3.2 Precedents ............................................................................................ 14
3.2.1 United States ............................................................................... 14
3.2.2 Canada ........................................................................................ 15
3.2.3 Britain........................................................................................... 16
3.2.4 Australia ....................................................................................... 16
3.3 Rate Caps and Marketing Flexibility ...................................................... 16
3.3.1 Need for Marketing Flexibility ....................................................... 16
3.3.2 How Rate Caps Help.................................................................... 17
3.3.3 Marketing Flexibility Precedents................................................... 19
3.3.4 Evaluation .................................................................................... 32
4.
Revenue Caps ....................................................................................................... 34
4.1 Comprehensive Revenue Caps............................................................. 34
4.1.1 Description ................................................................................... 34
4.1.2 Precedents................................................................................... 35
4.1.3 Evaluation .................................................................................... 36
4.2 Non-Comprehensive Revenue Caps ..................................................... 37
4.2.1 Basics .......................................................................................... 37
4.2.2 Precedents................................................................................... 38
4.2.3 Evaluation .................................................................................... 39
5.
Index Design Issues ............................................................................................ 40
5.1 Overview ............................................................................................... 40
5.1.1 Index Formulas ............................................................................ 40
5.1.2 Inflation Measures ....................................................................... 41
5.1.3 X-Factors ..................................................................................... 44
5.1.4 Z-Factors ..................................................................................... 44
5.2 Index Design Methods .......................................................................... 46
5.2.1 The North American Approach .................................................... 46
5.2.2 The British Approach to Index Design ......................................... 60
6.
Service Quality Provisions ................................................................................ 63
6.1 Benchmarking Basics............................................................................ 64
6.2 Quality Indicators .................................................................................. 65
6.3 Quality Benchmarks .............................................................................. 66
6.4 Award and Penalty Rates...................................................................... 68
6.5 Plan Symmetry...................................................................................... 69
6.6 Informal Quality Provisions ................................................................... 71
6.7 Precedents ............................................................................................ 71
7.
Benefit Sharing Provisions ............................................................................... 73
7.1 Introduction ........................................................................................... 73
7.2 Enhanced Rate Trajectory .................................................................... 74
7.3 Initial Rate Cuts..................................................................................... 75
7.4 Earnings-Sharing .................................................................................. 77
7.4.1 Description................................................................................... 77
7.4.2 Precedents................................................................................... 78
7.4.3 Evaluation .................................................................................... 80
7.5 Plan Termination Provisions ................................................................. 82
7.5.1 Plan Term .................................................................................... 83
7.5.2 Rate Reset Provisions ................................................................. 84
III. PBR for Power Transmission ........................................................................... 88
8.
The Power Transmission Business ................................................................ 88
8.1 Transmission Service Supply ................................................................ 88
8.2 Transmission Service Demand.............................................................. 91
8.3 Implications of Power Transmission PBR.............................................. 95
9.
Precedents For Transmission PBR................................................................. 99
9.1 United States......................................................................................... 99
9.1.1 An Introduction to the FERC ........................................................ 99
9.1.2 PBR at the FERC ....................................................................... 106
9.2 Canada................................................................................................ 122
9.2.1 Jurisdiction ................................................................................. 122
9.2.2 Industry Structure....................................................................... 122
9.2.3 Regulatory System..................................................................... 125
9.3 Australia .............................................................................................. 128
9.3.1 Industry Structure....................................................................... 128
9.3.2 Transmission Regulation............................................................ 129
10. PBR for HQ TransÉnergie................................................................................ 131
10.1 Features of Québec’s Transmission Industry .................................... 131
10.1.1 HQ TransÉnergie ..................................................................... 131
10.1.2 Importance of Transmission ..................................................... 131
10.1.3 Structural Considerations ......................................................... 132
10.1.4 Québec Regulation .................................................................. 132
10.1.5 Québec’s Power Market........................................................... 134
10.2 Indicated Regulatory Strategy ........................................................... 136
10.2.1 Cost of Effective COSR............................................................ 136
10.2.2 Cost of Effective PBR............................................................... 138
10.2.3 Prospects for Performance Gains ............................................ 142
10.2.4 Conclusion ............................................................................... 144
1
I. INTRODUCTION
Hydro-Québec TransÉnergie (hereafter “HQ TransÉnergie”) provides
power transmission services in the province of Québec. Its rates and services
are regulated by Québec’s Régie de l’énergie (the “Régie”).
The Régie has
asked the company to consider performance-based regulation (“PBR”) for its
transmission services.
Pacific Economics Group (“PEG”) is the leading North American
consultancy on PBR for energy utilities.
Our clients have included several
Canadian utilities, regulatory agencies, and trade associations.
PEG was
retained by HQ TransÉnergie to consider the desirability of adapting PBR in the
near future.
This is the report on our work. In the next section, I discuss criteria for the
design of regulatory systems and then consider PBR and its advantages over the
traditional approach to regulation. I then discuss at some length the major issues
in the design of a PBR plan. In Section 9, I consider important features of the
power transmission business and their implications for PBR. There follows in
Section 10 a discussion of precedents for power transmission PBR in the United
States, Canada, and Australia. The last section of the report considers special
circumstances of the power transmission business in Québec and details an
indicated regulatory strategy for HQ TransÉnergie.
2
II. PERFORMANCE BASED REGULATION
PBR is now a well-established alternative to traditional regulation of
energy utilities. In North America, PBR plans have been approved for energy
utilities in such diverse jurisdictions as British Columbia, California, Florida, Iowa,
Massachusetts, North Carolina, and Ontario. The Federal Energy Regulatory
Commission (FERC) and the National Energy Board (NEB) use PBR to regulate
oil pipelines and some gas lines. PBR is also extensively used in North America
in other industries, most notably railroads and telecommunications. Overseas,
PBR is even more ubiquitous. Due in part to the many jurisdictions using PBR
and the varied industries involved, diverse approaches have developed. This
means that many established mechanisms are available today to craft a PBR
plan.
In Section 2 of the report, I propose a sensible set of criteria for the design
of regulatory systems and use it to discuss the rationale for PBR.
In later
sections I turn to a consideration of major plan design issues. The main issues
addressed are the choice between price caps and revenue caps, the design of
rate and revenue cap indexes, service quality provisions, and benefit sharing
provisions.
2. RATIONALE FOR PBR
2.1 System Design Criteria
2.1.1 Efficiency
One of the most important criteria for evaluating alternative regulatory
systems is their ability to promote economic efficiency. A regulatory system is
(economically) efficient to the extent that it generates the maximum possible net
economic benefits for society.
Economic efficiency has several dimensions.
One is the operating efficiency of the utility. This has a marketing as well as a
3
production dimension. The cost efficiency of the regulatory system also matters.
I discuss here the concepts of production efficiency, marketing efficiency, and
regulatory cost efficiency in turn.
PRODUCTION EFFICIENCY
Regulation encourages a utility’s production efficiency to the extent that it
induces it to produce the services that it provides at minimum cost. In the short
run, capital inputs are substantially “fixed” in the sense that adjustments in the
amounts used are quite expensive. Productive efficiency then depends primarily
on the extent to which services are provided with a minimum-cost mix of other,
variable inputs such as labor. In the long run, all inputs are variable and the costeffective use of capital is also an efficiency concern.
MARKETING EFFICIENCY
Regulation encourages a utility’s marketing efficiency to the extent that it
induces it to provide the right mix of services to the right customers in the right
amounts.
Simply put, we want utilities to play the right role in our evolving
economy. The right role is the one in which they help households, businesses,
and other customers pursue their goals at the lowest cost. A good measure of
the success of marketing is the difference between the value of services to
customers and cost of service provision. In the short run, the adjustment in the
rates and other terms of existing services to reflect changing market conditions is
the main marketing challenge. In the long run, the mix of services offered by a
utility becomes an important concern.
Service quality is an important aspect of the terms of service. Customers
care about the quality of services as well as their prices. Customers’ need for
quality varies and changes over time. Unregulated markets often involve an
array of products with different price-quality attributes. Since social benefits from
regulation depend on both price and quality, the encouragement of appropriate
quality levels is a proper regulatory objective.
The marketing efficiency of a utility does not depend solely on the terms
on which it offers services to markets that, due to the essential character of the
4
services and the lack of competitive pressures, are regulated. Utilities may also
be able to enhance welfare by supplying customers in unregulated markets.
These are sometimes referred to as “non-core” markets. Almost every utility has
some involvement in such markets. The rental of underutilized real estate under
transmission lines is a good example. Utility participation in unregulated markets
can lower prices and make valuable new products available to customers. These
advantages are especially attractive in markets, like those for local telecom
services, where additional competition is needed.
REGULATORY COST
Costs are incurred in utility regulation. These include, most obviously, the
resources (e.g. lawyers, accountants, engineers, administrators, and services) of
utilities, intervener groups, and government agencies that are dedicated to the
regulatory process. Senior company officials are also drawn into the regulatory
arena. This can impair utility performance to the extent that it distracts managers
from their operating responsibilities.
The cost of regulation varies with the
amount of work performed. Generally speaking, the cost will be higher the larger
is the number of utilities regulated and the more inherently controversial are the
activities subject to oversight.
2.1.2 Fairness
A second fundamental criterion for appraising regulatory systems is
fairness. This concerns the manner in which the benefits of utility operations are
divided among the stakeholders in the regulatory process. A minimum condition
is that the chief parties to regulation, shareholders and customers, fare no worse
under PBR than they would under traditional regulation.
A more aggressive
standard would be for the chief parties to share in the benefits of improved
performance that PBR makes possible.
In assessing the fairness of the regulatory system, it is important to
remember the outcomes that matter to stakeholders. Customers benefit, most
5
obviously, from low prices and high service quality. Customers also benefit from
rate stability and the availability of tailored rate and service offerings.
2.1.3 Conclusion
Regulation should encourage good utility performance, use regulatory
resources efficiently, and share the benefits of good performance between
utilities and their customers. Utility performance has a marketing as well as a
cost containment dimension. Good performance and a fair sharing of benefits
both point to the need for a proper balance between a utility’s operating risk and
expected return.
2.2 The Regulatory Challenge
2.2.1 Cost of Service Regulation
DESCRIPTION AND PRECEDENT
Cost of service regulation (“COSR”) is a convenient term for the traditional
approach to the regulation of North American energy utilities. Under this system,
the rates approved by a commission are expected to recover the company’s
prudently incurred cost of providing regulated services.
return on capital.
1
This cost includes a
Rate cases are held periodically in which estimates are made
of the prudent cost of capital, labor, and other inputs that are used to provide
regulated services. This becomes the base rate revenue requirement.2 To the
extent that a utility sells some products in unregulated markets, its regulated cost
will be less than its total cost. Alternatively, the revenue obtained from such
services may be netted off of total cost.
1
2
This characterization of cost of service regulation is, of course, stylized. The terminology and
precise procedure for setting rates under COSR varies considerably across regulated
industries and regulatory jurisdictions.
The volatility of energy prices has prompted some regulators to provide for a shorter lag
between the purchase of energy inputs and the addition of these costs to the revenue
requirement.
6
Once the revenue requirement is determined, it must be allocated for
recovery from tariffed services. The rate for each service recovers this assigned
cost given data on peak demand, and other billing determinants. The regulated
service offerings and rate designs require commission approval.
EVALUATION
COSR has played a vital role in the development, in several countries, of
utility industries that make service widely available at an affordable cost. Its
focus on the cost of service has two cardinal benefits. One is the satisfactory
resolution of the issue of fairness: the utility has a fair chance of recovering its
cost of service but has only a limited opportunity to earn more. The other is the
reduction in utility operating risk that results from the cost recovery. This ensures
that capital can be obtained for utility undertakings at a reasonable price.
The recovery of capital cost is especially important in the assessment of
utility risk. Capital goods provide services over many (e.g., 30-40) years. Once
capital goods are installed and become utility plant, their value in alternative
applications is often well below their cost. Companies owning such “relationshipspecific assets” are vulnerable to changes in the compensation allowed by
regulators. COSR is a good basis for a regulatory compact that reduces the
likelihood of such outcomes.
These benefits of COSR help to explain the sizable scale on which COSR
has been used. It was, for many years, the standard means by which investor
owned utilities were regulated. The chief alternative to COSR around the world
was not PBR but, rather, state enterprises. Large utility industries were built
under COSR, including those in the United States and Japan. In Canada, most
electric utilities were government-owned corporations but COSR was prevalent in
the regulation of natural gas utilities.
Despite this lengthy track record of effectiveness, there is mounting
evidence that COSR does not always achieve the maximum net benefit to
society that is achievable from utility services. One fundamental problem is the
high cost that must be incurred for regulators to learn about utility operations. If
7
they understood the changing constellation of production and marketing practices
that are ideal for the situations of specific utilities over time, they could in
principle mandate the services that should be provided and their terms.
Unfortunately, it is difficult even for experienced managers in an industry to
recognize best practices given the uncertainty that exists regarding future supply,
demand, and policy conditions. The challenge is much greater for regulators and
customers who lack operating experience in the industry. Economists call this
situation one of information asymmetry. A redressing of this asymmetry requires
substantial exchange, processing, and analysis of information.
Another
fundamental challenge in COSR is the allocation of the common costs that
utilities incur in providing miscellaneous services making this a potential source
of controversy.
Measures are naturally taken to contain the cost of COSR. One option is
to reduce the frequency between rate hearings. Another is to scale back the
scope or intensity of prudence reviews. For example, companies may be placed
at significant risk only for actions with conspicuously unfortunate outcomes. The
extent to which utilities fall short of best operating practices is rarely considered.
Regulatory cost can also be contained by restricting practices that
complicate regulation.
For example, companies may be discouraged from
offering diverse services or complex, changing rate structures. They may also be
discouraged from engaging in practices that are novel, risky, or inherently
controversial.
All of these measures can reduce regulatory costs. Unfortunately, some
of these economy measures can also compromise the productive and marketing
efficiency of utilities.
To the extent that prudence reviews are limited, for
instance, rate adjustments tend to reflect the trend in a utility’s own unit cost.
Efforts to trim costs or to improve the market responsiveness of rates and
services then lead eventually to lower rates. This weakens utility performance
incentives.
Another class of initiatives that is strongly discouraged is those
8
involving a significant risk of conspicuous failure. This would include many kinds
of innovations.
Restrictions on utility operations that are hard to regulate can also reduce
efficiency. For example, a regulatory system granting limited and inflexible rate
and service offerings hampers a utility’s ability to satisfy customers’ complex and
changing needs. Customers may not use utility services even when they can be
provided at a lower cost than the available alternatives.
The efficiency consequences of ineffective marketing are especially acute
where demand is elastic (sensitive) with respect to rates and other terms of
service. Situations in which demand is elastic include those in which customers
can obtain their service needs in other ways at a competitive cost. Demand is
also frequently elastic for incremental uses of utility services by existing
customers.
A third important source of demand elasticity is economically
distressed businesses that make extensive use of utility services. When elastic
customers do not make optimal use of a utility system, the margins from services
to them are lower than they can be and a larger share of the utility’s common
cost must be recovered from other services.
One economy measure that can increase the efficiency of COSR is a
reduction in the frequency of rate cases. As the period between rate cases,
sometimes called regulatory lag, lengthens the period of time during which the
company retains the benefits of performance gains increases.
Performance
incentives are thereby strengthened, especially for projects with longer pay back
periods. The externalization of rates also makes it easier for regulators to afford
utilities greater marketing flexibility.
The ability of a utility to operate without rate adjustments depends critically
on the extent to which its unit cost exhibits a flat or declining trend and does not
fluctuate around its trend. Unit cost is more likely to exhibit a flat or declining
trend when input price growth is slow and utilities are able to realize rapid
productivity growth. Unit cost is more likely to exhibit stability around its trend to
9
the extent that input prices and demand are not volatile and investments tend to
be spread evenly over time.
The unit costs of energy utilities, unfortunately, tend to rise over time and
are sometimes volatile. The chief reason for rising unit costs is that productivity
growth in the energy utility industries, as in most sectors of our economy, cannot
keep pace with input price inflation.
The problem of unit cost volatility is
especially pronounced in the procurement of price volatile energy inputs.
The
end result is that under COSR rate case cycles in the energy utility industries
typically do not exceed three years. Recovery of fuel and purchased power costs
often occurs even more rapidly using special fuel adjustment clauses.
This
situation is not conducive to strong performance incentives and limits the
operating flexibility that regulators are comfortable granting.
2.2.2 The PBR Alternative
The term PBR applies to a variety of regulatory mechanisms and
procedures that differ from COSR in relying less on a utility’s own cost, output,
and service quality to establish rates and other terms of service.3 An economist
might call the resultant decoupling of rates from a utility’s own operating data an
externalization of the regulatory system.
Externalization can be achieved in
several fundamentally different ways. One is to reduce the frequency of rate
cases which, as we have seen, cause a company’s rate trend to more closely
match its unit cost trend. Another is to avoid a complete unit cost true up when
such adjustments are made.
There are several “active ingredients” in this new approach to regulation.
One is automatic rate adjustment mechanisms that are established in advance of
their operation.
Such mechanisms are often represented by mathematical
formulas. The use of such mechanisms can reduce the frequency and scope of
regulatory interventions that would tend to make a utility’s rate trends more
similar to its unit cost trend.
A second source of progress is a reliance for
10
ratemaking purposes on data that are insensitive to the actions of utility
managers. Data that are useful in this regard include indices of price inflation
and information on the operations of other utilities in the industry.
To the extent that restrictions on terms of services are established by
external means, they are less sensitive to a utility’s own performance
improvement initiatives.
Utilities will then find that their performance has a
greater impact on earnings. This strengthens incentives to improve operating
efficiency. The externalization of the rate setting process also lessens concerns
about cost shifting and cross subsidies. With stronger incentives and lessened
ability to shift costs, utilities can be given more operating flexibility.
Economic research is a third active ingredient of PBR. Theoretical and
empirical research can be brought to bear on the appropriate combination of
automatic mechanisms and external data. One example is research on external
rate adjustment mechanisms that yield revenues sufficient to compensate a
competently managed utility.
Another is research on what plan provisions
provide balanced and strong performance incentives.
The combined effect of these attributes is a regulatory system that, in
many cases, can stimulate better utility performance despite lower regulatory
cost. PBR can thus increase the size of the economic “pie” that is available for
sharing between utilities and customers.
It constitutes an advance in the
“technology” for utility regulation.
While results to date have been encouraging, the state of the art is not so
far advanced that PBR is markedly superior to COSR in all cases. One problem
area is risk. Utilities under PBR will often bear more of the brunt of conventional
business risks. Their situation in this regard is much like an airline that, faced
with soaring jet fuel prices, can hope for some relief from market-based fares but
is by no means ensured full compensation.
There is, additionally, a greater
regulatory risk that restrictions on rate and service offerings will be established in
3
Other names for this approach to regulation that are sometimes used include incentive
regulation and alternative regulation (Altreg).
11
an arbitrary manner that denies a well-managed utility a reasonable chance of
recovering its cost. The recovery of capital cost is a particular concern.
Any increase in utility operating risk will ultimately be recognized by capital
markets and reflected in the cost that utilities under PBR incur to attract funds.
The increase in the cost of funds can significantly erode the net benefits of PBR.
It can also cause utilities to oppose PBR if they feel that it does not offer a
reasonable balance of risk and return. One consequence of this general problem
is that PBR still involves occasional true-ups of a utility’s rates to its cost.
Our analysis suggests that the advantages of PBR over COSR depend on
the particulars of its application. PBR will generally be more advantageous to the
extent that effective COSR is unusually costly. For example, when input price
inflation is rapid or input prices are unusually volatile, frequent rate cases are
required under COSR whereas PBR can offer automatic inflation adjustments.
COSR can also be unusually costly, as we have seen, when rate cases involve
unusually difficult issues of cost allocation, transfer pricing, or operating
prudence. Consider, lastly, that COSR is unusually costly when regulators have
jurisdiction over a large number of companies.
PBR will also be more advantageous to the extent that the effective
mechanisms that have been developed to date are amenable to implementation.
As discussed further in Section 5, for instance, a common approach to rate
indexing requires good estimates of historical productivity trends of utilities. The
calculation of productivity indexes requires, in turn, a considerable amount of
historical operating data.
Estimates of the historical productivity trend must,
furthermore, be reasonably good estimates of future productivity trends. These
conditions do not hold in every possible application. For example, good historical
data may be unavailable and past productivity trends may not continue in the
future.
A third set of circumstances that affects the relative advantage of PBR is
the opportunities for utility performance gains. The extent of performance gains
achieved depends in part on the performance gains that can be achieved.
12
Generally speaking, the potential for performance gains is greater to the extent
that more of the activities that contribute to performance can be controlled by
utility personnel. The potential for performance gains is also larger to the extent
that subject utilities are substandard performers.
A quick review of where PBR is prevalent around the world reveals that it
is indeed most popular in situations where its advantages are larger.
For
example, PBR is especially prevalent in activities that are difficult to regulate
under COSR.
Most notably, it is the standard approach to the regulation of
railroads, oil pipelines, and telecom utilities which, as we discuss further below,
need substantial marketing flexibility if they are to serve diverse markets with
varied competitive pressures from a common set of assets. In the United States,
PBR is also widespread in the regulation of natural gas procurement, which
involves a price-volatile input and difficult issues of operating prudence.
It is also interesting that PBR is the standard approach to the regulation of
newly privatized utilities. Decades of operation as public enterprises make it
likely that many of these utilities are capable of unusual short-term performance
improvement.
This gives them a margin for error in the event that rate
adjustment indexes are poorly calibrated. In North America, in contrast, most
IOUs have operated under COSR for many years. Lengthy intervals between
rate cases were common in the 1990s due to slow input price growth and capital
investment. There is thus not an expectation there, in the general case, that
PBR can trigger dramatic short term performance gains.
There is also some evidence that PBR is more common where regulators
have jurisdiction over a large number of companies. For example, PBR is still
more the exception than the rule for the regulation of energy utilities in the United
States. Most regulation of this industry occurs at the state level. With fifty states,
most regulators don’t have jurisdiction over a large number of utilities.
13
3. RATE CAPS
Most approved PBR plans involve multiyear caps on the growth of utility
rates or revenues.
This section addresses the rate cap approach.
This
approach generates stronger incentives to improve marketing performance and
often involves greater marketing flexibility provisions.
I discuss marketing
flexibility provisions of rate cap plans at some length.
The following section
addresses the revenue cap approach.
3.1 Overview
Under a rate-cap plan, restrictions are placed on the escalation of rates for
utility services. The restrictions can be placed on annual rate escalation or on
the cumulative escalation since a certain base period. The limits are called caps
since utilities are usually free to charge rates that are less than the maximum
allowed.
The mechanisms for limiting rate growth are diverse, but all have the
attribute of being external to the company’s operation. The simplest approach is
to hold rates constant for the plan duration. This approach is called, variously, a
rate freeze or rate case moratorium. A simple variant of the rate freeze is a set
of pre-scheduled rate adjustments, which may be increases or decreases.
Rate growth is also commonly capped using indexes.
Under this
approach, growth in baskets of the utility’s prices may be measured using actual
price indexes (“APIs”). Growth in each API is limited using a price cap index
(“PCI”).4 Here is a formula for limiting the growth in annual rate escalation.5
growth API < growth PCI
4
5
[1]
The useful acronyms API and PCI appear to have developed in U.S. Federal
Communications Commission proceedings.
A formula for limiting growth in cumulative escalation is
API t / API o ≤ PCI t / PCI o
where API o and PCI o pertain to the base period.
14
Price cap indexes are largely external to the company’s operations. Their
growth is typically driven by price inflation measures.
The design of such
indexes is discussed further in Section 5 below.
3.2 Precedents
3.2.1 United States
RATE INDEXING
In the United States, the first large scale PBR plan involving rate indexing
was that for class I line haul railroads under the terms of the Staggers Rail Act of
1980.6 An index was used to adjust a zone of rate freedom in which rates to
captive shippers were free from challenge.
The U.S. telecommunications
industry was another rate indexing pioneer.
The Federal Communications
Commission (FCC) played a leadership role in this regard, approving rate cap
plans for AT&T in 1989 and for interstate services of local exchange carriers
(LECs) in 1991.7
Index-based rate caps are now widely used in state-level
telecom regulation. Extended rate freezes are also common.
In the U.S. energy industry, indexing has been featured in rate plans for
several utilities.
Boston Gas was the first gas utility to operate under rate
indexing. Plans with rate indexing have also been approved for gas delivery
services of Bangor Gas, Berkshire Gas, San Diego Gas and Electric, and
Transwestern Gas Pipeline. Oil pipelines are also regulated using index-based
rate caps.
The first rate plan with indexing for a U.S. electric utility was that for the
bundled power services of PacifiCorp (CA).
Since then, plans have been
approved for the bundled power service of Central Maine Power (ME) and the
power distribution services of Bangor Hydro Electric (ME), Bay State Gas (MA),
6
7
Pub. L. No. 96-448, 94 Stat. 1895 (October 14, 1980).
“Report and Order and Second Further Notice of Proposed Rulemaking,” FCC89-91, CC
Docket No., 87-313 (April 17, 1989); and “Second Report and Order.” FCC90-314 CC Docket
No. 87-313 (September 19, 1990).
15
Central Maine Power, National Grid (MA), San Diego Gas and Electric (SDG&E)
and Southern California Edison (CA).
Extended periods of operation without rate cases have been achieved at
one time or another by many U.S. energy utilities. These sometimes result from
commitments to formal rate freezes. Rate freezes are sometimes occasioned by
PBR initiatives but also result from initiatives with other goals such as mergers or
retail competition. The FERC’s PBR plans for International Transmission and
Michigan Transco involve rate freezes.
Also noteworthy are plans for the
bundled power services of AmerenUE (MO), Black Hills Power & Light (SD),
Carolina Power and Light (NC), Duke Power (NC), several Michigan utilities, and
Florida Power and Light (FL); for the power distribution services of
Commonwealth Electric (MA), National Grid (MA and NY), and NSTAR (MA); and
for the gas distribution services of Consumers Energy and Michigan
Consolidated Gas (MI).8
3.2.2 Canada
In Canada, rate indexing began in the telecommunications industry. The
Canadian Radio-television and Telecommunications Commission (CRTC)9
approved a rate indexing plan for jurisdictional utilities in 1997.
In the electric power industry, a rate indexing plan was approved for the
power distribution services of EPCOR in the year 2000. A plan was approved for
the power distribution services of Ontario utilities in the same year and later
suspended. Union Gas was the first Canadian gas company to operate under
rate indexing. A plan has since been approved for the gas distribution services
of Torasen. Non-indexing approaches to rate caps are much less common in
Canada than in the other three countries surveyed. A PBR plan approved for
ATCO Gas North featured prescheduled rate adjustments.
8
9
The plan for National Grid (MA) involves a rate freeze period as well as an indexing period.
“Price Cap Regulation and Related Issues.” Telecom Decision CRTC 97-9 (May 1, 1997).
16
3.2.3 Britain
Rate indexing has been extensively used by regulators in Britain. It was
first applied to British Telecom in 1984. Since then, rate indexing has been
applied to the country’s electric, gas, and water utilities.
3.2.4 Australia
Rate indexing is also common in Australian regulation.
The country’s
telecommunications industry has been under price controls since 1989. Rates
for energy distributors in the states of Queensland, New South Wales and
Victoria are also subject to indexing.
3.3 Rate Caps and Marketing Flexibility
A major attraction of rate cap plans is the potential for enhanced utility
marketing flexibility.
In this section I first address the need for marketing
flexibility. There follow discussions of marketing flexibility provisions under rate
caps and their precedents.
3.3.1 Need for Marketing Flexibility
The terms on which many utilities offer their services are inconsistent with
what is known about the demands for these services and the cost of providing
them.
Services are sometimes priced below the cost of service.
This
encourages excessive consumption. When capacity is fully utilized, it will fail to
allocate available capacity to users who value the services most highly. On other
occasions services are priced well above the cost of provision. This discourages
cost effective uses of utility services, especially in cases where demand is price
elastic. Utilities also typically fail to offer the complex array of price and service
options that customers’ desire.
17
3.3.2 How Rate Caps Help
Rate caps strengthen incentives for utilities to increase the market
responsiveness of their rate and service offerings. Profits can be bolstered by
reducing rates in situations where rates exceed cost and demand is price elastic.
Utilities may also wish to use rates to discourage service requests that are
unusually costly to fulfill and to encourage requests that are less costly. To the
extent that they are external, price caps can also enhance the marketing
flexibility that regulators can responsibly allow. That is because external rate
adjustment mechanisms reduce potential concerns with cost shifting and crosssubsidization that arise when a utility’s own cost and output data are used to set
prices.
The amount of marketing flexibility afforded by a price cap plan depends
greatly on the plan details.
Two approaches are commonly used.
One is
automatic rate adjustments through the price cap mechanism. The other is light
handed regulation of optional tariffs. We discuss each in turn.
AUTOMATIC RATE ADJUSTMENTS
Price cap mechanisms like those detailed in relation [1] are one means of
affording utilities greater marketing flexibility.
The amount of automatic rate
adjustment flexibility afforded by a price cap plan depends in part on the
specification of the actual price index.
Generally speaking, an API that
summarizes the escalation in several prices gives the utility some discretion in
the implementation of the price escalation restrictions. In North American plans,
the API is typically an explicit function of the prices of the individual services that
it covers. For example, the growth in the API can be a weighted average of the
growth in the prices of individual services.
The weights would in this case
typically be the shares of the services in total revenue.
To better understand the marketing flexibility afforded by price cap
mechanisms consider, first, the case in which growth in the prices of individual
services are capped but not the growth in specific rate elements of the services.
There is in this case a separate API for each service. Each API summarizes the
18
growth in the rate elements for a service. The utility can then escalate some rate
elements more rapidly than the PCI so long as other elements grow less rapidly.
Suppose, for example, that a utility’s charge for residential energy distribution
service consists of a customer (access) charge and a volumetric charge. If the
PCI permitted the charge for the service to rise by 1%, the utility might then raise
the customer charge by 3% and lower the volumetric charge by 1%. The price
cap mechanism in this case permits automatic rate redesign.
Consider next the case in which the API summarizes the growth in the
prices for a “basket” of regulated services. A utility might then raise the prices of
some services by more than the PCI growth so long as rates for other services in
the basket grow less rapidly. The price cap mechanism in this case permits
automatic rate rebalancing as well as rate redesign.
Regulators often recognize the need for rate redesign and/or rebalancing
but wish to control it. The price cap mechanism can provide such controls. We
have already noted that it is possible to permit the redesign of rates for individual
services but not rate rebalancing by placing each service in a separate basket.
The degree of rate rebalancing can be limited by the design of service baskets.
Less rebalancing is achievable to the extent that there are multiple baskets.
More price elastic services might, for instance, be placed in separate baskets
from less price elastic services.
Side conditions are also added to mechanisms to control the degree of
marketing flexibility.
A common condition is to limit the inflation in rates for
certain services to, say, the growth in the PCI plus a fixed percent. Alternatively,
rates for certain services may be frozen.
The approach to price cap indexing that allows the least flexibility is to limit
the growth in each individual rate element of each tariff to the growth in the price
cap index. In this case, individual rate elements of tariffs will typically grow at the
same rate. This effectively discourages rate redesign as well as rebalancing.
The lesson to be learned from this discussion is that the indexing
mechanism provides a ready vehicle for controlling redesign and rebalancing of
19
rates. Given the freedom to redesign rates, utilities will move them in directions
that better reflect variations in cost causation and demand elasticity. API design
can control the pace and character of rate design.
OPTIONAL RATES AND SERVICES
A second common provision for marketing flexibility in rate cap plans is
the ability to offer optional rates and services. These can be subject to lighter
handed regulation or, in the extreme, decontrolled. Several kinds of optional
offerings may reasonably be considered.
One is optional tariffs for standard
services. Another is new services. A third is non-essential services. A fourth is
unusually complex service packages that may include standard services as
components. A fifth is services to price elastic markets.10
Rate caps can substantially mitigate the cross-subsidy concerns that
these offerings raise under COSR.
That is because prices charged are not
linked directly to costs.11 By way of example, a discount offered for a service in
one basket can affect the rates for services in another basket, if at all, only after
the next rate case. In the meantime, the utility would only lose money if it priced
its service at less than the market would bear. This encourages it to strike a
price that yields the greatest possible margin. Concern about cost shifting can
be further mitigated by placing a floor on the optional rate that equals the
incremental cost of service.
3.3.3 Marketing Flexibility Precedents
There are many precedents for marketing flexibility in rate regulation.
Flexibility provisions have to date been most extensively used in the regulation of
railroads, telecom utilities and oil pipelines, where the need for them is greatest.
I begin with these examples to build intuition before considering precedents in
energy utility industries.
10
11
Some services may qualify for light handed regulation under more than one of these criteria.
For example, a utility might wish to offer a service that is new and inessential to a competitive
market.
See, e.g., R. Brauetigam and J. Panzar, “Diversification Incentives Under “Price-Based” and
“Cost-Based” Regulation,” RAND Journal of Economics, Autumn 1989, 20:3, 373-391.
20
RAILROADS
The railroad industry provides one of the most interesting case studies of
the potential impact of marketing flexibility.
The need for marketing flexibility in
the industry stems from both demand and supply side considerations.
The
demands for railroad services have varied degrees of demand elasticity. The
chief source of elasticity is competition. Truckers, airlines, pipelines, barge lines,
and lake and ocean shipping lines, as well as other railroads, may compete for
cargos that a particular railroad might haul.
Railroads also face indirect
competition from suppliers of alternatives to the products of potential shippers.
Consider the case of steam coal. The demand for shipments of steam coal is
sensitive to the delivered cost of natural gas, an alternative fuel, at possible
generation sites. This places railroads in competition with natural gas pipelines
indirectly.
Railroads also face considerable price elasticity at the margin of use. For
example, an electric utility that uses coal must typically purchase both coal and
coal delivery services.
transport bills.
Purchase from more distant fields involves higher
The competitiveness of long distance shipments is especially
sensitive to the price of transportation.
Economically marginal customers are another source of demand elasticity
for railroads. That is, many customers have marginally profitable businesses and
rely on the railroads for the delivery of their products or important inputs. A high
cost coal mine is an example.
On the supply side, railroads must grapple with differences in the cost of
requested services. For example, it is cheaper for railroads to provide service if
customers ask for fewer pick up and drop off points and make fewer shipments.
The distance of pick up and drop off points from major rail corridors is another
important cost consideration.
Policymakers have in the last thirty years recognized the marketing
challenges facing railroads and afforded them extensive marketing flexibility.
Confronted with an industry that provided vital services but was failing to earn its
21
allowed rate of return, the federal governments of the U.S. and Canada have
passed a series of acts that reformed railroad regulation. In the United States,
the most notable legislative initiatives have been the Railroad Revitalization and
Regulatory Reform Act of 1976, the ICC Termination Act of 1995, and the
Staggers Rail Act of 1980. In Canada, the Canada Transportation Act is salient.
The Surface Transportation Board has promoted marketing flexibility through a
series of decisions.
Consider first the U.S. regulatory system.
COSR has been largely
abandoned for railroads since allowed rates are almost never based on an
allocated portion of a railroad company’s actual cost. Instead, rate restrictions,
where applied, are based on the hypothetical notion of stand-alone cost.
Regulation of the terms of U.S. railroad services is limited to markets
where railroads have demonstrated dominance. Services to numerous markets
have been officially exempted from regulation. In other markets, simple tests are
used to gauge railroad dominance.
U.S. railroads enjoy substantial marketing flexibility even where they have
market dominance.
For example, they are free to enter into confidential
contracts with shippers and most traffic moves under such contracts. Railroads
must produce formal tariffs only if a shipper requests it. The contracts commonly
have tailored pricing and service quality provisions.
challenge the terms of service railroads offer.
Captive shippers can
However, the regulations
governing maximum rates to captive shippers give railroads substantial pricing
discretion. Most notably, “differential pricing” is sanctioned in which rates can
vary with elasticities of demand in different markets.
Marketing flexibility is also extensive in Canadian rail regulation. Under
the Canada Transportation Act, regulation is limited to those services and
regions where it is necessary to serve the needs of shippers and must not
unfairly limit the ability of any carrier or mode to compete freely. As in the States,
railroads can enter into confidential contracts with customers. Additionally, grain
22
shipments are subject to an index-based revenue cap plan that gives railroads
considerable marketing discretion.
These policy measures have made possible a fascinating experiment in
how utilities facing complex and changing demands might use marketing
flexibility. One striking result has been the pervasiveness of special contracts.
According to one Canadian author, “confidential contracts have allowed railways
and shippers to craft rate and service arrangements particular to their own
needs. The concept, allowing shippers and carriers to effectively tailor their own
transportation regimes, which they agree to keep confidential, has been an
overwhelming success, garnering strong support from both shippers and
carriers.”12
In the United States about 70% of the tonnages of class I line haul
railroads by 1997 occurred under special contracts.
tonnage was exempt from economic regulation.
Another 12% of 1997
Only 18% of tonnage was
subject to rate reasonableness regulation.
There is abundant evidence that U.S. railroads use the marketing flexibility
they are allowed to engage in differential pricing. An example is a 1999 study by
the U.S General Accounting Office (GAO) of the Carload Waybill sample that the
regulator maintains.13 The study found marked differences in the margins from
services with different demand elasticities. For example, margins were greater
on shipments of wheat on routes where there were few competitive transport
options (e.g. Great Falls to Portland) than where there was competition from
other railroads and other forms of transportation (e.g. Minneapolis to New
Orleans). Low margins on motor vehicle shipments (e.g. Ontario to Chicago)
reflect trucking industry competition.
There is also evidence that railroads adjust rates to reflect change in the
markets of shippers. For example, the GAO report states that rates for some
shippers rise and fall with export demand.
12
D.W. Flicker, “Canada-United States Railway Economic Regulation Comparison: Research
Conducted for the Canada Transportation Act Review”, mimeo, November 2000.
23
The volatility in commodity markets can affect railroad rates because it
affects the demand for rail transportation. As demand changes, railroads
adjust rates to attract or retain business. For example, officials at one
Class I railroad told us that it has a wide range of pricing policies for
chemicals that allow it to react to changes in world chemicals markets.
Officials from the same railroad said that export demand can play a
particularly strong role for grain.14
The western coal industry of the United States provides a case study of
the manner in which a transportation industry with marketing flexibility and strong
marketing incentives can transform the market for shippers’ products. Changes
in U.S. environmental policy have stimulated demand for the low sulfur coal that
is abundant in many parts of the west. However, generating companies have
other means of controlling sulfur emissions as well.
These include scrubber
facilities, low sulfur coal from eastern fields, and the use of gas- and nuclearfueled generation.
Western railroads have responded to this marketing
challenge by offering attractive prices for long distance shipments. This has
encouraged the use of western coal as far afield as Michigan and Louisiana.
Railroads also use marketing efficiency to encourage shippers to use their
systems in cost effective manners.
Western coal haulers, for instance,
encourage shipment in lengthy “unit trains” devoted to particular customers.
Shippers have also been encouraged to move receipt and delivery points closer
to trunk lines.
In Canada, railroads offer rebates to grain shippers who can
assemble large numbers of cars on their own.15 Advance ordering systems are
in place that offer discounts to shippers that can make advance commitments to
ship certain volumes. An example is the Canadian Pacific’s Max Trax plan.
TELECOMMUNICATIONS
Incumbent telecommunications utilities (“telcos”) have also faced serious
marketing challenges in recent years. They have, like the railroads, faced varied
degrees of competition in the major markets they serve.
13
14
Markets in which
United States General Accounting Office, Railroad Regulation, Changes in Railroad Rates
and Service Quality since 1990, GAORCED-99-93, Washington, DC, April 1999.
Ibid p. 37
24
competition is especially strong include those for long distance service generally
and for local exchange services to larger volume customers in urban areas.
Competition is much less severe in markets for small volume customers but even
here there are challenges from cellular and PCS companies, cable television
networks, and competitive local exchange carriers.
To complicate matters
further, prices to business customers have traditionally subsidized service to
small volume customers (and also rural customers) in many regions.
As in the railroad industry, regulators have recognized the marketing
challenges facing incumbent telcos and have granted them substantial marketing
flexibility. The provision of long distance service has now been substantially
decontrolled. In local exchange service, extensive use has been made of the
marketing flexibility provisions discussed above.
In the United States, marketing flexibility was featured in the very first telco
price cap plan, that for AT&T.16
Service baskets were established and the
growth in the API for each basket was a revenue-share weighted average of the
rate elements of services in the basket. The mechanism afforded the company
automatic rate redesign and rate rebalancing flexibility. However, the degree of
flexibility was controlled. Separate baskets were established for residential and
small business users, 800 service, and other, more competitive services.
In
establishing multiple baskets, the FCC explained that
Imposing an aggregate cap on a basket of services assures regulatory
control over prices charged to the class of consumers within the basket,
and prevents cross-subsidization of services outside the basket by those
inside.17
15
16
17
See James Nolan, “Assessing the Impact of Bill C-34 on the Grain Handling and
Transportation System in the Province of Saskatchewan”, University of Saskatchewan, 2002.
See, for example, “In the Matter of Policy and Rules Concerning Rates for Dominant
Carriers”, CC Docket No. 87-313 (March 1989). Regulation of AT&T rates was abandoned
after long distance competition strengthened.
Ibid 337, p. 166.
25
Furthermore,
Our baskets…approach can and should be tailored to give AT&T less
flexibility in its pricing of residential and various less competitive services,
and greater flexibility to price efficiently in more competitive areas.18
Side conditions provided additional controls on the extent of rate rebalancing.
For example, the average residential rate was allowed to grow by only 1% more
than the price cap index each year. This general approach to marketing flexibility
has since been approved in many other telco plans.
Marketing flexibility has also been featured in Canadian telecom
regulation. In the first CRTC price cap plan, all capped services were placed in a
single basket. Growth in the API was a revenue-share weighted average of rate
elements. Numerous side conditions were imposed to control rate rebalancing
and redesign. For example, the escalation in the prices of services in two “subbaskets” (Basic Residential Local Services and Other Capped Services) were
each restricted to rise by no more than the inflation rate each year. Additionally,
escalation in individual rates for residential and single-line basic services in
smaller exchanges was limited to 10% each year. A few services, such as 9-1-1
service, were subject to a rate freeze.
No caps were imposed on the terms of optional services. In the CRTC’s
words, “Given the discretionary nature of this class of services, the Commission
is of the view that an upper pricing constraint is not warranted.”19
The
Commission also elected to remove from price caps certain services, such as
Special Facilities Tariffs, that were “redundant or impractical” to include.
However, it did not exclude services to competitive markets.
In the second CRTC price cap plan, which is now under way, the price cap
mechanism and light handed regulation were both still employed to afford telcos
marketing flexibility. However, automatic rate rebalancing was further restricted
by the establishment of more numerous service baskets. Side conditions were
18
19
Ibid 360, p. 180.
CRTC (1997) op cit, 142 p. 21.
26
employed. For example, there were 5-10% annual caps on the escalation of
individual rate elements. Rates for several services are, once again, frozen.
Optional residential services were placed in separate baskets covered by
caps. However, several services were excluded from the caps. These included,
as before, business optional local services, certain complex service bundles that
contained price-capped services, and certain Special Assembly Tariffs.
In
discussing the latter group, the Commission notes that “these services are
generally offered to a limited number of customers and the rates are often
developed having regard to factors such as long term customer commitments.”20
The Commission permitted telcos to offer certain services, such as Centrex, to
competitive markets free from price caps.
The marketing flexibility granted to Canada’s incumbent telcos brought
marked changes in their rate and service offerings. Most notably, they elected to
discount rates for services to larger volume customers in major metro areas
substantially. Rates for residential customers, meanwhile, typically escalated by
the maximum rates allowed.
OIL PIPELINES
The oil pipeline industry comprises a fairly diverse set of businesses.
Some pipelines transport crude oil from producing fields to refineries or storage
facilities.
Petroleum product pipelines transport diverse refined products
(e.g., gasoline, kerosene, home heating oils, jet fuels and diesel fuels) from
refineries to marketing terminals.21
An oil pipeline can face competition from other, substantially unregulated
modes of transportation, as well as from other pipelines.
In 2002, crude oil
pipelines carried 74.7% of the total crude oil transported while water carriers,
motor carriers, and railroads accounted for 24.9%, 0.3% and 0.1%, respectively,
of the total. In the same year, product pipelines carried 62.3% of the total while
20
21
CRTC (May 2002) 457.
Oil Pipelines of the United States: Progress and Outlook, Association of Oil Pipelines,
Washington, D.C.
27
the other three modes carried 26.3%, 3.5% and 2.3% of the total.22
It is plain
from these statistics that water carriers are the main competitors to pipelines.
However, they are able to compete with pipelines only where waterways are
available.
Pipelines serve markets with varied competitive pressures. An example
would be a product pipeline running from the Gulf Coast along the eastern
seaboard to the Northeast. It faces much less competition in Georgia and the
Carolinas than it does in major cities of the Northeast, which are served by
marine carriers, other pipelines, and local refineries.
The 1906 Hepburn Act mandated that interstate oil companies be
common carriers and that they charge just and reasonable rates. From 1906 until
1977, the rates and terms of service offered by the oil pipelines were regulated
by the Interstate Commerce Commission (ICC).
In 1977, the Department of
Energy Organization Act transferred regulatory authority of oil pipelines from the
ICC to the Federal Energy Regulatory Commission (FERC). The FERC initially
adopted a cost based approach to regulation. In 1988, following a dispute over
disclosure of confidential cost information by a pipeline, the FERC allowed a
market-based rate alternative for pipelines that can show lack of significant
market power.23
The Energy Policy Act of 1992 directed the FERC to develop a “simplified
and generally applicable methodology” to regulate pipeline rates. In Order No.
561, the FERC set out a new ratemaking methodology, which uses indexing.24
The indexing methodology caps individual pipeline rates using a price cap index
that is based on an inflation measure and a productivity offset.25 Although the
22
23
24
25
Shifts in Petroleum Transportation, 2002, Association of Oil Pipelines, Washington, D.C.
Buckeye Pipe Line Co., 44 FERC ¶ 61,066 (1988).
Order No. 561, Revisions to Oil Pipeline Regulation Pursuant to Energy Policy Act of 1992, III
FERC Stats. & Regs. ¶ 30,985 (1993).
Following a review of the indexing rate, as required by Order No. 561, the Commission
issued a December 2000 order affirming the method. Five-Year Review of Oil Pipeline Pricing
Index, 93 FERC ¶ 61,266, Docket No. RM00-11-000 (December 14, 2000). The Association
of Oil Pipelines challenged this order in court and upon further review the FERC limited the
index to track an inflation measure only. Five-Year Review of Oil Pipeline Pricing Index:
28
indexing method froze in place patterns of rates that existed upon its adoption,
the order also permits cost-of-service proceedings that allow pipelines to request
a rate above the index ceiling. In cases where pipelines can show a substantial
divergence between actual cost and revenues based on rates at the ceiling level,
they are allowed to charge cost-of-service rates. The FERC allows pipelines to
charge market-based rates if they can demonstrate that they do not exercise
significant market power in relevant markets.26
Market-based rates allow
pipelines substantial pricing flexibility in competitive markets, where rates they
charge shippers fluctuate in response to changing supply and demand
conditions. Order No. 561 also sets out provisions for pipelines to charge rates
on a negotiated basis. A version of this method that applies to existing rates,
called settlement rates, allows pipelines pricing flexibility as long as they obtain
“unanimous agreement” from all shippers using the rate; settlement rates can be
filed that exceed the index ceiling as long as pipelines and all shippers agree on
the rate. A second version of this method, which applies to new rates and is
simply called negotiated rates, requires a pipeline to secure an agreement with a
non-affiliated shipper to file this rate offering. Both methods allow pricing flexibility
for pipelines as long as they do not use market power to ‘coerce’ agreement from
shippers.
These policy measures give pipeline companies substantial flexibility to
respond to competition and to develop tailored service packages for customers.
Market-based or negotiated rates allow pipelines to meet competition and take
advantage of business opportunities.
Indexation protects customers in less
competitive markets and provides a potentially useful means of updating the
terms of special contracts.
26
Order on Remand, 102 FERC ¶ 61,195, Docket Nos. RM00-11-000 and RM00-11-001
(February 24, 2003).
The FERC issued Order No. 572, which details the requirements for application of marketbased rates. It also indicates that pipelines can not charge rates above the index ceiling until
the Commission finds they lack significant market power. Order No. 572, Market-Based
Ratemaking for Oil Pipelines, III FERC Stats. & Regs. ¶ 31,007 (1994).
29
GAS AND ELECTRIC UTILITIES
The precedents for marketing flexibility are not as extensive in the gas and
electric utility industries. This reflects, in part, less acute competitive challenges.
However, all of the major marketing flexibility provisions discussed above do
have precedent. The use of the price cap mechanism to permit automatic rate
redesign and rebalancing, for instance has been approved in a number of
jurisdictions.
Price cap plans for Boston Gas, for example, have permitted automatic
rebalancing of gas distribution rates to reduce interclass subsidies and increase
price signal efficiency. The company has been prohibited, however, from pricing
services below marginal cost. Automatic rate rebalancing was proposed by the
OEB staff for provincial power distributors in the draft Rate Handbook.
The
proposal was rejected by the Board in its final decision.
Light-handed regulation of optional rate and service offerings has
considerable precedent in North American energy regulation.
The California
Public Service Commission has allowed Southern California Gas to offer
negotiated rates and optional tariffs provided the price is not less than the longrun marginal cost. In Ontario, the OEB has approved light handed regulation for
certain services of Union Gas.
In 1996, the FERC issued a policy statement supporting expedited
approval of negotiated gas transmission services provided that customers
continued to have recourse to a rate that is based on cost of service principles. It
also concluded that “where a natural gas company can establish that it lacks
significant market power, market-based rates are a viable option for achieving
the flexibility and added efficiency required by the current marketplace.”27
In
discussions before the FERC, pipelines have cited the need for flexibility to
address a number of marketing challenges, including competition from other
pipelines, the dual-fuel capability of many large volume customers, the existence
27
74 FERC 61,076 (January 1996) p. 8.
30
of a secondary market for firm capacity, and the desire of new electric generators
for price certainty.
In the electric power industry, pricing flexibility was featured in rate plans
for two Maine electric utilities, Central Maine Power (CMP) and Bangor HydroElectric (BHE) in the mid-1990s. Both companies were bundled power service
providers at the time and had high operating costs and a number of economically
marginal and/or price sensitive large volume customers. Special contracts with
customers had previously been subject to lengthy investigations.
The Maine
Public Service Commission would approve them if it determined that the
customer would not have remained a customer at the tariffed rate and the
discount agreed to was not larger than necessary to keep the business.
A change in state law expressly permitted the Commission to authorize
pricing flexibility programs where companies could discount rates with more
limited Commission oversight. A price cap plan was approved for CMP that gave
it flexibility to discount rates for standard services, develop new customer classes
for targeted services, and to enter into special rate contracts with individual
customers without Commission approval. All offerings were subject to marginal
cost floors. The Maine Public Utilities Commission, in approving the plan, stated
that
Captive customers are protected by the rate cap and revenue deficits
borne by shareholders... Because CMP will have substantial exposure to
revenue losses due to discounting, the Company will have strong
incentive to avoid giving unnecessary discounts, and it will have a strong
incentive to find cost savings to offset any such losses. Pricing flexibility
gives CMP the opportunity to use price to compete to retain customers.
These features of the ... pricing flexibility program simulate conditions in
competitive markets and will help the Company adapt to increasing
competition in its industry.28
28
Re Central Maine Power Company, Docket 92-345 (II), January 10, 1995, p. 24.
31
Similar language appeared in the Order approving BHE’s plan.
The Maine
Commission has since approved pricing flexibility to power distributors in the
state who operate under price caps using similar reasoning.29
Both companies used the marketing flexibility granted under the first plan
to offer special discounts to customers. This created the issue of who was to
absorb the lost margin (called “revenue delta”) from discounts at the time of the
next rate case: the companies or other customers. In addressing this issue for
BHE the Maine Commission stated that
We remain convinced that pricing flexibility decisions should not be treated
like ordinary utility expenditures in which prudence investigations provide
the insurance that utility actions have been reasonable. The best means
to protect ratepayers from unreasonable price discounts is to adopt an
incentive mechanism like a price cap in which future rate increases are
unrelated to the amount of discounts granted. It is simply too difficult and
expensive to realistically review the utility’s actions and customers’
alternatives that resulted in the utility’s granting a price discount.30
After considering the risk of unnecessary price discounts that BHE faced under
its first plan the Commission decided to allocate the lost margins 85% to
ratepayers and 15% to shareholders.
In Ontario, Energy Board Staff recommended pricing flexibility for
provincial power distributors in its proposed electric distribution Rates
Handbook.31 Staff stated in the draft Handbook that
One overall price cap for a utility that imposes an average adjustment to
all prices may prove unsatisfactory from several perspectives including
limiting a distributor’s ability to fine tune its cost allocation and its
responsiveness to pricing pressures in particular sub-markets.32
Staff also cited the usefulness of pricing flexibility in achieving gradual rate
harmonization after mergers. Staff provided an example of automatic rate
rebalancing through a price cap mechanism. The pace of rebalancing would be
29
30
31
32
See, for example, Central Maine Power Company: Annual Price Change Pursuant to the
Alternative Rate Plan, Docket No. 99-155, 13 July 1999
Bangor Hydro Electric: Proposed Increase in Rates, Docket No. 97-116, March 24, 1998.
Ontario Energy Board Staff, Proposed Electric Distribution Rates Handbook, mimeo, June 30,
1999.
Ibid, p. 4-9.
32
controlled via side conditions. The Board rejected Staff’s proposal in its Rates
Handbook decision.33
3.3.4 Evaluation
Rate caps can generate utility performance incentives much stronger than
those obtained under typical cost of service regulation.
One reason is that
incentives are comprehensive so that a wide range of cost containment and
marketing initiatives are encouraged. Another is that indexing can facilitate an
extension of the period between rate cases.
To the extent that this is true,
improved unit cost performance does not reduce allowed price escalation during
the term of the plan. The benefits of improved performance can thus go straight
to the bottom line. The potential impact on productive and allocative efficiency is
substantial. The actual incentive effects of rate caps depend greatly on plan
details. For example, incentives increase with the length of the indexing period
and with the introduction of post plan sharing provisions.
Rate caps can provide a further boost to efficiency by permitting a
relaxation of operating restrictions. The case of marketing flexibility is illustrative.
To the extent that rate restrictions are external, customers of monopoly services
can be insulated from the effects of a company’s operations in markets with
price-elastic demand. This reduces concerns about cross subsidization. Lighthanded regulation of utility rates for non-core services is then possible.
A
company can also have more leeway in its purchases from affiliates and its
depreciation practices.
Rate caps can reduce regulatory cost.
Some startup costs must, of
course, be incurred to master the new regulatory system. These may include a
close monitoring of the company’s operations during the terms of the first
indexing plans.
reduced.
But the frequency of future rate cases can be substantially
Furthermore, reliance on external indexes diffuses inherently
controversial cost allocation and transfer pricing issues.
33
Decision with Reasons, RP-1999-0034, January 18, 2000.
On the other hand,
33
controversy can be considerable over alternative methods for measuring input
price and productivity growth.
The numerous inherent advantages of rate caps are offset to some degree
by disadvantages. One is regulatory risk. The novelty of rate indexing could
encourage the selection of key plan parameters arbitrarily.
Utilities may
reasonably worry that that regulators will choose plan terms that prevent the
recovery of prudently incurred cost. Customers may reasonably worry that plan
terms will deny them a fair share of plan benefits. Concerns about arbitrary
selection of key plan parameters reduce the willingness of parties to try the rateindexing option and can weaken the incentive benefits of price cap plans
substantially. A rate freeze is a sensible alternative to indexing in jurisdictions
where this is a concern but is not suitable in all times and places, as has been
noted.
Rate caps also involve business risk: the possibility that price restrictions
will not track trends in external business conditions that affect a company’s unit
cost. Relevant business conditions include weather, the business cycle, input
prices, financial markets, and government policy. Windfall gains and losses may
occur if rate caps don’t reflect changes in these conditions.
34
4. REVENUE CAPS
4.1 Comprehensive Revenue Caps
4.1.1 Description
Under a comprehensive revenue cap the revenue requirement of the
company and not its rates is the focus of regulation. The growth of the revenue
requirement is usually limited to the growth in a revenue cap index (RCI), as in
the following formula:
growth Revenue Requirement < growth RCI
[2]
Like PCIs, RCIs often feature measures of price inflation.
RCIs may
include, additionally, a measure of output growth.34
The addition of a balancing account mechanism can ensure that actual
revenues are similar or equal to the revenue requirement.
The balancing
account contains the value of any mismatch between actual revenue and the
revenue requirement until rates can be adjusted to eliminate it.
These
arrangements are sometimes called revenue-decoupling mechanisms since they
sever the link between revenue and efforts to market regulated services.35
Revenue cap mechanisms typically do not specify how revenue limits are
translated into rate limits. The regulation of rate and service offerings can, in
fact, continue using traditional methods. The utility can, in principle, be afforded
some flexibility in the provision of rate and service options. However, incentives
for efficient marketing are weaker than under a rate cap mechanism, as I discuss
further below.
34
35
This is discussed further in Section 5.
Revenue decoupling mechanisms have also been used in the absence of indexing.
Prominent examples include the electric revenue adjustment mechanisms that have been
used in California and Maine.
35
4.1.2 Precedents
UNITED STATES
A comprehensive revenue per customer indexing plan was approved for
the gas delivery services of Southern California Gas (CA) in 1996. The company
had proposed price caps but a revenue cap was deemed more consistent with its
previous regulatory commitments. The CPUC has since approved revenue cap
plans for the power and gas distribution services of Pacific Gas & Electric and
San Diego Gas and Electric, and the power distribution services of Southern
California Edison. A comprehensive revenue cap plan was approved in 1998 for
the power distribution services of PacifiCorp in Oregon. A revenue per customer
indexing plan was approved for the gas distribution services of Baltimore Gas &
Electric in 1998.
CANADA
The
National
Energy
Board
(NEB)
of
Canada
has
approved
comprehensive revenue caps for two oil pipelines, Enbridge Pipelines (formerly
Interprovincial Pipe Line) and TransMountain Pipe Line.
companies resulted from settlement agreements.
Plans for both
There is no evidence that
industry unit cost trends were explicitly considered in the development of these
plans.
BRITAIN
The power transmission services of National Grid have been subject to
revenue caps since 1993. All regulated transmission services were subject to
revenue caps under the first plan.
Dispatching and other system operation
services have since been exempted from revenue caps.
AUSTRALIA
Revenue caps are used by the ACCC to regulate power transmission
services of Energy Australia, Powerlink Queensland, Powernet Victoria, and
Trans Grid in Australia. The inflation factors in all of these plans are consumer
price indexes. Revenue caps are also used to regulate power distributors in New
South Wales.
36
4.1.3 Evaluation
Comprehensive revenue caps can create strong incentives for cost
containment by permitting operation for an extended period with an externalized
revenue requirement. There are incentives for a wide range of cost containment
initiatives.
The external basis for the cap also encourages some forms of
operating flexibility. For example, extended utility operation under a revenue cap
could permit a regulator to relax restrictions on purchases from affiliates.
One important difference between the consequences of rate and revenue
indexing lies in the marketing of utility services.
Incentives for improved
marketing are general weaker than under rate caps. Marketing incentives may,
in fact, be weaker than under COSR. For example, reducing rates for services in
price elastic applications may, by raising total revenue, lower rates promptly.
Utilities may, as a consequence, be less aggressive at promoting system uses,
including efforts to avoid uneconomic bypass.
They do however, have an
incentive to raise rates for services that are especially costly to provide.
Revenue caps can raise more concerns than rate caps about the quality of
core services.
As with rate caps, quality may suffer since there are strong
incentives to cut costs. While the pressures to minimize costs are similar under
rate and revenue caps, under a revenue cap revenues that are lost if poor
service leads to fewer sales can be recovered through price increases on
remaining customers using the balancing account. Since this is not possible
under rate caps, the incentives to maintain service quality are weaker in the
absence of counterbalancing incentive provisions. This concern will be greater to
the extent that customers care about quality and lack cost-competitive
alternatives.
Revenue cap plans reduce windfall gains and losses from demand
fluctuations. This is an important consideration for utilities that face unusually
volatile demand due, for instance, to sensitivity to weather, prices of competing
products, or prices in the end product markets of business users.
37
Stabilization of revenue can lower a utility’s capital cost but in the process
destabilizes rates. For example, a recession in the service territory can place
upward pressure on rates at a time when rate increases are especially
unwelcome.
Another important attribute of revenue caps is their ability to strengthen
the incentives to promote energy conservation. Under rate caps, the promotion
of conservation can reduce a utility’s operating margins. Under revenue caps,
rates rise automatically to offset this effect.
Conservation is an important goal in some jurisdictions. However, there
are other methods for promoting energy conservation. One is a commitment by
the utility before the start of a plan to achieve certain conservation objectives.
Consideration should also be given to the issue of regulatory cost.
Revenue caps can permit economies in the cost of regulation relative to COSR.
However, regulatory cost is likely to be somewhat greater than under rate caps.
One reason is the need for periodic filings to implement the balancing account
mechanism.
There may, additionally, be a continued need to consider the
allocation of revenue requirements between customer groups, service offerings,
and rate design. Note that the addition to the indexing formula of an output
growth factor creates another potential plan design controversy.
4.2 Non-Comprehensive Revenue Caps
4.2.1 Basics
Under non-comprehensive revenue caps there are caps on only a portion
of the company’s revenue requirement.
An example might be a cap on the
revenue requirement (allowed cost) for O&M expenses. Partial revenue caps
are, like comprehensive caps, usually fashioned using indexes. In the event of
indexing, an adjustment for output quantity growth is once again needed. As with
comprehensive revenue cap plans, partial indexing plans typically do not address
38
rate and service offerings. Utilities therefore typically require authority outside of
partial rates and revenue caps to alter these offerings.
4.2.2 Precedents
UNITED STATES
An important early example of non-comprehensive revenue caps is the
first PBR plan for San Diego Gas and Electric. This plan, which applied to both
gas and electric services, was approved in 1994.36 It featured separate indexbased adjustments for revenue requirements corresponding to allowed O&M
expenses and capital spending. Separate O&M indexing mechanisms were
specified for gas and electric operations. The mechanisms included inflation
factors, X-factors, and adjustments for output growth.37
CANADA
Non-comprehensive revenue caps have been more widely used in
Canada than in the U.S.
BC Gas began operating under caps for certain
categories of base rate revenue in 1994. The caps pertained to O&M expenses
and small capital expenditures.
BC Gas also operates under a revenue
decoupling mechanism called the Revenue Stabilization Adjustment Mechanism.
It applies only to revenues from residential and commercial sales.
The NEB approved a non-comprehensive revenue cap plan for gas
transmission services of Westcoast Energy in 1996. Indexing limited growth in
the revenue requirement components covering O&M expenses and small capital
additions. The formula for growth in both revenue cap indexes was forecasted
inflation in a CPI. There was no explicit X or output factors in the formula.
The Alberta commission has approved non-comprehensive revenue caps
for NOVA Gas Transmission.
The caps apply to O&M expenses and small
capital additions. A plan was approved by the OEB for the gas delivery O&M
expenses of Toronto-based Consumers Gas in 1998.
36
37
It has been claimed that the term “performance based ratemaking” was coined by San Diego
personnel during this plan’s development.
This plan was succeeded by the rate cap plan that is mentioned above.
39
4.2.3 Evaluation
Non-comprehensive revenue caps can make revenues in the targeted
areas less sensitive to the operations of the subject utility. This can substantially
strengthen incentives to contain the associated costs.
It can also permit
increased operating flexibility in the targeted areas. Suppose, by way of example
that a utility wishes to purchase many of its O&M services from unregulated
affiliates. A cap on allowed O&M expenses can then permit more light-handed
review of service transfers while permitting continued COSR for capital costs.
Non-comprehensive revenue caps can make sense in situations where
comprehensive caps do not.
One example is a situation where a company
expects to make a sequence of large capital additions in the next few years. The
company has a legitimate concern about the recovery of these costs, and may
wish for this reason to see them approved and included in the rate base. On the
other hand, a sequence of traditional rate cases will weaken incentives for O&M
cost management.
One potential problem with partial revenue caps is the unevenness of
performance incentives that result. There may be less incentive to control cost in
non-targeted areas. The company may, in the extreme, be given an incentive to
improve performance in the targeted areas at the expense of performance in
other areas. If a utility were subject only to a cap on O&M revenue, for instance,
excessive capital spending could be undertaken to reduce O&M expenses.
Overall, the company’s performance might not improve.
This problem is
mitigated to the extent that the partial caps cover most areas of controllable cost.
For example, plans covering both O&M expenditures and capital expenditures
can be defended on the grounds that they cover all “controllable” costs.
Partial revenue caps share with comprehensive caps several other
attributes. One is relatively weak incentives for better marketing. Another is
stronger incentives to promote energy conservation.
40
5. INDEX DESIGN ISSUES
Rate and revenue cap indexes can have an important impact on the
welfare of utilities and their customers. The indexes are commonly determined
using formulas. The design of such formulas is therefore a salient issue in many
PBR proceedings. In this section I discuss key capping index design issues.
5.1 Overview
5.1.1 Index Formulas
Price cap index formulas vary from plan to plan but have the following
general structure. The PCI growth rate (growthPCI ) is the difference between an
inflation factor (P) and an X-factor (X), plus or minus a Z-factor (Z).38
The
standard formula may be stated succinctly as
growth PCI = P − X ± Z
[3]
Compared with price cap indexes, a growth rate formula for a revenue cap
index requires an additional adjustment to reflect the effect of output growth on
cost. Some approved RCI formulas have an explicit term for such an adjustment
which may be called an output factor is here denoted by Y.
growth RCI = P − X + Y ± Z
[4]
The X and Y terms as here described are sometimes captured in a
consolidated X so that the growth rate formula resembles that for a price cap. If
X happens to be similar to the expected growth of output (i.e.,Y = X ), the formula
can be further simplified to
growth RCI = P ± Z
38
[5]
The term Z-factor appears to have developed in the FCC proceeding to develop a price cap
plan for AT&T. It was so called because the PCI for AT&T also included an X-factor as here
described and a “Y” factor to affect a specific category of price cap adjustments.
41
If, alternatively, the inflation rate is deemed to be similar to the
unconsolidated X factor, the rate cap growth formula can be reduced to
growth RCI = Y ± Z
[6]
Some plans restrict growth in revenue per customer. This can be shown
to be mathematically equivalent to revenue requirement indexing where the
growth rate in the number of customers is the output factor.
5.1.2 Inflation Measures
The inflation factor, P, provides an automatic adjustment to the PCI for
price inflation. It is sometimes fixed in advance but is more commonly the recent
growth rate in an external price inflation measure. Three basic kinds of inflation
measures have been used in approved rate-cap plans. These may be termed
macroeconomic, industry-specific input price, and output price measures. We
discuss each in turn.
MACROECONOMIC MEASURES
Macroeconomic inflation measures are summary measures of growth in
the prices of a wide range of the economy’s goods and services. Those used in
PBR plans are typically output price indexes computed by government agencies.
Examples include price indexes for gross domestic product (GDPPIs) and
consumer price indexes (CPIs).
Macroeconomic measures are almost universally used in telecom utilities’
rate-cap plans. For example, the GDPPI has been employed in both price cap
plans of the CRTC.
Macroeconomic inflation measures have also been
employed in several PBR plans for energy utilities. The price cap index for Union
Gas, for instance, used as an inflation measure the Ontario CPI. Consumer price
indexes such as Britain’s retail price index (RPI) are used in almost all also
overseas indexing plans overseas.
One advantage of macroeconomic inflation measures is their simplicity.
Another is their credibility, since they are typically computed with some care by
government agencies.
Still another is their familiarity to stakeholders in the
42
regulatory process. The main concern with macroeconomic inflation measures is
their ability to track growth in the prices of utility inputs.
INDUSTRY SPECIFIC MEASURES
Industry-specific input price indexes are expressly designed to track
inflation in the prices of the relevant utility inputs. Such measures summarize the
growth in subindexes that are chosen to track trends in the prices of major input
categories.
The index formula customarily assigns weights to the subindex
growth rates which reflect the shares of the input categories in utility cost. This is
the approach to index weighting which best captures the impact of growth in
various input prices on cost.
An industry-specific input price index was first used in the PBR plan for
U.S. railroads.
It was a weighted average of the growth rates in indexes of the
prices of railroad inputs, including labor, fuel, materials, equipment rentals,
depreciation, interest, and a miscellaneous input category.
Each input was
assigned a weight that reflected its share of the cost of railroad operations. An
industry-specific input price index was first approved in the energy industry for
the bundled power services of PacifiCorp (CA). The staff of the California Public
Service Commission (CPUC) played an instrumental role in the index design.
Industry-specific inflation measures have since been approved for the gas
delivery services of Southern California Gas (CA), the gas and electric power
delivery services of San Diego Gas and Electric (CA), and the power distribution
services of Ontario utilities.
By design, an industry-specific input price index can track industry input
price fluctuations better than an economy-wide measure. Such an index can
thus do a better job of reducing windfall gains and losses that might result from
the failure of a macroeconomic index to track input price inflation. Business risk
can be lessened thereby. These advantages are important to the extent that the
input price growth of a utility industry differs from that of the economy.
For
example, energy transmission and distribution are unusually capital intensive
businesses and are therefore unusually sensitive to change in the cost of funds.
43
This has a pattern of fluctuation that can differ from that of other utility inputs for
extended periods.
One disadvantage of the industry-specific approach is the complexity of
the design challenge. No official source computes input price indexes for energy
utilities.
However, the construction of accurate indexes is aided by well-
established theory and publicly-available data.
An interesting issue in considering industry-specific inflation measures is
their effect on regulatory risk. Industry-specific measures can in principle reduce
operating risk and help sidestep controversy over possible adjustments needed
to a PCI with a macroeconomic inflation measure to help it better track industry
input price trends. On the other hand, approved industry-specific measures may
not do the best possible job of tracking industry input price inflation. A good
example is the measure approved in Ontario for power distribution. This counted
only half of the calculated growth in the capital price in the name of rate
stabilization. Since the capital price had the largest weight in the growth rate
formula, the effected on allowed price escalation was substantial. This could
matter greatly in the long run given the capital intensiveness of the distribution
business.
OUTPUT PRICE MEASURES
Industry specific output price indexes are indexes of the prices charged by
other service providers. For example, plans for two Midwestern U.S. electric
utilities linked their industrial rates to those of neighboring utilities. The PBR plan
for power distribution services of National Grid USA in Massachusetts will adjust
its rates for five years using an index of the distribution rates of northeast utilities.
Indexes of this kind reflect the unit cost trend of the industry.
They
therefore reflect productivity as well as input price trends. An advantage of their
use is avoidance of controversy over how these trends should be measured. In
North America, it has until recently been difficult to regulate most transmission
and distribution services using peer price indexes due to the lack of unbundled
price data on the services.
44
5.1.3 X-Factors
The X-factor is an external parameter in the PCI formula that typically
causes the PCI to grow more slowly than the inflation measure, to the benefit of
customers. Thus, prices for regulated services are likely to decline in real terms.
X is sometimes called a “productivity factor” or a “productivity offset” to the
inflation measure since considerations of productivity growth are sometimes
involved explicitly in choosing its value.
Various methods have been used to ensure the external character of X.
Most commonly, its value in each plan year is set in advance and is constant
throughout the plan. However, in several approved plans the X-factors are set in
advance but scheduled to vary from year to year. For example, X-factors in
several cases have been scheduled to rise gradually over the term of the plan. X
may also be recomputed periodically to reflect new information as long as the
calculation method is established in advance. The best known precedent for this
approach is the X-factor in the price cap index for U.S. railroads.39 This was an
annually updated rolling average of the recent productivity growth of the railroad
industry.
5.1.4 Z-Factors
The Z-factor term of a price cap index adjusts the allowed rate of price
escalation for external developments that are not reflected in the inflation and
X-factors. It is apt to differ from period to period. One of the primary rationales
for Z-factor adjustments is the need to adjust price limits for the effect of changes
in tax rates and other government policies on the company’s unit cost. Absent
such adjustments, policymakers can adopt new policies that increase the
company’s unit cost confident in the knowledge that its earnings, rather than its
rates, will be affected. Another rationale for Z-factors is to adjust for the effect of
miscellaneous other external developments on industry unit costs that are not
captured by the inflation and X-factors.
39
This is discussed in more detail below.
Z-factors can potentially reduce
45
operating risk and discourage opportunistic behavior without weakening
performance incentives since they are triggered only by external developments.
A disadvantage is that they can significantly raise regulatory cost.
Most index-based price cap plans have explicit rules for Z-factor
adjustments. Those approved by the OEB for provincial power distributors and
recorded in the distribution Rates Handbook are illustrative.
•
Causation – the expense must be clearly outside of the base upon which
rates were derived.
•
Materiality – the cost must have a significant influence on the operation of
the utility, otherwise they should be expensed in the normal course and
addressed through organizational productivity improvements.
•
Inability of Management to Control – to qualify for Z-factor treatment, the
cost must be attributable to some event outside of management’s ability to
control. Examples include a tax change or requirements of the IMO that
result in expenditures by the distribution utility. On the other hand, an ice
storm that causes extensive damage in a system with sub-par
maintenance would not qualify for Z-factor treatment.
•
Prudence – the expense must have been prudently incurred. This means
that the option selected must represent the most cost-effective option (not
necessarily least initial cost) for ratepayers. For example, some utilities
will need to upgrade their billing systems to deal with market opening.
The prudence standard requires that the utility justify purchasing a new
system versus outsourcing the function to a vendor, association, or
utility.40
While there is no general language about the relevance of government
policy changes, the explicit mention of tax changes and Independent Market
Operator requirements in the Rate Handbook is noteworthy. The Board shed
40
OEB, Draft Distribution Rates Handbook, 1999.
46
further light on events it deems relevant for Z factoring in its Union Gas decision.
It states that
The Board agrees with the intervenors that the use of Z-factors limited to
changes in legislative and regulatory requirements and generally accepted
accounting principles specific to the natural gas business is appropriate.41
5.2 Index Design Methods
Two general approaches to the design of rate and revenue cap indexes
have now been established: the North American approach and the British.
These are so-named because of their region of origin.
5.2.1 The North American Approach
Although index-based PBR is associated in the minds of many with Great
Britain, North America actually has a longer history with this regulatory system.
E. Fred Sudit of Rutgers University outlined the approach to PCI design that has
become common in North America in a 1979 paper.42 William Baumol, then at
Princeton University, elaborated on the idea in a 1982 paper.43 These early
treatises influenced the American approach to PCI design, but credit must also
go to other individuals who were involved in the early regulatory proceedings and
supporting legislation.
LOGIC OF PRICE CAP INDEXES
The North American approach to index design is founded on the logic of
economic indexes. The analysis begins with consideration of the growth in the
prices charged by an industry that earns, in the long run, a competitive rate of
return. In such an industry, the long-run trend in revenue equals the long-run
trend in cost.
41
42
43
OEB, Decision with Reasons, RP-199-0017, July 2001.
E. Fred Sudit, “Automatic Rate Adjustments Based on Total Factor Productivity Performance
in Public Utility Regulation,” in Problems in Public Utility Economics and Regulation ed. M.
Crew, Lexington Books, 1979.
William J. Baumol, “Productivity Incentive Clauses and Rate Adjustment for Inflation,” Public
Utilities Fortnightly, July 22, 1982, pp. 11-18.
47
trend Revenue = trend Cost
[7]
The assumption of a competitive rate of return is applicable to utility
industries and even to individual utilities. It is also applicable to unregulated,
competitively structured markets.
Consider, now, that the trend in the revenue of any firm or industry is the
sum of the trends in appropriately specified output price and quantity indexes.
trend Revenue = trend Output Quantities + trend Output Prices
[8]
Relations [7] and [8] together imply that the trend in an index of the prices
charged by an industry earning a competitive rate of return equals the trend in its
unit cost index.
trend Output Prices = trend Cost - trend Output Quantities = trend Unit Cost [9]
The long run character of this important result merits emphasis.
Fluctuations in input prices, demand and other external business conditions will
cause earnings to fluctuate absent adjustments in production capacity.
Fluctuations in certain expenditures that are made periodically can also have this
effect. An example would be a major program of replacement investment for a
distribution system with extensive asset depreciation.
Since capacity
adjustments are costly, they will typically not be made rapidly enough to prevent
short-term fluctuations in returns around the competitive norm. The long run is a
period long enough for the industry to adjust capacity to more secular trends in
market conditions.
The result in [9] provides a conceptual framework for the design of a rate
adjustment index that we call the industry unit cost paradigm. For example,
growth in a utility’s rates can be measured by an actual price index. A PCI can
limit the growth in this index. A stretch factor established in advance of plan
operation can be added to the formula which slows PCI growth to the benefit of
48
customers.44 A PCI is then calibrated to track the industry unit cost trend to the
extent that
trend PCI = trend Unit Cost Industry − Stretch Factor
[10]
A properly designed PCI provides automatic adjustments for trends in external
business conditions that affect the unit cost of utility operation. It can therefore
reduce utility operating risk without weakening performance incentives.
This
constitutes a remarkable advance in the technology for utility regulation.
The design of PCIs that track the industry unit cost trend is aided by an
additional result of index logic. It can be shown that the trend in an industry’s
total cost is the sum
of the trends in appropriately specified industry input price and quantity indexes.
trend Cost = trend Input Prices + trend Input Quantities
[11]
It follows that the trend in an industry’s unit cost is the difference between
the trends in industry input price and TFP indexes.45
trend Unit Cost = trend Input Prices − trend TFP
[12]
Furthermore, a PCI can be calibrated to track the industry unit cost trend if
it satisfies the following formula:
trend PCI = trend Input Prices Industry − (trend TFP Industry + Stretch Factor )
[13]
An important issue in the development of a PCI is whether it should be
designed to track short run or long run unit cost growth. An index designed to
track short run growth will also track the long run growth trend if it is used over
many years. The alternative is to design the index to track only long run trends.
44
45
Mention here of the stretch factor option is not meant to imply that a positive stretch factor is
warranted in all cases.
Here is the full logic behind this result:
trend Unit Cost = trend Cost - trend Output Quantities
= (trend Input Prices + trend Input Quantities ) − trend Output Quantities
= trend Input Prices
− (trend Output Quantities - trend Input Quantities )
= trend Input Prices − trend TFP
49
Different approaches can, in principle, be taken for the input price and
productivity components of the index.
One issue to consider when making the choice is the manner in which
short-run input price and productivity fluctuations affect the prices charged by
unregulated industries.
Inflation in the prices charged by such industries
sometimes accelerates (decelerates) rather promptly when input prices
accelerate (decelerate).
Airlines and trucking companies, for instance,
sometimes hike prices in periods of rapid fuel price growth.
An analogous result does not obtain for TFP. For example, TFP typically
falls (rises) in the short run in response to a slackening (strengthening) of
demand. These same developments typically have the reverse effect on prices
in unregulated markets.
A second consideration is the effect on risk. A price cap index that tracks
short-term fluctuations in industry unit cost increases rate volatility but reduces
utility operating risk. This can permit an extension of the period between rate
reviews that strengthens performance incentives.
Consider, next, the criterion of implementation cost. This depends in large
measure on data availability. Data on price trends are available more quickly
than the cost and quantity data that are needed, additionally, to measure TFP
trends. Final data needed to compute the TFP growth of U.S. power distributors
in 2004, for instance, was not available until the fall of 2005. The longer lag in
the availability of cost and quantity data is due chiefly to the fact that these data
typically come from annual reports whereas price indices are often calculated
and reported on a monthly or quarterly basis.
It is also germane that the
calculation of TFP indexes can be quite a bit more complicated than the
calculation of price indexes.
Implementation cost also depends on the feasibility of calculating current
long run trends accurately.
Methods have been developed to measure the
recent long run trend in the TFP of the industry. For example, a sample period
suitable for calculating the recent long run trend can be chosen using research
50
on the drivers of TFP index volatility. The recent long run trend in an industry’s
TFP is, moreover, often if not always a good proxy for the prospective trend over
the next several years. 46
The use of historical data on industry input price trends to calculate the
prospective future trend is more problematic. Industry input price indexes are
often volatile. The calculation of an average annual growth rate thus depends
greatly on the choice of the sample period. It can be difficult to reach consensus
on what sample period would yield a long term input price trend. One reason is
that research on the short run drivers of fluctuations in utility input prices is not
well advanced. Absent a scientific basis for sample period selection, the choice
of a sample period can engender controversy and raise the risk of PBR for
utilities. Higher regulatory risk can raise the cost of funds and reduce thereby the
net benefits of PBR.
Historical trends in input prices are, furthermore, sometimes poor
predictors of the trends that will prevail in the near future. Suppose, by way of
example, that there has been rapid input price inflation in the last ten years but
that the expectation is for more normal inflation in the next five years. In this
situation, regulators would presumably be loath to fix PCI growth at a rate that
reflects the 10-year historical trend.
Examination of input prices in the power distribution industry is useful for
illustrating these concepts.47 Since power distribution is capital intensive, the
summary input price index is quite sensitive to fluctuations in the price of capital.
These result from fluctuations in plant construction costs and the rate of return on
capital. The rate of return on capital depends on the state of the economy and
on expectations regarding future price inflation. 48
A sensible weighing of these considerations leads us to conclude that
different treatments of input price and productivity growth are in most cases
46
47
48
Reliance on the long run trend can be problematic, however, when applied to utilities that
contemplate major capital additions.
This analysis also applies to power transmission, as we discuss further below.
The rate of return on capital also reflects return on equity.
51
warranted when a PCI is calibrated to track the industry unit cost trend. The
price inflation index should track short term input price fluctuations. The X factor,
meanwhile, should generally reflect the long run historical trend of TFP.
This general approach to PCI design has important advantages. The price
inflation index measure exploits the greater availability of inflation data. Making
the PCI responsive to short term input price growth reduces utility operating risk
without weakening performance incentives.
Having X reflect the long run
industry TFP trend, meanwhile, sidesteps the need for more timely cost data and
avoids the chore of annual TFP calculations.
Given that the price inflation index should track recent input price growth,
other important issues of its design must still be addressed. One is whether it
should be expressly designed to track industry input price inflation as per relation
[13]. There are several precedents for the use of an industry-specific inflation
measure in rate adjustment indexes.
The majority of rate indexing plans
approved worldwide, however, feature macroeconomic inflation measures When
a macroeconomic inflation measure is used, the PCI must be calibrated in a
special way if it is to track the industry unit cost trend. Suppose, for example,
that the inflation measure is the GDP-PI. This was noted above to be an index of
output price inflation. Due to the broadly competitive structure of our economy,
the long run trend in the GDP-PI is the difference between the trends in input
price and TFP indexes for the economy.
trend GDPPI = trend Input Prices Economy - trend TFP Economy
[14]
Equations [12] and [14] together imply that
trend Unit Cost Industry = trend GDPPI
Industry
⎡
⎤
- trend TFP Economy
- ⎢ trend TFP
Economy
Industry ⎥
- trend Input Prices
⎣+ trend Input Prices
⎦
(
(
)
)
[15]
When the GDP-PI is used as the inflation measure, it follows that the PCI already
tracks the input price and TFP trends of the economy. X factor calibration is
52
warranted only to the extent that there are differences in the input price and TFP
trends of the utility industry and the economy.
This analysis suggests that when the GDP-PI is employed as a price
inflation index the PCI can be calibrated to track the industry unit cost trend when
the X factor has two calibration terms: a productivity differential and an input
price differential. The productivity differential is the difference between the TFP
trends of the industry and the economy. X will be larger, slowing PCI growth, to
the extent that the industry TFP trend exceeds the economy-wide TFP trend that
is embodied in the GDP-PI. The input price differential is the difference between
the input price trends of the economy and the industry. X will be larger (smaller)
to the extent that the input price trend of the economy is more (less) rapid than
that of the industry.
The input price trends of a utility industry and the economy can differ for a
number of reasons. One possibility is that prices in the utility industry grow at
different rates than prices in the economy as a whole. For example, labor prices
may grow more rapidly to the extent that utility workers have health care benefits
that are better than the norm. Another possibility is that the prices of certain
inputs grow at a different rate in some regions than they do on average
throughout the economy. It is also possible that the industry has a different mix
of inputs than the economy. Power distribution technology is, for example, noted
above to be more capital intensive than the typical production process in our
economy. It is therefore more sensitive to fluctuations in the price of capital.
The difficulties, discussed above, in establishing a long-term input price
trend also complicate identification of an appropriate input price differential. For
example, the difference between the average annual growth rates of input price
indexes for of the industry and the economy is sensitive to the choice of the
sample period. It is less straightforward to establish the relevant sample period
for a comparison of long-term industry and economy input price trends than it is
for an analogous TFP trend comparison. Even if we could establish a differential
between the long term trends it could differ considerably from the trend expected
53
over the prospective plan period. This situation invites gaming over the sample
period used to calculate the input price differential.
Controversy is possible,
additionally, over the method used to calculate the price of capital.
LOGIC OF REVENUE CAP INDEXES
The extension of index logic to the case of revenue caps is
straightforward. A revenue cap index that is based on index logic would be
calibrated to track the cost trend of the industry rather than the unit cost trend.
The cost trend of an industry is the sum of the unit cost trend and the output
quantity trend. Recalling the results in [12] it follows that the cost trend is the
difference between the input price and productivity trends plus the output quantity
trend. A revenue cap index ("RCI") is then calibrated to track the industry cost
trend if
trend RCI = trend Input Prices - trend TFP
+ trend Output Quantities + OUtput Quantities
[16]
The RCI growth formula thus differs chiefly from the PCI growth formula chiefly in
considering a provision for output growth.
PRECEDENT
The earliest use of index logic in regulation design emerged from hearings
before U.S. federal regulatory commissions. As early as 1980, the Interstate
Commerce Commission (ICC) proposed to determine allowable increases in rail
freight rates using the average increase in rail carrier costs.49 The Staggers Rail
Act of 1980 was noted above to require index-based regulation for larger
railroads. The law established a Zone of Rate freedom for certain rail services.
Under Section 203 of the Act, the boundary of this zone was to be adjusted each
quarter by an “Index of Railroad Cost compiled or verified by the commission with
appropriate adjustments to reflect the changing composition of railroad cost,
including the quality and mix of material and labor”. The growth rate of this index
came to be called the Rail Cost Adjustment Factor (RCAF).
54
There was vigorous and protracted debate before the ICC regarding the
appropriate form of this index. The most fundamental issue was whether the
index should reflect the trend in the TFP of the industry as well as the input price
trend. An index reflecting both would track the unit cost of the industry, as noted
above.
In 1989, the ICC concluded that the index should reflect the TFP trend of
the railroad industry as well as its input price trend.50 The X-factor it adopted is a
moving average of the growth rate in an index of railroad industry TFP, as noted
above. The index measured the productivity of the very companies that were
subject to the PBR plan.
The U.S. Federal Communications Commission has issued landmark
decisions on PCI design that are broadly consistent with index logic.
In
approving the price cap plan for AT&T in 198951, inflation measures and industry
TFP trends were discussed extensively.52 The X-factor reflected the industry
productivity trend and an inflation measure adjustment.
In approving rate indexing for the interstate services of local exchange
carriers the need to calibrate the PCI to the industry unit cost trend was explicitly
recognized. For example, in a 1995 order dealing with PCI, the FCC states that
“the indexes are adjusted each year in accordance with a formula that accounts
for industry-wide changes in unit costs”.53
Since the approval of the first plans at the federal level, rate-indexing
plans have been adopted by a number of other regulatory commissions. The
industry unit cost standard is frequently observed in PCI design. Commissions
sometimes recognize the standard explicitly.
49
50
51
52
53
Thus the Massachusetts
ICC, Advanced Notice of Proposed Rulemaking, “Railroad Cost Recovery Procedures,”
Ex Parte No. 290 (Sub-No. 2), April 28, 1980, 49 CFR 1135.
ICC, “Decision, Railroad Cost Recovery Procedures-Productivity Adjustment,” Ex Parte
No. 290 (Sub-No. 4), March 22, 1989.
“In the Matter of Policy and Rules concerning Rates for dominant Carriers.” 4FCC
Rcd 27763; CC Docket No. 87-313 (March 15, 1989).
The affected rates of AT&T were subsequently decontrolled.
Federal communications Commission, First Report and Order in the Matter of Price Cap
Performance for Local Exchange Carriers, cc Docket 94-1, April 7, 1995.
55
Department of Public Utilities, in approving a rate-cap plan for NYNEX, noted in
1995 that,
Price cap regulation…replaces company specific test year cost based
control of a firm’s rates with an index representing the expected changes
in the cost of the average firm in the industry.54
The California Public Utilities Commission noted in the same year in
approving the rate-cap plan for Southern California Edison that
The price and productivity values should come from national or industry
measures and not from the utility itself. The independence of the update
rule from the utility’s own costs allows PBR regulation to resemble the
unregulated market where the firm faces market prices which develop
independently of its own cost and productivity. The productivity measure
should come from a forecast of industry-specific productivity.”55
In Canada, the CRTC has also subscribed to the industry unit cost
standard. In its order approving the PBR plan for the Stentor companies, the
CRTC states that, “the price cap formula is composed of three basic components
which, in total, reflect changes in the industry’s long run unit costs.56
The price cap index approved by the OEB for Ontario power distributors is
constructed from industry specific input price and TFP trends. It is thus expressly
designed to track the industry unit cost trend. Note, however, that the OEB did
not elect to base the X-factor solely on the long-term TFP trend.
As for RCI logic precedents, the RCI approved by the OEB for the O&M
expenses of Enbridge Gas Distribution was based on index logic. So too was the
revenue per customer index approved by the CPUC for Southern California Gas.
TOTAL FACTOR PRODUCTIVITY
Since TFP play an important role in North American style rate indexing it
may prove useful to explain their workings in more detail.
54
55
Petition of New England Telephone and Telegraph Company dba/NYNEK for an Alternative
Regulatory Plan for the company’s Massachusetts Intrastate Telecommunications Services.
DPU 94-50. May 12, 1995.
Application of Southern California Edison to adopt a Performance Based Rate Making
Mechanism Effective January 1, 1995, Alternate Order of Commissioners Fessler and Duque,
July 21, 1996.
56
The TFP index of an industry is the ratio of output and input quantity
indexes.
TFP =
Output Quantities
Input Quantities
[17]
The output quantity index of an industry summarizes trends in the amount
of work that it performs. If output is multidimensional, the growth in each output
quantity dimension considered is measured by a subindex. The growth in the
output quantity index is typically a weighted average of the growth in the quantity
subindexes.
The input quantity index of an industry summarizes trends in the amounts
of production inputs that it uses. Growth in the usage of each input category
considered is measured by a subindex. For example, growth in the amount of
labor services employed can be measured by a labor quantity subindex. The
growth in the summary input quantity index is typically a weighted average of the
growth in the quantity subindexes.
The TFP index of an industry captures the wide range of developments
that can cause its unit cost to grow at a different rate than its input prices. These
developments include technological progress and the realization of scale
economies. TFP is volatile but typically trends upward so that an industry’s unit
cost grows more slowly than its input prices over time.
Economic research has shown that the sources of TFP growth are diverse. One
important source is technical change. New technologies permit an industry to
produce given output quantities with fewer inputs.
Economies of scale are a second source of TFP growth.
These
economies are available in the longer run when cost characteristically grows less
rapidly than output. In that event, output growth can slow unit cost growth and
raise TFP. A company’s potential for scale economy realization depends on its
current operating scale and on the pace of its output growth. Incremental scale
56
Ibid paragraph 29.
57
economies will be greater the more rapid is output growth and the smaller is the
initial operating scale.
A third important source of TFP growth is X inefficiency.
This is the
degree to which individual companies operate at the maximum efficiency that
technology allows.
TFP will grow (decline) to the extent that X inefficiency
diminishes (increases). The potential of a company for TFP growth from this
source is greater the greater is its current level of operating inefficiency.
An important source of TFP growth in the shorter run is the degree of
capacity utilization. Producers in most industries find it uneconomical to adjust
production capacity to short-run demand fluctuations. The capacity utilization
rates of industries therefore fluctuate.
TFP grows (declines) when capacity
utilization rises (falls) because output is apt to change much more rapidly than
capacity.
Another short-run determinant of TFP growth is the pattern of
expenditures that are more occasional than even in character. Expenditures of
this kind include those for certain kinds of maintenance and investments.
A
surge in expenditures can slow productivity growth and even result in a
productivity decline. Uneven spending is one of the reasons why the TFP growth
of individual utilities is often more volatile than the TFP growth of the
corresponding industry.
The TFP trend of a utility industry is an empirical issue.
Results of
productivity research have been presented in several PBR proceedings.
Regulators often choose X-factors without stating their views on the components.
There are, however, several cases in which they have explicitly acknowledged
the long run industry productivity trend. Here is a summary of the results.
58
Industry
Company
TFP Trend
Gas distribution
Boston Gas
0.4
Gas distribution
Southern California Gas
0.5
Gas distribution
San Diego Gas and Electric
0.7
Power distribution
0.9
Power distribution
Ontario
0.9
Power distribution
Southern California Edison
0.9
Telecommunications
Canadian telcos
2.6
Telecommunications
SNET – CT
2.1
Telecommunications
Ameritech – IL
1.3
Telecommunications
Nynex – ME
2.2
Telecommunications
Nynex – MA
2.0
Telecommunications
Ameritech – OH
2.8
Telecommunications
Bell Atlantic – PA
2.9
These figures have important implications for energy utility regulation.
One is that X-factors can reasonably be expected to be much higher in indexing
plans for telecom services than in plans for many energy services. The current
TFP trend for telecom utilities is apparently around two hundred basis points
higher than that for energy distributors.
This reflects, in the main, rapid
technological change and demand growth in the telecommunications industry. It
should not be surprising, then, to find approved telecommunications price cap
plans with X-factors at least two hundred basis points above those in approved
energy utility plans.
These productivity figures also help to explain why a multi-year rate freeze
may not financially stress telecom utilities as much as it would an energy utility.
Given input price growth in the 2-3% range, index logic suggests that telecom
utilities have recently experienced steady or moderately declining unit costs.
This permits them to prosper under rate freezes.
While energy utilities face an input price growth trend broadly similar to
that of telecom utilities, their TFP growth is typically much slower. Accordingly,
their input price growth is more likely to exceed their TFP growth, and their unit
cost is more likely to rise over time. This is a common situation in our economy
59
as can be seen by the tendency of the consumer price index to rise over time.
Many energy utilities will therefore have difficulty remaining financially viable for
an extended period of time without nominal rate increases. An American-style
PCI could address this situation by allowing utility rates to rise moderately each
year in nominal terms to keep pace with industry unit cost growth. The fact that
utility prices are apt to rise in nominal terms should by itself cause no more
concern than in competitive sectors of the economy.
REGIONAL RESEARCH FOCUS
An important issue in North American style index calibration is the choice
of a region for indexing research.
Many network industries are, like gas
distribution, natural monopolies. There is therefore no regional trade in most
distribution services that could be used to identify an appropriate regional
grouping. A sensible alternative is to choose a region with similar input price and
productivity trends. Industries in different countries can exhibit different unit cost
trends even if they are adjacent. Industries in different regions in countries of
some size can also exhibit different unit cost trends, for several reasons. One is
differences in regional economic growth. Variation in regional growth patterns is
evident in both Canada and the U.S. Differences in government policies can
lead to differences in the unit cost growth of utilities. For example, governments
can differ in support for demand-side management efforts that affect volume
growth.
This analysis suggests that the region surrounding a utility will tend to
have more similar input price and productivity trends than regions further afield.
These considerations suggest that the unit cost trend in the region surrounding
the subject utility can be the appropriate focus of input price and productivity
research. However, circumstances can render this option unworkable as well.
Some or all of the surrounding region may be in a different country. Additionally,
the surrounding region may have few peer utilities, lack good utility operating
data, or be dominated by just one or two utilities.
60
A review of North American indexing plans is useful in illustrating the
region selection challenge. In Ontario, regulators elected to base the inflation
and X-factors in the rate cap plan for power distributors on the input price and
productivity trends of the provincial industry.
The Massachusetts regulator
explicitly approved the calibration of the Boston Gas X-factor using the TFP trend
of northeast distributors.
The California gas distribution industry is dominated by three companies.
Neighboring states have much smaller economies and important differences in
operating conditions.
This consideration has been important in the CPUC’s
approval of X-factor calibrations for Southern California Gas and SDG&E that
were based on national industry productivity trends.
In
the
telecommunications
industry,
X-factors
in
a
number
of
telecommunications price cap plans for U.S. LECs have been established in
proceedings where the company’s own productivity trend was the featured
evidence. The FCC based its X-factor for interstate services of LECs on national
TFP research but has acknowledged its potential inappropriateness for certain
regions. The CRTC based its X-factor for LECs on national data in its first price
cap plan.
5.2.2 The British Approach to Index Design
The British approach to PCI design is that typical of utility rate regulation in
Britain. It has since been adopted in several other countries. Most notable,
perhaps, is its widespread use by regulators in Australia.57 Most British utilities
were formerly public enterprises.
British Telecom (BT) was the first to be
privatized, in 1984. Since then, privatization has extended to Britain’s electric,
gas, and water utilities.
The decision to use rate indexing in British utility regulation was strongly
influenced by the recommendations of Stephen Littlechild of the University of
Birmingham. In a report released in 1983, he proposed to adjust BT’s rates
57
Other countries using the building block approach include Ireland and Mexico.
61
using an index with a growth rate formula of “RPI-X” form.58 A specific value for
X was not recommended, nor was there significant discussion in Littlechild’s
paper of the appropriate framework to be used to determine X. Rather, the value
for X was described as “a number to be negotiated.” The lack of a well-defined
framework has given British regulators considerable discretion in determining
X-factors.
Over time, however, broadly similar approaches have developed
across Britain’s utility industries.
Under “British-style” rate indexing, rate cases are typically held every five
years. In contrast to North American practice, which focuses on a single test
year, the rate case involves detailed multi-year cost and output forecasts. The
principle “building blocks” of the total cost forecast are the forecasts of the value
of the current capital stock and of capital spending, depreciation, the rate of
return on capital, and O&M spending. A macroeconomic inflation index such as
the RPI is used as the inflation measure of the price cap index.
Given the
forecasts of growth in total cost, billing determinants, and the RPI, it is possible to
choose a combination of initial rates and an X-factor such that forecasted
revenue equals forecasted cost.
The British approach to the design of rate and revenue cap indexes has
several advantages over the North American approach. One is the possibility of
implementing it in situations where the North American approach is hampered by
a lack of historical data that could be used for productivity calculations. This was,
apparently, the situation in most of the British industries at the time of their
privatization. The British approach is also advantageous in a situation where
there really is no sizeable group of peers that could provide the basis for industry
productivity trends even if data were available. This was and continues to be the
situation in the British telecommunications and power and natural gas
transmission industries.
The British approach is also advantageous in situations where the
expected forward looking productivity trends of individual utilities are markedly
58
Stephen Littlechild, Regulation of British Telecommunications’ Profitability:
Report to the
62
different from the recent long-run TFP trend of the industry. This situation is
often encountered in industries, like power generation and transmission, that
have unusually bunched intertemporal patterns of investment. In that case, an
individual utility might, for example, anticipate large scale investments in the next
few years that will slow productivity growth markedly even though the recent
productivity growth of the industry was fairly rapid.
These advantages of the British-style approach to rate indexing should be
weighted against some important disadvantages.
One is the higher level of
regulatory cost that it involves. A five year test rate case is substantially more
complicated than the one-year test cases that are made possible by productivity
indexing.
The uncertainties of long term forecasts may also be said to
discourage plans with longer terms, such as the seven- to ten-year plans now
being approved in North America.
Another serious problem with the British
approach is the incentive that it provides to utilities to exaggerate their future cost
growth.
These disadvantages have spurred considerable innovation in British style
regulation in recent years. For example, statistical benchmarking is frequently
used to appraise O&M expense forecasts, and appraisals are often solicited from
outsiders concerning capital spending projections. An Australian regulator, the
Essential Services Commission in Victoria, now uses industry productivity
research to forecast future O&M expenses.
Secretary of State, February 1983.
63
6. SERVICE QUALITY PROVISIONS
The attainment of appropriate quality standards is a critically important
consideration in PBR plan design. Utilities can often save money by trimming
maintenance expenditures and capital investments that affect quality. In many
cases, the local utility is a monopoly provider and stands to lose fewer sales than
a competitive firm if service quality is off the mark.
The OEB notes the importance of service quality oversight in its Rates
Handbook decision. It states that
Any reduction in the quality and/or reliability of service represents a
reduction in the value of that service. Therefore, as part of its function in
regard to approving or fixing just and reasonable rates, the Board has a
responsibility to oversee that service quality is preserved and improved.59
Formal service quality incentive mechanisms have been approved for
numerous utilities.
They are a form of benchmark PBR which rewards or
penalizes a utility depending on the relationship between its measured quality of
service and quality benchmarks. There are three basic elements in a service
quality incentive plan: a series of indicators of the company’s quality of service;
an associated set of quality benchmarks; and an award mechanism that leads to
changes in utility rates or allowed returns. The indicators are measurable service
quality dimensions.
The benchmarks are the standards against which the
indicators are judged.
They can in principle be based on the company’s
historical performance, industry norms, or levels that are deemed to be
acceptable for other reasons. The award mechanism determines the adjustment
in rates that is warranted by the change in service quality. Important design
issues include the symmetry of awards and penalties and the customers’
valuation of specific quality indicators.
59
OEB, Rates Handbook Decision, ibid p. 50.
64
6.1 Benchmarking Basics
The benchmarking approach to PBR involves the evaluation of one or
more indicators of company activity using external performance standards
(benchmarks). The standards are external to the extent that they are insensitive
to the actions of subject utility managers. Evaluations and rate adjustments are
accomplished by formal mechanisms that are established in advance of use and
typically function for several years.
The key features of a benchmark plan are the performance indicators,
performance benchmarks, and the rate adjustment mechanism.
The
performance indicators used in approved benchmark plans vary greatly in scope.
Plans are comprehensive to the extent that they cover all of the utility
performance dimensions that matter to customers.
The performance benchmarks used in benchmark plans are also varied.
A common benchmark is a company’s activity level in a period just prior to plan
commencement.
A company is then rewarded for improvement in its
performance relative to recent history.
An alternative approach, which is an example of “yardstick regulation” or
statistical benchmarking, is to use the corresponding performance indicator of a
group of utilities. Under this approach, a company is rewarded for improving its
performance indicator relative to the group. The utility group is sometimes called
a peer group, and can consist of all utilities in the same region as the company
subject to the plan. In that event, the peer group may be viewed as a proxy for
the regional industry. In principle, the region can also be the entire nation.
The rate adjustment mechanisms in approved benchmark plans vary. A
major design issue is the customer sharing percentage. The mechanism may or
may not feature a deadband in which deviations from the benchmark do not
induce rate adjustments.
65
6.2 Quality Indicators
A critical issue in the development of effective service quality provisions in
a PBR plan is the choice of indicators on which performance will be judged.
Ideally, individual quality indicators should satisfy four criteria: 1) They should be
related to the relevant aspects of service; 2) focus on monopoly services; 3)
cover all major quality dimensions, and 4) be no more complex than necessary to
provide effective incentives.
First, since measured service quality can ultimately affect customer rates,
indicators should be linked to aspects of utility service customer’s value. This
may seem obvious, but a strict application of these criteria excludes indicators
that have been included in some plans.
For instance, the knowledge and
courtesy of phone center employees may be a legitimate quality indicator, but the
goal of establishing worker training programs to build these skills is not.
Second, indicators should focus on the quality of the activities for which
there are few if any alternative suppliers. This is consistent with the principle that
regulation, including regulation of service quality, is less necessary in competitive
markets.
Market forces are likely to create acceptable quality levels when
products are available from multiple providers. Third, quality indicators should
not focus on some important areas while ignoring others because performance
may deteriorate in the non-targeted areas.
Comprehensiveness can be achieved simply by adding indicators to a
plan. However, regulatory costs often rise accordingly since more utility and
commission resources must be devoted to quality monitoring, measurement, and
the reconciliation of findings related to quality indicators. Some commissions
have been sensitized to the regulatory costs of complex service quality plans. In
these jurisdictions, service quality incentives have been simplified by relying on
fewer, but more broadly-based, indicators. While the specific indicators may vary
widely among approved service quality incentive plans, there are broad
similarities between the types of indicators used for energy utilities. The most
common categories of indicators are reliability and customer service.
66
6.3 Quality Benchmarks
Quality benchmarks are the standards against which measured quality is
judged.
Benchmarks should be ideally sensitive to the external business
conditions which influence a utility’s quality. These business conditions may be
called quality “drivers”. The list of relevant factors includes weather (e.g. winds,
lightning, extreme head and cold), vegetation (contact with power lines), the
amount of undergrounding mandated by local authorities, the degree of
ruralization in the territory (typically increasing the exposure of lines to the
elements and lengthening response times when faults occur), the difficulty of the
terrain served, and regulatory changes such as a restructuring of the industry to
promote competition. These drivers can vary considerably between utilities and
over time.
Universally accepted quality standards do not exist for utility industries, so
commissions have considerable latitude in setting benchmarks. For any given
indicator, one straightforward benchmark is the utility’s average performance
over a recent period. Quality assessments would then depend on measured
quality levels that differ either positively or negatively from recent historical
experience.
Using past utility performance to set benchmarks is appealing in many
ways.
The data are of known quality and reflect local cost drivers.
The
construction of benchmarks from a utility’s past quality level should reflect the
fact that a company’s measured quality can be affected by quality drivers that are
volatile and prone to fluctuations that are hard to predict. Utilities should not
ideally be subject to penalties or rewards because random factors have affected
their measured service quality.
PBR plans can be designed to mitigate the
impact of random factors that might lead to inappropriate penalties or rewards.
One way to handle the impact of fluctuations in quality drivers is through a
deadband around the quality benchmark in the award/penalty mechanism.
Statistical methods can provide a rigorous foundation for setting deadbands that
reduce the probability of inappropriate penalties or rewards to specified levels
67
(e.g. 5%). Such statistical methods have been used in several service quality
PBR plans for telecom utilities and have been proposed by energy utilities in
some states.60
Statistically based dead bands should reflect historical fluctuations in
indicator values.
This is commonly measured by the standard deviation of
sampled values. The greater the fluctuations have been, the higher the standard
deviation and the wider the deadbands.
Statistically based deadbands also
reflect the size of the sample. The deadband should be wider the smaller is the
sample.
Regulators may not consider a utility’s past performance to be an
adequate quality standard, especially if recent service levels were deemed poor.
Some utility managers may also view the company’s history as inappropriate
when its performance is exceptionally good. In this case, it may be considered
unfairly demanding to expect the utility to match its historically superior
performance on an ongoing basis.
An alternative to basic benchmarks on the Company’s own history is to
base them on the service quality performance of the industry. The industry may
take the form of a national or regional sample or a peer group selected by other
means. In principle, industry-based benchmarks may be attractive in PBR. They
are clearly external to the subject utility, which creates strong performance
incentives. Industry benchmarks also tend to be consistent with the operation of
competitive market, where customer choices are driven by the cost and quality of
products relative to available substitutes.
In practice, however, industry-based benchmarks are often problematic.
One reason is that uniform and publicly-available data on quality are not
collected for large numbers of energy utilities.
Another reason that industry-
based benchmarks are problematic is differences in the operating conditions of
utilities. Optimal quality levels reflect such key conditions as the cost of providing
60
“Investigation by the Department of Telecommunications and Energy on its own motion to
establish guidelines for service quality standards for electric distribution companies and local
gas distribution companies.” Massachusetts D.T.E. 99-84 (June 29, 2001).
68
quality service and the demand for quality. These conditions vary across service
territories. The issue of key importance is whether a company’s quality level is
good given the quality drivers that it faces.
It is difficult to obtain a sizable
amount of quality data from companies that are similarly situated.
6.4 Award and Penalty Rates
Another significant plan design issue is the magnitude of any rewards or
penalties levied. In practice, empirical evidence is rarely presented to justify the
amount of potential penalties or rewards in a plan. Instead, penalty levels are
sometimes chosen with the idea that they are “significant” enough to prevent
service quality declines.
Ideally, a service quality incentive requires information on how customers
value different quality indicators, so that the potential rewards and penalties for
performance will reflect the value of the service provided. Given its importance, it
is somewhat surprising that little empirical work has been done on customer
valuations of quality indicators included in incentive plans. In part this is because
quality is inherently difficult to value.
But while this information may not be
readily available, it can be gathered from a number of sources.
Although a complete discussion of the topic is beyond the scope of this
report, three basic methods are used to estimate the value of service quality.
One method uses proxy data related to the service attribute. For example, the
value of having to wait for a field service representative to arrive can be
approximated as the customer’s lost wages (i.e., the opportunity cost of the
customer’s time). Proxy prices have the advantage of simplicity, but they can be
imprecise and bear a tenuous link to actual service valuations.
A second method of estimating customer valuation uses market-based
measures for the value of service. The difference between firm and interruptible
rates is one example of market-based data that reflects some customers’
valuations of reliability. Another example of market-based measures is the use
of hedonic price indexes, which are developed by regressing market prices on
69
identifiable quality attributes. Hedonic price indexes reflect the notion that price
differences are due to implicit markets for individual product characteristics.
Some official statistics utilize hedonic methods.
For example, the Bureau of
Labor Statistics adjusts for quality changes of some products when computing
the Consumer Price Index. While market-based methods are often conceptually
sound, they can be controversial, are often not well-understood, and can produce
divergent estimates of underlying quality valuations.
In addition, hedonic
methods are less likely to capture the underlying quality valuations in utility
markets since prices often reflect regulatory decisions rather than market forces.
Finally, quality valuations can also be obtained through customer surveys.
An advantage of this approach is that surveys can focus on specific aspects of
utility services that might be included in an incentive plan. However, survey
results reflect subjective perceptions rather than actual consumer behavior, and
hypothetical valuations may not be a good guide to how consumers would
actually act in markets.
6.5 Plan Symmetry
The symmetry of the award mechanism is another important design issue.
It has been argued that symmetric awards (i.e. both rewards and penalties are
possible) are not needed when quality incentives are designed only to maintain
quality levels which might otherwise decline due to the stronger incentives to cut
costs under PBR. However, symmetric plans can be calibrated to incent only the
maintenance of current quality standards.
The encouragement of better quality may, in any event, be desirable. All
types of PBR, including service quality incentives, are fundamentally motivated
by a desire to improve utility performance and not simply to prevent performance
from slipping.
Asymmetric plans generally do not create incentives for
companies to improve quality and thus may limit the total customer benefit that is
available from utility operations.
70
The impact of external business conditions on measured service quality
performance also tends to support symmetric service quality incentives.
As
noted, some business conditions can be quite volatile and may lead to
inappropriate penalties or rewards. Symmetric service quality incentives reduce
the likelihood that random factors will lead to inappropriate net penalties or
rewards over the course of a multi-year incentive plan. That is because random
changes in business conditions can lead to rewards as well as penalties. Over
time, the magnitudes of any inappropriate penalties and rewards can therefore
be expected to cancel each other out. This leads to reasonable penalties and
rewards that on average reflect a utility’s underlying quality performance. This
would not be the case with an asymmetric service quality incentive, where
external factors may subject a company to penalties without the chance of being
compensated with offsetting rewards.
Symmetric plans are also more consistent with the workings of
unregulated markets. Customers in such markets routinely pay higher prices for
higher quality products. Many farmers, for instance, do not have full control over
the quality of their produce from year to year and earn quality premia when
production conditions are favorable as well as lower prices when they are
unfavorable.
However, competitive markets usually offer an array of goods with varying
quality levels, and not all customers choose to consume high-quality goods. In
some cases, incentive plans lead to price increases on monopoly services.
Where this is the case, at least some customers may be paying for quality
improvements that they do not want.
The uncertainties related to the magnitude of rewards or penalties lend
additional support for symmetric service incentives over asymmetric incentives.
Since regulators often use considerable discretion in setting penalty rates, a
symmetric plan may discipline regulators into choosing more appropriate rates.
That is, with an asymmetric plan, regulators may err on the side of choosing very
high penalties to assure that quality does not decline under the plan. This is less
71
likely under a symmetric plan, which would require an equally high reward due to
performance improvements. Hence, even if an asymmetric plan is ultimately
approved, a symmetric service quality proposal may be beneficial if the prospect
of symmetry leads to more appropriate magnitudes for penalty payments.
6.6 Informal Quality Provisions
Service quality PBR is becoming more important in utility regulation.
Quality incentive mechanisms can play an important role in ensuring that
incentives for quality and unit cost containment are balanced.
Despite their
importance, research to place these plan provisions on a solid foundation of
reason and empirical research is not well advanced.
The many challenges encountered in the design of benchmark incentive
mechanisms for quality, combined with the dearth of good research in the field,
make it reasonable to question whether such mechanisms are the best way to
regulate quality in PBR plans. Continuation of traditional quality regulation, which
holds the utility responsible for quality and obliges it to address any deficiencies,
remains a sensible alternative. A hybrid system is also worthy of a consideration
in which the utility is obligated to make regular reports on a set of quality
indicators.
6.7 Precedents
There are a large number of formal service quality provisions in approved
rate plans. Service quality PBR is especially well established in New York and
California.
Generic proceedings on service quality PBR have been held in
several states.61
Symmetric service quality plans have been approved for energy utilities.
For example, both the California and New York commissions have adopted
symmetric service quality plans based on explicit findings that the underlying
61
See, for example, Massachusetts D.T.E. 99-84, op cit.
72
principles are sound. However, asymmetric service incentives are somewhat
more common.
Despite the many precedents for formal service quality incentive
mechanisms, many PBR plans do not have them. The absence of incentive
mechanisms is especially common in first generation plans. For example, the
OEB did not approve a formal mechanism for power distribution in its Rates
Handbook decision. It stated in the decision that
The Board recognizes that electricity industry restructuring introduces
many unknown factors that could impact on performance levels and
customer expectations. Further, there is a lack of consistent information
on historical performance. Therefore, the Board is of the view that, for first
generation PBR, a cautions approach to introducing service quality
performance indicators and standards is warranted. The proposed
approach in first generation PBR appropriately focuses on data collection,
reporting, and monitoring of service quality and reliability performance by
all distribution utilities.62
The Board also elected not to approve a formal quality incentive mechanism in
the first general Union Gas PBR plan.
62
OEB, Rates Handbook decision, ibid p. 50.
73
7. BENEFIT SHARING PROVISIONS
7.1 Introduction
As I explained in Section 2, a well-designed PBR plan generates stronger
performance incentives with fewer operating restrictions than cost of service
regulation. Performance is expected to improve under such a plan, and utilities
can earn more and their customers pay less – at the same time – than could be
the case under cost of service regulation.
The details of a PBR plan will
influence the allocation of plan benefits between utilities and their customers, and
the proper mechanism for sharing plan benefits is a controversial issue in many
PBR proceedings.
Benefit-sharing provisions should allow both shareholders and customers
to fare better than under standard rate regulation. If PBR is voluntary, utilities
have little incentive to agree to a plan unless it offers a reasonable chance for
higher earnings, especially in view of the higher risk entailed. It is incorrect, then,
to point to higher utility earnings under PBR as evidence of its “failure.” Higher
utility earnings are consistent with successful PBR as long as customers also
benefit compared with a continuation of the status quo.
The selection of a benefit sharing mechanism should be based on
sensible criteria. I evaluate alternative sharing mechanisms primarily in terms of
their effect in three areas: performance incentives, cross-subsidization, and risk
reduction. Other attributes considered include simplicity and “salability,” (i.e., the
ability to convincingly demonstrate benefit sharing).
Various PBR plan provisions influence on customer benefits. These can
be grouped into two general categories.
One is predetermined sharing
provisions such as initial rate cuts and enhanced rate trajectory. These are so
called because they are determined in advance of plan operation and are
delivered to customers whether or not performance actually improves. A second
general category of benefit sharing provisions is “real time provisions.” These
74
include earnings sharing and cost-based rate resets.
Customer welfare also
depends, of course, on the market responsiveness of rate and service offerings
and on service quality. In this section, I focus on the measures that are most
expressly devoted to benefit sharing.
In this section, I analyze the salient benefit-sharing provisions. I describe
the basic features of each approach, detail important precedents, and evaluate
its advantages and disadvantages as a means of benefit-sharing.
7.2 Enhanced Rate Trajectory
One way to share the benefits of PBR is to enhance the rate trajectory so
that it is more favorable to customers. Consider first how this might be done in
the context of a rate or revenue requirement index. The X-factor in such indexes
influences allowed rate escalation. A higher value for X benefits customers of
regulated services. An X-factor designed in accordance with North American
principles is calibrated to reflect the TFP trend of the relevant industry. One way
to share expected plan benefits with customers, then, is to set the X-factor at a
level above the calibration point. This component of the X-factor was noted
above to be called a stretch factor. It is set in advance to help ensure an external
character for X. However, it can be allowed to vary from year to year. Stretch
factors have been featured in many North American indexing plans. They are
sometimes explicit and sometimes implicitly added to the X-factor.
Rate freezes do not involve explicit stretch factors but often contain
sizable implicit ones. Suppose, for example, that input price inflation is 2% and
normal productivity growth is 1%. The stretch factor implicit in a rate freeze
would in this case be 1%.
The growth trend in rates is not the only way that customer welfare is
affected by the rate trajectory. Customers are also affected by the extent to
which the company absorbs risk. The base productivity factor, for example, is
more than just the offering of the benefit of normal productivity growth. It is,
furthermore, a commitment by the company to provide said benefit over a multi
75
year period during which actual productivity growth may be quite different. A rate
freeze offers the customer protection against input price as well as productivity
risk.
An important advantage of stretch factors is that their values can be
assigned independently of a company’s activities during the plan. Stretch factors
therefore do not compromise performance incentives or operating flexibility.
Valuations made prior to the first indexing period clearly have this attribute
The appropriate stretch factor depends in part on the prospects for
productivity growth during the plan term. Expected productivity growth should by
this logic be lower the greater is the efficiency of the company. Benchmarking
studies can shed light on a company’s operating efficiency.
However, such
studies invite controversy and good studies are expensive. Absent such work,
regulators should take careful note of the regulatory system under which a
company has operated.
Regarding their salability, stretch factors are appealing to regulators
insofar as they represent an advance commitment to customer benefits.
Customers therefore benefit whether or not performance improvements are
realized—at least during the term of the plan. On the other hand, customers and
their representatives may not understand that stretch factors are designed to be
insensitive to a utility’s current earnings and may resent high earnings if they
occur. It is helpful in this regard for regulators to acknowledge the value of
stretch factors and the long run benefits of high earnings when approving PBR
plans.
7.3 Initial Rate Cuts
A less common approach to sharing plan benefits is to lower the initial
(base year) rates or revenue requirement below the levels that would otherwise
result.
When this is done, consumers immediately reap a plan benefit.
Moreover, benefits continue to be created in subsequent years since, with lower
76
initial rates, lower prices result from index-based rate adjustments. This approach
has been more widely used in Great Britain than in North American PBR to date.
The advantages and disadvantages of initial rate cuts as a benefit sharing
mechanism are similar to those for stretch factors. To the extent that rate cuts do
not deepen in successive plans in response to performance improvements,
performance incentives are strong. Cuts at the outset of the first plan do not
affect incentives. The concern is, instead, with the size of initial rate cuts that
might occur at the start of subsequent plans and their linkage to past
performance improvements under PBR. As with stretch factors, initial rate cuts
do not mitigate business risk and can actually increase regulatory risk absent a
proper conceptual and empirical foundation. Customers benefit whether or not
utility performance improves but may resent high earnings if they occur.
A unique advantage of initial rate adjustments is the immediacy of the
benefits.
On the other hand, a unique disadvantage is the difficulty of
demonstrating that rate cuts are in fact being made when, as is common,
companies propose rate increases just prior to indexing. Utilities are then in the
awkward position of claiming that they could have asked for even larger price
increases and that customers have benefited from the company’s restraint.
Since other parties will have differing opinions about the warranted rate hike, the
benefits may be less convincing.
Regulators considering initial rate cuts should recognize that they are in
lieu of other benefit sharing provisions. For example, any initial rate cut should in
principal reduce the appropriate stretch factor.
recognized in British-style PBR.
This principal is clearly
Regulators in Britain and Australia explicitly
discuss how plan benefits are to be divided between rate cuts and higher
X-factors.
77
7.4 Earnings-Sharing
7.4.1 Description
An earnings-sharing mechanism (ESM) adjusts a company’s price
restrictions when its rate of return (ROR) has been in a certain range over a
recent historical period. A typical ESM provides for rate adjustments when the
actual (pre-sharing) ROR differs from a target ROR by certain prescribed
amounts. The mechanisms are established in advance of their use and typically
function for several years. The most widely-used rate of return in ESMs is return
on equity (ROE)
Approved ESMs vary significantly in several ways. The most important
difference is the shares of surplus (and/or deficit) earnings assigned to
shareholders and customers. These shares may differ in different ranges around
the target ROE. Many plans feature a deadband around the target in which rates
are insensitive to ROE fluctuations.
Immediately beyond the deadband, the
customer share is commonly 50%.
In some plans, it increases substantially
when ROE is extraordinarily high and falls substantially when it is extraordinarily
low. Thus, the company share falls with the extent of surplus earnings. ESMs
with this attribute are sometimes called “regressive.”
Alternatively, a
“progressive” ESM increases the company’s share of benefits as surplus
earnings increase.
Some plans are symmetric in the sense that they provide for rate
decreases when earnings are high and similar rate increases when earnings are
commensurately low.
Other plans provide for rate adjustments only when
earnings are high or low. For example, a plan approved for a Maine utility shares
earnings deficits but not surpluses.
Other plans share only surpluses.
The
symmetry of an ESM can, naturally, have a major impact on the risk-return
balance of a PBR plan.
78
7.4.2 Precedents
UNITED STATES
ESMs are one of the oldest approaches to PBR.
They were used in
England as early as 1855 to regulate local gas companies.63 A plan was adopted
in Canada in 1877 to regulate Consumers Gas. An early American plan was that
established in 1905 for Boston Consolidated Gas.
A plan for the Potomac
Electric Power, approved in 1925, remained in effect until 1955. ESMs have
been used recently by many U.S. energy utilities. Most recent PBR plans for
U.S. and Canadian energy utilities involve ESMs. However, ESMs were not
included in the PBR plans for National Grid (MA) or the plans approved by the
FERC for oil pipelines or the power transmission services of International
Transmission.
Experience with ESMs in the North American telecommunications industry
is also interesting. Most of the early price cap plans at both the federal and state
level included an earnings sharing mechanism (ESM) as an adjunct to the price
cap mechanism. For example, the original FCC plan for the LECs included an
ESM to provide a “backstop” in the event that the X-factors established by the
FCC were substantially in error or in the event that a particular LEC’s productivity
significantly differed from the average.64 In addition, the first price cap plans in
California (Pacific Bell and GTE-California in 1990), New York (Rochester
Telephone in 1991), Rhode Island (1992), and New Jersey (1993) all featured
ESMs.
However, the FCC’s later LEC price cap plan, adopted in 1997, did not
include earnings sharing.
The FCC believed that ESMs blunt the efficiency
incentives created by price caps since companies must immediately share the
benefits of efforts to reduce their unit costs.65 The FCC also noted that “the
63
64
65
For further discussion of the early precedents see Harry Trebing, “Toward An Incentive
System of Regulation:, Public Utilities Fortnightly, July 18, 1963, p. 22-37.
For example, see Second Report and Order, CC Docket 87-313, September 19, 1990, FCC
90-314, paras 120-165.
For example, see Fourth Report and Order, CC Docket 94-1, May 7, 1997, FCC 97-159, para
148.
79
removal of sharing also removes a major vestige of rate-of-return regulation that
created incentives to shift costs between services to evade sharing in the
interstate jurisdiction.”66 The FCC went on to state that the cost-shifting and
cross subsidy incentives inherent in rate-of-return-based sharing mechanisms
were at odds with the goal of promoting greater competition and eventually
deregulating LECs, as envisioned by the Telecommunications Act of 1996:67
Not only is sharing inconsistent with the general competitive paradigm that
was established in the 1996 Act, but sharing might make it more difficult to
deregulate services that become subject to substantial competition by
creating an opportunity for LECs to misallocate costs from deregulated
common carrier services to services that remain subject to sharing
requirements. As more and more incumbent LEC services become
subject to competitive pressures, the public interest detriments of the
cross subsidy incentives inherent in sharing become worse as the costs
that can be misallocated to services that remain subject to sharing
requirements increase. Without the elimination of sharing, it might
become necessary to adopt new structural or nonstructural safeguards to
prevent or limit these misallocations. Rather than consider adopting such
administratively burdensome requirements, I conclude that eliminating
sharing is the more reasonable course.68
Similarly, in state jurisdictions, ESMs are becoming increasingly rare as an
adjunct to price cap plans. Few states currently use rate indexing in conjunction
with an ESM to regulate the dominant LEC. In U.S. energy utility regulation,
ESMs are more common but many recent plans do not have them.
66
67
68
Id
Id, para 151.
Some FCC Commissioners were even more adamant in their opinion about the negative
features of earnings sharing. For example, Commission Chong stated that:
“I am particularly pleased that this Report and Order puts a stake through the heart of
‘sharing,’ the requirement that incumbent LECs earning more than specified rates of
return must ‘share’ half or all of the amount above those rates of return with their access
customers in the form of lower rates the following year. Since sharing continues the
inefficiencies of a rate-of-return era, I have long believed that a system of pure price caps
without sharing would be preferable. I believe that I have correctly found today that
sharing tends to blunt the efficiency incentives I sought to create through the price cap
plan.”
Separate statement of Commissioner Rachelle B. Chong, Fourth Report and Order, CC
Docket 94-1, May 7, 1997, FCC 97-159, p.2.
80
CANADA
In Canada, ESMs have been fairly common in PBR plans for energy
utilities. The OEB, for instance, approved the use of an ESM in its price cap
plans for Union Gas. Several plans that lack ESMs have featured benchmarkstyle sharing mechanisms.
Neither CRTC rate indexing plan for Canada’s
telecom utilities featured ESMs.
BRITAIN AND AUSTRALIA
Regulators in Britain have considered the adoption of ESMs on several
occasions. One review of a British Gas plan featured an especially thorough
deliberation of this issue. However, few ESMs have been adopted to date in
Britain. There are also no ESMs in the approved index plans for Australia’s
power transmission and distribution utilities.
7.4.3 Evaluation
ESMs have some important advantages as benefit sharing mechanisms.
One is their ability to mitigate risk. This property is, of course, greater when
ESMs are symmetric. ESMs are an automatic means of adjusting rates for a
wide range of risky external developments. This can be appealing where risks
are substantial or Commissions lack the technical expertise to approve
alternative risk mitigation measures such as industry-specific input price indexes.
As an alternative to initial rate reductions and X-factors, ESMs also reduce
regulatory risk.
In effect, benefits are shared as realized and there is less
pressure on regulators to choose stretch factors and initial rate reductions that
share the unknowable plan benefits. There is, however, some regulatory risk to
the utility in proposing an ESM: principally, the risk that the Commission will
approve an asymmetric ESM in which earnings shortfalls aren’t shared.
In addition to risk management, another benefit of ESMs is their
popularity. Many stakeholders appear to believe that ESMs align shareholder
and customer interests. If a distributor had a 14% ROE last year, for instance,
the ESM might reduce the revenue from regulated services by the value of 100
81
basis points of ROE.
ESMs also help keep utility earnings within politically
acceptable bounds.
On the downside, ESMs do not by themselves guarantee that customers
benefit from a PBR plan. Stakeholders may complain if utility earnings fail to
reach the sharing range. Failure to reach the sharing range is especially likely
when there are low initial rates or a high stretch factor. Stakeholders must also
remember that their rates may go up during an earnings shortfall.
Another disadvantage of ESMs is that the continued focus on earnings
keeps alive inherently controversial issues like utility-affiliate transactions and
cost allocations between a utility’s various regulated services and any
competitive market services.
This can give rise to controversies in ESM
implementation hearings. Regulators may anticipate this and deny the company
operating flexibility.
The effect of ESMs on performance incentives is complicated. Compared
to a multiyear plan in which rate restrictions are completely insensitive to a
utility’s performance, a plan with an ESM should in theory weaken performance
incentives. After all, utility managers have less incentive to improve performance
if half of the after-tax benefits go to customers. On the other hand, the practical
reality is that the inclusion of an ESM in a plan may encourage interested parties
to agree to an extension of the period between plan reviews. ESMs may also
help the parties agree to plan termination provisions that have less deleterious
incentive consequences. For example, it can be agreed that in the event of any
cost based true-up of rates at the end of the plan, a company is entitled to keep
its share of any surplus earnings and is not entitled to compensation for its share
of surplus losses.
The analysis of the impact of ESMs on the direct cost of regulation has a
similar flavor. ESMs increase regulatory costs during periods where companies
are not otherwise subject to regulatory intervention, such as a multi-year rate
plan.
For example, with ESMs it may be necessary to compute the cost of
regulated services, and therefore to allocate total cost between regulated and
82
unregulated services.69 This effect is offset to the extent that the inclusion of an
ESM in a plan can persuade stakeholders to agree to extend the period between
formal rate cases.
The reasons for the prevalence of ESMs in the approved PBR plans of
North American energy utilities and their relative paucity in the PBR plans of
telecom utilities merit brief consideration.
Two explanations seem plausible.
First, cost allocation issues have historically loomed larger for telecom
companies than for energy utilities due in part to the greater competitive
pressures. Because customers have so many alternatives to utility service, the
marketing and cost allocation issues that result from ESMs may be more costly
for telecom utilities.
A second reason for the discrepancy in the use of ESMs may be the
relative novelty of PBR for energy utilities. As noted above, many early PBR
plans for telcos featured ESMs, but earnings-sharing in the industry has become
rarer over time. Similarly, ESMs may become less common for energy utilities as
regulators and parties gain experience with PBR, including better knowledge as
to all the costs associated with sharing mechanisms.
7.5 Plan Termination Provisions
Plan termination provisions are provisions for what happens to regulation
on the occasion of a PBR plan’s termination. These typically involve a formal
rate case under both North American and British style index plan design
methods.
Two issues are salient in the specification of plan termination
provisions. One is the plan term, which is the duration of time between formal
rate cases. The other is the degree to which rate resets reflect other, external
considerations.
69
This is a major concern for telecom utilities, which typically provide extensive regulated and
unregulated services from the same facilities.
83
7.5.1 Plan Term
Most PBR plans specify the term of their application. Formal rate cases
will typically not be held during this term.
PRECEDENTS
The trend in PBR has clearly been towards plans of longer term. Plans of
three year’s duration were typical during the 1990’s. More recently, five year
terms have become standard and some plans of considerably longer duration
have been approved. Especially noteworthy in this regard are the ten year plans
for power distribution services of National Grid in Massachusetts and New York
and gas
distribution services of Berkshire Gas and Boston Gas in
Massachusetts.
EVALUATION
The rate case typically held at the termination of a plan is an important
opportunity to share plan benefits with customers. Thus, short plan terms let
customers share in benefits sooner. Short plan terms also reduce business and
regulatory risk. This makes them more suitable for businesses undergoing rapid
change or for regulatory jurisdictions where there is exceptional risk of unusual
stretch factors or initial rate adjustments.
On the other hand, plans of longer duration strengthen performance
incentives and alleviate concerns about cross-subsidies and novel operating
practices that can lead to operating restrictions. Longer terms are especially
useful in encouraging initiatives that involve up front costs to achieve long-run
efficiency gains. That is one reason why longer plan terms are of interest in PBR
plans occasioned by utility mergers.
Both of the National Grid plans just
mentioned involved mergers. The risk of a longer plan term can be reduced by
several other plan provisions, including industry-specific inflation measures,
Z-factors, marketing flexibility, and earnings sharing mechanisms.
84
7.5.2 Rate Reset Provisions
DESCRIPTION
The rate reset provisions of PBR plans can in principle involve widely
varying degrees of externalization. At one extreme, rates may be reset entirely
on the basis of a rate case and thus reset the company’s rates to its cost and
output. At the other, a plan could be reset entirely on the basis of external data.
For example, a rate or revenue cap index could be revised only to better reflect
the recent unit cost trend of the relevant industry.
The middle ground includes a number of possible options. One idea is to
set the new rates as an average of the rates resulting from a new rate case and
the rates resulting from one year’s continuation of the old PBR mechanism. If the
company has been operating under an ESM, another idea is to permit the
company to keep its share of surplus earnings.
PRECEDENTS
Rate cases are a common input into the resetting of rates for energy
utilities worldwide. For example, AmerenUE was permitted to keep some surplus
earnings under an ESM at the time of a PBR plan update. These permit the
company to keep some of the benefits of efforts to engineer long-term
performance gains. In Britain and Australia, where rates reflect multi-year cost
forecasts, several approved plans provide for companies to keep a share of
lower-than-forecasted cost during the next plan.
These provisions are some
times called “efficiency carryover” mechanisms.
EVALUATION
Rate reset mechanisms have a major impact on customer benefits from
PBR. A rate reset that is based entirely on a rate case passes to customers the
full benefit of cost savings achieved. Risk is reduced.
Yet rate reset mechanisms also have a major impact on the incentives to
make long term performance gains. To the extent that a full cost-based rate trueup is not ensured, performance incentives are strengthened and there are
reduced concerns about cross subsidies and novel practices that can lead to
85
operating restrictions. Incentives for initiatives involving up front costs and long
term benefits are, once again, especially affected. On the other hand, partial rate
resets could prove problematic if the reset occurs when the utility is embarking
on a program of major capital investments.
86
7.5.3 Concluding Remarks on PBR Plan Design
Our discussion has revealed that many tools are available for the
construction of PBR plans for energy utilities.
These tools have differential
impacts on performance incentives, operating flexibility and customer benefits.
It’s challenging to design a plan that strikes the right balance.
The benefits from PBR are maximized by plans that generate strong and
balanced incentives for a wide array of activities. For example, plans should
encourage utilities to strike the right balance of attention between cost
containment and service quality. Benefits are typically greater for comprehensive
rate or revenue cap plans than for non-comprehensive plans.
Benefits are
greater for price cap plans with marketing flexibility than for revenue caps,
especially when they facilitate better utility marketing.
Our analysis has also highlighted the importance of encouraging energy
utilities to undertake initiatives that involve up-front cost to achieve long term
performance gains. Plan termination provisions play an especially critical role in
the incentives for such initiatives. The greater risk of provisions that strengthen
such incentives can be offset by more careful attention to eliminating
unnecessary sources of operating risk under the plan.
Regarding the risk-return balance, careful plan design can help to achieve
a risk-return balance that is right for utilities and their customers. Tools that
reduce risk without unduly raising concerns about performance incentives and
operating practices are especially desirable. For example, an industry-specific
input price index can track fluctuations in a company’s input prices better than a
macroeconomic output price index. An X-factor based on a regional rather than
a national TFP trend may better reflect the realistic expectation for unit cost
growth.
The Z-factor can reflect changes in government policy and other
worrisome external developments.
The importance of tailoring plans to fit the circumstances of a utility must
also be stressed. When it comes to PBR plan design, one size does not fit all.
Utilities vary in their productivity growth expectations, risk exposure, and need for
87
marketing flexibility. Different plans are therefore indicated if all are to properly
balance risk, return, and customer benefit considerations.
88
III. PBR FOR POWER TRANSMISSION
In the final sections of the report I consider the application of PBR to
power transmission.
In Section 9, we provide a general discussion of the
transmission business and its implications for PBR. There follows in Section 10
a review of precedents for PBR in the United States, Canada, and Australia. We
conclude the report in Section 11 with an examination of the situation of
HQ TransÉnergie and the potential advantages of a PBR approach to its
regulation.
8. THE POWER TRANSMISSION BUSINESS
8.1 Transmission Service Supply
Power transmission is the long distance transportation of electricity.
Power is moved over stationary conducting lines. These are usually elevated
above the ground by towers or poles.
In urban areas, however, they are
sometimes routed through underground conduits.
Transmission is conducted most economically at much higher voltages
than those at which power is generated or locally distributed. The transmission
business therefore involves an extensive amount of voltage transformation. This
occurs at transmission substations, where voltage is raised in preparation for
long distance transport or lowered in preparation for local delivery.
The operation of transmission grids is extremely complicated.
The
preservation of system integrity requires that the quantity of power receipts must
be matched almost exactly by the quantity of deliveries at each point in time.
System integrity can also be jeopardized if power flows at any point on the
system exceed available transfer capacity. A transmission system operator must
also control the quality of power.
The complexity of the task increases with the
number of receipt and delivery points and power shippers.
89
The management of power flows and the control of power quality requires
sophisticated software and an array of specialized equipment. Power supplies
can also be useful. Suppose, for example, that a shipper requests delivery of
power across a congested interface. The transmission operator can enable such
a trade by inducing an increase in the supply of power in the area of the
requested delivery.
This can be a cost-effective alternative to additional
investments, especially in cases where the congestion is only occasional. Power
supplies are also used to balance the system and to control power quality.
Power transmission technology is highly capital intensive. The combined
cost of conductors, structures, substations, and other transmission plant typically
accounts for more than 70% of the total cost of service. This cost share is
considerably greater than the corresponding share for power distribution. One
reason is that the transmission business involves only large-volume customers
so that there is less need for labor-intensive billing cycle and customer
information (e.g. call center) services.
The relationship of transmission cost to output is another important
consideration.
In the long run, the important output-related drivers of
transmission cost are peak load, distance shipped, and the number of locations
at which pickups and deliveries must be made. Load is generally more peaked
to the extent that it is ultimately used for either air conditioning or space heating
and is not used in interruptible business applications.70 The distance power is
shipped depends chiefly on the distance between sources of system supply and
demand. It also depends on the tendency of receipt and delivery points to lie
along a few linear routes. The number of receipt and delivery points matters
because each point requires a substation, much as interstate highway entry and
exit requires specialized ramps.
The impact of output on transmission cost has a radically different impact
in the short run. The great bulk of cost is fixed and fluctuations in volume or peak
load often have little impact on cost up to point at which capacity is fully utilized.
90
At that point, system integrity can be compromised in the absence of plant
additions. Line losses are the chief variable cost of system operation. These are
greater the greater is distance shipped but nonetheless account for only a
modest share of total cost over typical transmission distances. The fixedness of
most transmission costs in the short run means that productivity growth is thus
highly sensitive to output growth that doesn’t tax capacity.
Economies of scale can be realized in transmission. Special features of
power transmission that encourage scale economies include the coordination
problems that would otherwise form long distance trade over more balkanized
systems.
The benefits of scale economies help to explain why, in many
countries, there are far fewer power transmission utilities than distribution utilities.
In Canada, for example, transmission is regulated at the provincial level and
these regulators typically have oversight over only one transmission utility, which
is often sizeable. In Ontario, this contrasts with a responsibility to regulate more
than a hundred power distributors. In Britain, there are three power transmission
utilities and fourteen power distributors.
In the Netherlands, the national
regulator has jurisdiction over one transmission utility and more than a dozen
distributors. In Australia, the national regulator has oversight over five major
transmission utilities. There are around fifteen power distributors in that nation.
The intertemporal pattern of investment in power transmission facilities
differs considerably from that in gas or electric distribution.
Investments in
distribution tend to be spread rather evenly over time because the growth of
urban areas, where consumption is typically concentrated, involves a horizontal
expansion of the area of economic activity into areas that were previously rural.
This attribute of distribution is not shared by power transmission.
In the
populated parts of North America, for instance, a transmission grid spanning the
continent was constructed years ago in a flurry of construction that was largely
completed by 1970. This grid supported the growth in U.S. power consumption
for many years. Transmission construction was also slowed for many years by
70
Extensive use in both air conditioning and space heating in a region tends to reduce load
91
an overbuild in the generation sector of the industry.
A rapid increase in
transmission construction is foreseen in the coming decade as new power plants
are built and the grid is strengthened to support more long distance trade.
As in other utility businesses, opportunities exist to outsource certain
transmission services. Some tasks can be economically outsourced. Mergers
may produce scale economies. New technologies are available for adaptation.
These include improved conducting materials and the use of monitoring and
communications equipment to permit real-time rating of transmission facilities.
8.2 Transmission Service Demand
The demand for power transmission arises chiefly from the fact that it is
often efficient to locate generation at sites that are distant from major load
centers. This is so for many reasons.
•
Generation cost is typically lower closer to sources of primal energy such
as coal and gas fields and hydropower sites. These locations are often
distant from consumption centers.
•
There are certain economies to be realized from larger generating stations
and the capacity of these stations often exceeds local needs.71
•
Different regions sometimes have different demand peaks so that it is
economical to meet demand peaks in one region from idle generating
facilities in another. A good example of this is the trade between eastern
Canada and the northeast United States.
•
Generation in cities involves higher land and labor costs.
It can also
involve undesirable safety, noise, visual, and air and water quality
externalities.
Air quality is a special concern with coal and oil-fired
generation. Safety is a special concern with nuclear generation.
•
Transmission can, by facilitating long distance trade, consolidate local
power markets into larger and more competitively structured regional
71
peakedness.
Large power plants are, for this reason, sometimes called “central” generating stations.
92
markets. This can be a lower cost way to promote low power prices than
alternative measures, such as the regulation of rates for generation or the
prohibition of local generation concentration.
The demand for power transmission differs in important ways from the
demand for power distribution. One reason is the greater importance of industrial
establishments and other large volume consumers of power in the transmission
business.
These customers typically account for about one third of North
American power consumption.
However, they often make little use of utility
power distribution services. Some large volume customers generate their own
power or take delivery of power at high voltage. In either case, they may bypass
the distribution system almost entirely. Others take delivery at subtransmission
or primary voltage but undertake local delivery services themselves. In contrast,
large volume customers typically make extensive use of transmission systems
unless they generate their own power.
It follows that the finances of the transmission business are more
dependent on the demand of large volume customers than are the finances of
the distribution business. These customers often have more elastic demands for
delivery services than the residential and small-volume business customers that
dominate distribution demand. This is true chiefly because they are more likely
to have access to cost-competitive alternatives to the transmission services of
the local utility. These alternatives include, most notably, self generation. This
alternative is especially cost competitive where the resultant waste heat can be
used in the production process of a customer or neighboring establishments.
Cost competitive alternatives to the use of a transmission system by a largevolume customer can also include the relocation of output to facilities in other
regions where the cost of power is lower. A customer can also have elastic
demand if it uses large amounts of power and its operations are economically
marginal due to an uncompetitive cost structure or unusually low prices for the
products that it makes.
93
Another
important
difference
between
the
demands
for
power
transmission and distribution is the extensiveness of on-system generation. In
contrast to the power (or, for that matter, gas) distribution business, a sizable
share of the energy delivered by a typical power transmission system is
produced on-system. Extensive facilities are devoted to the receipt of power
from on system power plants, including substations and extra line investments.
In power exporting regions, a sizable share of EHV capacity is also devoted to
the movement of locally generated power.
Sensitivity to local generation activity matters because these operations
sometimes display considerable demand elasticity and/or volatility. In the short
run, generators may or may not produce depending on whether net back prices
cover their marginal cost of operation.
Production from oil and gas fired
generators is especially volatile due to their high marginal cost and the volatility
of the fuel prices. The recent dramatic decline of traditional gas fired power
generation in Texas is an example. Receipts of power from wind forms are often
volatile due to the volatility of wind conditions.
In the long run, generating companies can choose the location of their
plants as well as their level of operation. Chronically marginal plants may close.
While a certain number of plants are bound to be located on a given transmission
system, the economics of some projects will be more marginal and hence more
sensitive to the expected terms of service. Included in this group may be certain
older plants for which owners are considering major life extensions. Customers
considering major new investments are naturally concerned about how the terms
of service may change over the service life of plants
Another difference between the demand for transmission and energy
distribution lies in the greater choices that customers sometimes have in routing
power shipments. Most power retailers have little choice but to use the local
distribution system to affect final delivery. Quite often, the power that an end
user consumes enters the distribution system at the nearest receipt point. In the
transmission business, however, power merchants can obtain power from
94
numerous locations and can move power produced to numerous locations. In
some cases, they also have a choice between alternative routings for shipments
between two points.
Routing choices can have a significant impact on transmission system
use. Demand is hard to predict, as it often depends on regional differences in
power prices. Routing decisions can also be sensitive to the offered terms of
service. For example, the use of a grid for long distance shipments can be
especially sensitive to the transmission charge for that service. So too will be the
use of the grid when the shipper has cost competitive routing alternatives. The
demand uncertainty and elasticity challenge pertain to components of the grid as
well as its overall use. For example, a rebound in local generation can cause
disuse of facilities that heretofore had been used for imports.
Different transmission services involve quite different costs for the
provider. Cost will generally be higher to the extent that shipments involve:
•
longer distance;
•
greater use of the system at times and places of peak system use;
•
more numerous receipt and delivery points;
•
receipt and delivery points that are distant from major transmission
corridors.
The differences between the demands for power transmission and
distribution that we have discussed here have important implications for utility
owners. Demand is more complex and involves greater operating risk. Different
service requests can involve markedly different costs.
System
use is more
sensitive to the terms on which services are offered.
Sources of significant
elasticity include:
•
large volume users that can self generate, can consume power at
alternative locations, and/or are economically marginal
•
economically marginal generations
•
generating companies seeking new production sites
•
shippers with alternative means to ship power between fixed points.
95
In summary, the marketing challenge facing the power transmission
business is more like that facing railroads than that facing power distributors.
Power marketing performance can substantially reduce system use and thereby
reduce the contribution that transmission can make to the North American
economy. For example, long distance trade may be unnecessarily discouraged,
thereby limiting the competitiveness of bulk power markets.
Poor power marketing can also make system use unnecessarily costly.
For example, a failure to price deliveries across a congested interface high
enough may ultimately resort in additional investment that isn’t warranted.
One aspect of power transmission demand that does not differ between
power transmission and distribution is the importance of service quality.
Electricity is vital to the operation of households and business establishments.
An outright failure to deliver power therefore has a high cost to customers. Other
dimensions of service quality also matter to customers. For new generators,
these include a timely response to requests for connections.
The special role of the transmission industry in containing the cost of
generation and promoting competition means that the assessment of service
quality is more complex than in distribution. It is not enough simply to ensure
that the flow of power to distribution substations is uninterrupted.
The
transmission industry will be judged, additionally, on its ability to accommodate
power trades that reduce the cost of generation and promote power market
competition. This is not to say that the industry should be expected to execute all
shipment requests. The goal is, instead, to be able in the long run to execute
shipment requests that have a value to customers that equals or exceeds the
cost of service. Useful information on the value of such services can be obtained
from examination of inter-regional differences in power prices.
8.3 Implications of Power Transmission PBR
The special features of the power transmission business have important
implications for the design of an appropriate regulatory system. Consider first the
96
issue of regulatory cost. In Section 2, we made the point that PBR is more
advantageous to the extent that it makes possible sizable savings in the cost of
effective regulation.
COSR is especially costly to the extent that it involves
numerous utilities and/or is inherently complex and controversial. In this section
we have noted that many regulators have jurisdiction over only a small number of
power transmission utilities. On the other hand we have seen that, for many
transmission utilities, the effective marketing of transmission services can be
quite complex. Utilities can, through their rate and service offerings, encourage
customers to use the transmission system in less costly ways. For example, they
can be encouraged to build power plants close to existing transmission lines and
economize on the use of congested interfaces by buying power or building power
plants in load pockets. Tight capacity can be allocated to the highest valued
users.
The rates and terms of service can also be used to encourage
transmission system use. The need for market-responsive terms is especially
great in uses that are elastic with respect to rates and other terms of service.
Elastic uses of the system include, as we have seen, shipments that involve:
•
unusually long distances
•
alternative possible routings
•
new generation.
We have shown that the price cap approach to PBR is extensively used in
utility industries that need complex, market-responsive rate and service offerings.
This approach is potentially useful in transmission regulation as well.
A
transmission utility that owns and operates its system, for instance, can be
granted greater flexibility to redesign rates for tariffed services and to offer
various optional rates and services. The profit motive will encourage them to
strike the best balance between the complexity and frequent change of offered
terms of service and the desire to use terms to manage system congestion and
promote system use. Making revenues dependent on the extent of system use
bolsters incentives to maintain or improve service quality.
97
This advantage of price caps is not, however, equally strong in all
transmission settings.
•
System operators that are not independent of market participants may not
be eligible for extensive marketing flexibility.
•
Alternatives to the price cap approach to achieving marketing flexibility are
available. These include secondary markets for firm transmission rights
and organized markets for power in which prices reflect capacity
constraints.72
•
A “rough and ready” approach to system rate design can sometimes do an
adequate job of achieving essential marketing goals such as the
encouragement of price-elastic system uses.
•
Independent system operators are prevalent in the United States,
Australia and parts of Canada. All of these organizations are not-for-profit
entities that do not own the grids that they operate. Both conditions limit
the effectiveness of financial incentives.
•
Price caps may, for some transmission utilities, involve an unusual amount
of risk for which regulators aren’t prepared to offer appropriate
compensation.
PBR is also more advantageous to the extent that established PBR
mechanisms can be readily and effectively implemented.
In Section 2, we
commented that the active ingredients for PBR include the combination of
external rate adjustment mechanisms and external data using economic
reasoning and empirical research.
In Section 3, we explained that a major
example of this is the North American approach to economic indexing. This
works best when the industry productivity growth trend of the recent past can be
calculated and is a good predictor of the same trend during the prospective plan
period. Certain features of power transmission business complicate the use of
this method:
72
Both of these alternatives can, however, be integrated into a price cap framework.
98
•
Power transmission is capital intensive and major investments are more
sporadic than in power distribution.
When major investments are
bunched, a transmission utility is therefore likely to experience a
temporary surge in unit cost that can cause earnings to plunge absent rate
increases. Timely rate hikes can avoid this outcome and reduce utility
concerns about the riskiness of relationship specific assets. A British-style
approach to indexing can finesse this problem since the price cap index
would in this case be sensitive to the pace of expected investment.
•
The calculating of productivity growth for the calibration of a price cap
index is complicated by the fact that such growth is very sensitive to the
growth in system use. It can be difficult to identify a productivity trend that
is commensurate with the expected growth in system use during the plan
period. A British-style approach to indexing can finesse this problem since
it is based on an explicit forecast of output growth.
•
In Section 2, we also commented that the short term advantages of PBR
are greater to the extent that a utility can slow its unit cost growth. In this
section, we have noted that an unusually high percentage of transmission
cost is capital cost. Most of this cost is not controllable in the short run.
The short term performance gains from cost containment are thus limited
in the transmission business.
In Section 9.1, we noted that transmission system operation sometimes
involves large expenditures on power. When such purchases are needed they
can involve high prices due to local market concentrations, a tight market, and/or
the high marginal cost of generation. It is difficult although not impossible to
develop a price cap or revenue cap index that provides appropriate
compensation for such contingencies. Such costs could be recovered outside of
PBR but this would create imbalanced incentive problems. For example, an
operator might make excessive power purchases to finesse congestion in order
to avoid the cost of an investment that is actually cost effective.
99
9. PRECEDENTS FOR TRANSMISSION PBR
9.1 United States
9.1.1 An Introduction to the FERC
Most power transmission services of U.S. electric utilities are subject to
the jurisdiction of the Federal Energy Regulatory Commission (“FERC”) in
Washington.73
To understand the role of PBR at the FERC, it is helpful to
understand some other important dimensions of FERC regulation. We review
here three of the main ones.
COSR TRADITION
The chief responsibility of the FERC and its predecessor agency, the
Federal Power Administration (“FPA”), has been to regulate the interstate
commerce of U.S. energy utilities.
Most U.S. energy utilities are investor
owned.74 The FERC and the FPA have together been engaged in the regulation
of investor-owned utilities (“IOUs”) for many years. Since, additionally, PBR is a
comparatively recent development, the FERC can draw on many years of COSR
experience in deciding how to regulate transmission today. Indeed, the FERC
and the FPA have over the years been involved in some of the classic court
decisions that shaped the development of COSR in the States.75
COMPETITION POLICY
Like other Washington regulatory agencies, the FERC has devoted a
great deal of time and effort in the last twenty years to the promotion of
competition in segments of its jurisdictional industries where competition is
73
74
75
The FERC also regulates U.S. bulk power markets and interstate oil and natural gas
transmission. It does not have jurisdiction over power or natural gas retail sales or
distribution.
Publicly owned utilities (e.g. the Bonneville Power Administration, Tennessee Valley
Authority, and the Western Area Power Administration) have been established chiefly to
produce and transport power from federally-controlled hydropower sites.
See, for example, the landmark Supreme Court decision, Federal Power Commission et. al.
v. Hope Natural Gas Co. 320 U.S. 591 (1943).
100
feasible. Its efforts to create a competitive interstate market for natural gas have
been quite successful. Beginning in the 1980s the Commission induced pipeline
companies, through a series of measures, to make unbundled gas transmission
services widely available to independent marketers and consumers.
The
previously dominant and largely non-competitive role of pipelines in the interstate
sale of gas was sharply scaled back. State commissions subsequently took the
companion measure of inducing local gas distribution companies (“LDCs”) to
make unbundled distribution services to large volume customers widely
available. The role of LDCs in retail gas sales to these customers was sharply
scaled back
Since, additionally, the North American gas production industry is
competitively structured, the result has been the development of a national
wholesale market for gas that is one of the most efficient energy commodity
markets in the world.
In 1992, the U.S. Congress passed the National Energy Act in 1992. Title
VII of the Act required open access to the interstate power transmission system.
The FERC, as the chief regulator of U.S. power transmission, was encharged
with implementing these provisions.
The FERC has encountered many challenges in its efforts to promote bulk
power market competition. Most retail sales of power in the United States have
traditionally been provided by investor-owned utilities engaged in generation,
transmission, and distribution. The typical utility generated most of the power
that it sold and located most of its generating plants on its own transmission
system. However, IOUs did make some bulk power sales to other IOUs and to
municipal and cooperatively owned utilities.
A continuation of vertical integration was an obstacle to the development
of competitive bulk power markets. Utilities might be loath to purchase power
from independent suppliers, thereby reducing the size of the market.
Transmission owners (TOs) would sometimes be incented, absent policy
restrictions, to deviate from sound transmission operating practices in pursuit of
power market goals. For example, a TO might offer transmission services to
101
competing suppliers which, when compared to the terms on which it used its
transmission system, involved higher charges and/or inferior quality.
Structural measures are available to remedy this problem. For example,
utility companies can be induced by some combination of “carrots” and “sticks” to
sell or spin off either their generating facilities or their power delivery facilities.
Alternatively, they can be induced to transfer their transmission assets to
companies with bylaws that make them relatively passive owners76 . The result
of such policies would be specialized power transportation utilities that are
independent of market participants77.
Companies of this kind are called,
variously, “transcos” and “independent transmission companies” (“ITCs”).
This approach to power industry organization has, for the most part, not
been pursued in United States. The fact that most transmission services were
provided by IOUs rather than government enterprises is largely responsible.
Alternative arrangements were pursued that do not require a forced restructuring.
For example, the federal government’s Public Utilities Resource Procurement Act
encouraged vertically integrated utilities to purchase more power from
independent producers. The FERC has required vertically integrated utilities to
offer unbundled power transmission services. Structural separations could, in
principle, be urged by some of the fifty state governments or the District of
Columbia. It is these governments and not the federal government that decide
whether to implement retail power market competition. More than half of the fifty
states have, in fact, chosen not to pursue retail competition and none of these
states has encouraged a complete structural separation of generation and
transmission.78
Most states that have implemented retail competition have not
induced such separations either. In these states, which include Connecticut,
76
77
78
Entergy was an earlier advocate of this approach. See their April 1999 petition for a
declaratory order in Docket No. EL99-57-000.
We use the general term “transportation” since these companies could, in principle, provide
distribution as well as transmission services.
These are, for the most part, states in which the regulated price of power was not markedly
higher than prices in bulk power markets so that the benefits of restructuring seemed
insubstantial. Vermont and Wisconsin have encouraged the establishment of specialized
transmission utilities but these are owned by utilities involved in generation.
102
Illinois, Ohio, Maryland, New Hampshire, New Jersey, Pennsylvania, and Texas,
utilities have typically transferred their generating plants to unregulated affiliates.
Only two transcos have yet been established and both of these operate only in
the state of Michigan.
It follows from this history that most owners of United States transmission
facilities are still extensively involved in power generation. An analogue to this
situation in the gas transmission industry would be for a pipeline company to be
extensively engaged in the production and sale of natural gas and for its
shipments of such gas to account for a sizable share of all supplies that are
moved on its system.
Under these circumstances, the FERC has encouraged (but not required)
utilities that are unwilling to pursue structural separation to place their
transmission systems under the control of independent operating organizations.
This involves some form of long term contract, which in the States is called a
transmission operating agreement. Various approaches to the government of
such operating organizations are possible.
Independent system operators
(“ISOs”) are one option. These are characteristically run by non-profit boards
that represent the interests of a range of parties in addition to TOs. For-profit
operators are another possibility.
Independence is by no means the only complication that the FERC has
encountered in its efforts to develop competitive power markets. For one thing,
the service territories of most United States utility companies have traditionally
been limited to portions of a single state.79 While some consolidation of the
industry has occurred in recent years (e.g. the merger of American Electric
Power and Central & Southwest), ownership of the United States power
transmission grid is still highly balkanized. Long distance shipments of power
can then potentially involve serious coordination problems for affected utilities. It
can also involve sizable transaction costs for shippers.
79
For example, long-
Notable exceptions have included the multi-state transmission systems of Allegheny Power,
American Electric Power, Central and Southwest, Northeast Utilities, Pacificorp, and the
Southern Company.
103
distance shippers in the early days of open access sometimes paid charges to
multiple TOs along the contract path, a phenomenon called rate “pancaking”.
The balkanization of service territories has also meant that the United
States transmission system is not designed to support large volume, long
distance power flows. The capacity to deliver power between regions is in some
cases limited, and this can accentuate regional bulk power price disparities. Bulk
power prices can be especially high in certain load “pockets” in which the power
generation industry is not competitively structured and the ability of power trade
to provide price relief is limited.
Still another complication in restructuring is the loss of scope economies
that were once enjoyed from the vertical integration of generation and
transmission.
For example, vertical integration encourages the sitting of
generating plants and transmission lines where they can reduce the combined
cost of generation and transmission.
Once transmission and generation
functions are separated, pricing policies must be fashioned that encourage the
rational sitting of new generating plants and incent transmission owners to make
needed investments.
The FERC has issued a number of orders in an effort to address these
challenges. In a 1996 decision, Order No. 88880, the Commission found that
some utilities had engaged in unduly discriminatory and anticompetitive practices
in their provision of transmission services.
It required utilities to file non-
discriminatory open access transmission tariffs and to take transmission service
under the same tariff of general applicability as did others.
FERC Order 200081, issued in 1999, took the further step of encouraging
utilities to place their transmission assets under the control of regional
transmission organizations (“RTOs”) that would be independent of power market
participants. These organizations would be responsible for both the day to day
operation and the longer term investment decisions of the transmission systems
80
FERC Promoting Wholesale Competition Through Open Access Non-discriminatory
Transmission Services by Public Utilities and Recovery of Stranded Costs by Public Utilities
and Transmitting Utilities, Order No. 888 61 FR 21,540 (May 10, 1996).
104
in their region. They would establish the terms of transmission service and be
the point of contact for the utility and other shippers.
The Order offered some guidelines for RTO development. Of these, the
following are especially germane to the development of transmission PBR.
•
The FERC did not mandate a particular approach to RTO organization. In
fact, it explicitly indicated an openness to a range of organizational
structures that included ISOs, transcos, and “hybrid” structures in which
one or more transcos operated under the direction of ISOs.
•
A
preference
was
stated
for
“market mechanisms” rather than
administrative curtailment procedures such as transmission loading relief
to ease transmission system congestion. While not requiring a specific
approach, the FERC noted that two approaches, locational marginal
pricing (“LMP”) and tradable firm transmission rights (“FTRs”), appeared to
be sound. LMP is an indirect approach to the management of congestion
through pricing which involves power markets organized by the system
operator.
Prices may vary at different points on the system to reflect
system transfer constraints.
A number of RTO proposals were fielded in compliance with Order 2000.
Proposals were eventually approved for the Midwest Independent Transmission
System Operator (MISO) in 2001, the PJM Interconnection in 2002, ISO New
England in 2004, and the Southwest Power Pool in 2005.
organizations have an ISO structure.82
All of these
ISOs are also operating with FERC
blessing but without RTO designation in California, New York, and Texas.
While ISOs are ubiquitous, the Commission has retained some interest in
for profit system operators. In a 1999 declaratory order, it indicated an openness
to the passive ownership approach to Transco organization.
In 2001, it
conditionally approved the establishment of GridSouth Transco, a for-profit entity
81
82
Regional Transmission Organizations (RM99-2-00) (December 1999)
The FERC in fact chose the MISO proposal for the Midwest region over that of a competing
transco proposal by the Alliance Companies.
105
that would operate the transmission systems of TOs in the Carolinas.83 In 2002,
the FERC conditionally approved the establishment of Grid Florida, a for-profit
entity that would operate the transmission systems of some Florida TOs and
acquire and operate the systems of others in exchange for a passive ownership
interest. All of these initiatives sputtered, however, after the general idea of an
RTO performing the functions expected by the FERC (including the management
of power markets) proved unpopular in the southeast.
The FERC has also sanctioned the establishment within ISOs of for-profit
ITCs that lack the scale and scope to be RTOs. Such companies have been
allowed to perform certain functions that the RTO would otherwise perform. The
general concept of such “hybrid” organizational structures involving subordinate
ITCs has been approved by the Commission in decisions concerning the MISO84,
SeTrans, and ISO New England85.
Two subordinate ITCs, Michigan Electric
Transmission Company (“Michigan Transco”) and International Transmission, are
now operating. The FERC has, additionally, given its conditional approval to
other subordinate ITCs that have either disbanded or have not yet commenced
operation.
These include the Alliance Gridco, Grid America, Illinois Electric
Transmission Company, TRANSlink, and TransConnect.
Since the FERC has not mandated RTO participation, the transmission
systems in several areas of the United States are still operated by TOs subject to
the guidelines in FERC Order 888. The affected regions are the southeastern,
south central, southwestern, Rocky Mountain, and northwestern states, as well
as Alaska and Hawaii. Virtually all of the TOs in these states have continued
their past involvement in generation.
MARKETING FLEXIBILITY
The FERC has displayed a longstanding interest in granting its
jurisdictional utilities a measure of marketing flexibility. This reflects, in part, the
83
84
85
The FERC conditionally approved a similar arrangement for SeTrans Regional Transmission
Organization in 1992.
90 FERC 61,192 (2000).
106 FERC 61,032 (2004)
106
fact that many of its jurisdictional utilities serve markets with diverse competitive
pressures and seek the production economies that are possible from serving
diverse markets from a common set of assets.
•
Market-based rates are permitted for transmission services where the
existence of competitive market pressures can be demonstrated.
•
The negotiated/resource rates program allows jurisdictional carriers to
offer negotiated rates on market-determined terms so long as customers
have recourse to a standard rate that has been deemed just and
reasonable.
•
Secondary markets have been encouraged for firm transmission capacity.
Under this approach, pricing flexibility is achieved by trade between
transmission system users. For example, LDCs sell their surplus rights to
use pipelines during the summer months to other gas marketers at a
considerable discount.
INCREMENTAL PRICING
Note, finally, that the FERC has for many years been a practitioner of
incremental pricing for new transmission facilities.
Under this ratemaking
approach, the cost of new facilities may be recovered from the users of those
facilities and not “rolled in” to a common cost of service that is recovered from all
customers.
9.1.2 PBR at the FERC
Washington D.C. has over the years established a reputation as one of
the world’s leading centers of PBR innovation.
The Interstate Commerce
Commission (d/b/a the Surface Transportation Board), and the Federal
Communications Commission were amongst the first regulators in the world to
implement large scale PBR plans. These two agencies also pioneered the North
American approach to rate indexing. This approach involves, as we have seen,
research on industry input price and productivity trends.
107
EARLY FERC DECISIONS
The FERC and its predecessor agency, the Federal Power Commission,
have experimented with many alternatives to COSR over the years. An early
and large-scale experiment was the use of “area rates” to regulate the wellhead
prices for natural gas sold in interstate commerce. The FERC was driven to
experiment with such expedients after it was ordered by the Courts to regulate
the prices of tens of thousands of gas wells.86
Around 1990, the Commission began a more systematic review of
“incentive rate” options for utility services.
The ability of PBR to facilitate
marketing flexibility played a prominent role in this review. For example, the
Office of Economic Policy (OEP) released a report on incentive ratemaking for
gas transmission in 1989. The Staff acknowledged three basic goals of PBR:
reductions in pipeline cost, reduced administrative burdens, and improved
pipeline pricing and services. With regard to the third goal, staff commented that
pipelines presently have little incentive or ability to design demand-responsive
rates, market their services aggressively, or seek innovative ways to improve
service and minimize cost. An illustrative price cap plan was detailed in the
paper.
This plan would allow pipelines to negotiate alternative rates with
customers and offer new services without Commission approval so long as the
pipeline also offered certificated services at the indexed, FERC-approved rates.
Despite identifying certain advantages of PBR, staff stressed that incentive
regulation should be voluntary since “there is no consensus on a preferred
approach, nor a general agreement on how successful the various programs
have been in achieving efficiency.
In March of 1992 the FERC launched a generic hearing on PBR.87 A final
policy statement was issued in October 1992.88 The Commission maintained
that it “is not required to follow any specific type of ratemaking formula and is not
86
87
88
Federal legislation ultimately resulted in the decontrol of these prices.
FERC, Notice of Proposed Rulemaking on Incentive Regulation, Docket No. PL92-1-000
(March 13, 1992).
FERC, Incentive Ratemaking for Interstate Natural Gas Pipelines, Oil Pipelines, and Electric
Utilities, Decision No. PL92-1-000 (October 30, 1992).
108
limited to designing rates for the utilities it regulates based on traditional cost of
service ratemaking”.
The appraisal of PBR was generally positive. The FERC
commented that “incentive ratemaking is an alternative regulatory mechanism
that can reward utilities for efficiency and benefit customers with lower rates”. It
emphasized the multi-dimensional character of efficiency, stating that
Utilities should operate at optimal levels, allocate services efficiently,
invest only when economically justified, and capture expanding
markets…The Commission remains committed to adopting regulation that
promotes efficient use of existing facilities, efficient investment choices,
and aggressive marketing.
Perhaps the most striking feature of the 1992 document was the tough
standards that the FERC enunciated for PBR plan approval. Most importantly, it
required a quantification of the benefits of PBR relative to continued COSR. This
standard greatly complicated PBR filings, not least of all because it seemed to
require the presentation of cost of service evidence.
In 1995, the FERC approved a settlement agreement that established a
multi-year rate indexing plan for the current customers of Transwestern Pipeline
Company.89 This settlement avoided prolonged litigation that would have arisen
if Transwestern had filed for a general rate increase to reallocate costs
attributable to the relinquishment of a large block of capacity by a major
Transwestern customer, Southern California Gas. This is a good example of the
use of PBR to avoid an awkward application of COSR.
In 1995, the FERC approved five-year price cap plans for jurisdictional oil
product pipelines under the mandate of the Energy Policy Act of 1992. These
plans featured marketing flexibility and a price cap index.
The plans were
updated in the year 2000.
1995 was also the year in which the FERC began a generic proceeding on
alternatives to traditional cost of service ratemaking for natural gas pipelines. A
89
72 FERC 61,985 (1995).
109
policy statement was issued in this proceeding in 1996.90 The most noteworthy
development on the PBR front was the replacement of the standard that plan
benefits be quantified with the more workable standard that the benefits of PBR
be shared.
Apart from the oil pipeline plans which, as noted above, were mandated
by law, formal PBR proposals to the FERC were rare before Order 2000. A few
plans were proposed by gas pipeline companies (including Florida Gas
Transmission91, Viking Gas Transmission, and Northern Natural Gas) and
rejected by the FERC. A plan featuring an index-based cap on allowed O&M
expenses was approved for El Paso Natural Gas in 1998.
Several reasons may be ventured for the paucity of FERC-approved PBR
plans during this period. One is that this was a period of generally slow input
price growth. This situation helped companies to stay out of rate cases without
recourse to formal PBR. A related circumstance is that incremental pricing often
separated the regulation of older assets and capacity additions. Slow growth in
the rate base for older assets also reduced the need for rate cases. Difficult
evidentiary requirements for FERC PBR filings in the early 1990s have already
been noted. The marketing flexibility provisions of FERC policy reduced the
need for PBR. These provisions included, as noted above, negotiated rates,
market based rates, and secondary markets for transmission capacity.
ORDER 2000
Order 2000 devoted considerable attention to the issue of “innovative
ratemaking” practices”.
The term “innovative ratemaking” was intended to
encompass PBR and various of other ratemaking reforms that include “innovative
pricing”. We discuss PBR and innovative pricing in turn.
Regarding PBR, the Commission concluded that it can provide “significant
benefits” over COSR and encouraged its consideration by RTOs. It expressed a
90
91
FERC, Statement of Policy and Request for Comments, Docket No. RM95-6-000, RM96-7000 (January 31, 1996).
For FGT see Docket Nos. RP91-197 and RP95-103. For Northern Natural see Docket No.
RP00-152-000, 90 FERC 61,064.
110
willingness to consider various PBR plan design features.
Approaches
discussed in the NOPR include rate moratoriums and other kinds of rate caps,
revenue caps, and performance standards (benchmarks).
Despite the
advantages, the Commission stated at 541 that “there is almost no support for
making PBR mandatory, and we therefore will not require RTO filings to include
PBR proposals, although we encourage such proposals”.
The Commission
traced at 544-546 some principles for the design of PBR plan. Plans should be
comprehensive, symmetrical, share benefits with customers, and should not
compromise reliability.
With regard to who is eligible to operate under PBR, the Commission
stated at 506 that innovative rate treatments should be made by RTOs.
The
Commission further stated at 542 that
in the context of an ISO or a tiered ISO/Transco that has been described
by some commentators, the activities that contribute to performance may
be shared between the RTO and the transmission owners… the RTO
design would simply ensure that the rewards and penalties associated
with the activities performed by transmission owners would flow through to
the owners to achieve the desired results.
In addition to PBR measures, the FERC indicated in Order 2000 a
willingness to consider certain innovative pricing provisions for TOs that are
members of approved RTOs. The provisions sanctioned include: “formula rates”
(which decouple a TO’s earnings from its own equity valuation); the continued
use of (presumably favorable) older rates or authorized rates of return; “levelized
rates” (which recover all capital costs through a uniform payment over the life of
an asset); and accelerated depreciation on new investments.
By way of rationale for these provisions, the Commission stated a desire
to remove the “pricing disincentives” to RTO participation. Prominent among
these disincentives are the risks involved in ceding control over transmission
assets (e.g. planning and expansion decisions) to an RTO. The Commission
also acknowledged miscellaneous risks that are encountered in a world of
unbundled transmission. It is not clear in the Order why the consideration of
111
these risks would not be a normal part of ongoing COSR. The possibility is thus
raised that the FERC was threatening to withhold a fair ROE from firms that did
not participate in RTOs.
RECENT DEVELOPMENTS: PBR
PBR ideas were discussed in the Order 2000 compliance filings of several
TO groups. Most of these proposals were non-specific. In the year 2000, the
FERC conditionally approved a PBR plan for International Transmission.92 This
plan involved a four-year rate moratorium.93 However, delays in the spin-off of
International Transmission and in its participation in an RTO subsequently
delayed the start of the plan and significantly shortened its term.
In 2001 the FERC issued three decisions that clarified the nature of
acceptable PBR proposals for TOs. In a decision involving Southern Company
Services94 it found that PBR incentives are acceptable that
motivate the grid operator to perform in response to the market and to
improve grid operation… In other words, we would accept those incentives
that are properly configured in that they reward the grid operator and
decisionmaker for improved grid performance.
Incentives are not acceptable that
would flow to the transmission owners who, because they are proposed to
be passive owners of the RTO, do not make any incentives regarding grid
operation. Simply put, it is inappropriate to send a price signal to a
passive owner that cannot respond to a price signal.95
This decision was controversial inasmuch as PBR could in principle be used to
elicit better investment and cost management performances from passive TOs.
In fashioning this policy, the FERC may therefore have revealed a willingness to
use PBR as ”candy” to promote structural change in the transmission business.
92
93
94
95
92 FERC 61,276.
The company also requested some operating flexibility during the moratorium period.
Specifically, reserved the right to introduce new, innovative, and optional transmission
products and services on a pilot program basis and to pursue market-based transmission
projects.
94 FERC 61,271 (2001)
The Southern Company filing was, in any event, unacceptable on other grounds as well and
was rejected.
112
It also showed a certain nonchalance concerning the issue of transmission cost
containment.
In a 2001 decision involving RTO West96, the Commission addressed a
PBR proposal of TransConnect, a proposed subordinate Transco. It noted that
under Order 2000 the RTO, as the sole administrator of the transmission tariff for
the region, has the exclusive authority to file the rates for service under that tariff.
TOs are entitled only to make Section 205 filings with the FERC to recover the
costs that they incur under RTO operation. When a TO is independent of market
participants but is not the RTO, it can include in such revenue requirement filings
a request for PBR and other incentive-oriented rate recovery mechanisms.
However, such incentive provisions must reward or penalize the transmission
owners for actions that they control (e.g. incentives to reduce operating and
maintenance costs or incentives to expand the grid). The FERC later addressed
the PBR provisions of a TransConnect rate filing, accepting some and rejecting
others.97 However, TransConnect never became an operational utility.
In a 2001 order provisionally granting RTO status to PJM98, the FERC
rejected a PBR proposal by PJM TOs on the grounds that most of these
companies lacked the requisite independence characteristics and that TOs are
not, in any event, authorized to make rate filings under Order 2000. A request for
rehearing was denied.99
PBR ideas advanced by Entergy in its Order 2000 compliance filing were
not addressed by the FERC when it rejected the filing for other reasons in
2001.100 Entergy’s 2004 proposal to contract with an Independent Coordinator of
Transmission has no PBR content.
In 2002, the FERC conditionally approved a PBR proposal of Michigan
Electric Transmission Company (“METC”). The conditionally approved plan froze
the company‘s currently effective rates for approximately three years.
96
97
98
99
100
Avista Corporation et al, 95 FERC 61,114 (2001).
100 FERC 61,297 (2002).
PJM Interconnection, L.L.C. et al. 96 FERC 61,061 (2001)
101 FERC 61,345 (2002)..
96 FERC 61,062 (2001).
113
Additionally, the company was allowed to recover, on a deferred basis over five
years beginning at the end of the plan, the annual cost (depreciation and return
on investment) of any new transmission facilities incurred from Jan. 1, 20001
though December 31, 2005. METC had noted in its filing the need for substantial
capital spending.
The November 2005 notice of proposed rulemaking on incentive-based
rate treatments, discussed further below, contains a section on PBR. The FERC
states at 34 that
Because it is difficult to observe directly the level of effort a utility,
transmission company, ISO or RTO expends on cutting costs and
improving efficiency, performance-based regulation may provide a
valuable tool to motivate transmission entities to maintain and operate
their systems reliably and efficiently.
It notes that “common performance-based models” include price-cap
regulation, targeted incentives, and “benchmark incentives which establish
rewards based on the performance of a reference group performing similar
activities.” The characterization of this latter “model” as common is surprising
since incentive mechanisms of this kind are in fact fairly rare in PBR.
The
discussion of PBR is also noteworthy for not restating that PBR was reserved for
companies that are independent of market participants.
RECENT DEVELOPMENTS: INNOVATIVE PRICING
Progress on the innovative pricing front has been somewhat greater. In
2002, the FERC granted a 50 basis point ROE premium to TOs that participate in
the MISO.101 Its stated policy reason was “the level of operating independence
that the Midwest ISO provides”.
This decision was appealed by the Public
Service Commission of Kentucky to the U.S. Court of Appeals for the District of
Columbia. In 2005 the Court remanded the decision to the FERC for
reconsideration.102
101
102
100 FERC 61,292 (2002).
See U.S. Court of Appeals for the District of Columbia Circuit, Public Service Commission of
the Commonwealth of Kentucky v FERC, February 2005. The opinion was written by John
Roberts, who has since become Chief Justice of the U.S. Supreme Court.
114
In January of 2003, the FERC issued a notice of proposed policy
statement on innovative rate making.103
This paper has come to be known as
the transmission pricing policy statement.
It advanced for discussion the
following specific innovative pricing measures:
•
A generic 50 basis point adder to the allowed ROE of all transmission
facilities transferred to a FERC-approved RTO.
•
A lump-sum revenue requirement adder equal to an additional 150 basis
points for the transferred facilities of any utility that participates in an RTO
and meets the FERC’s independent ownership requirement.
•
A generic 100 basis point ROE adder for “investment in new transmission
facilities “which are found appropriate pursuant to an RTO planning
process”.
A deadline of 31 December 2004 was proposed to qualify for these
measures.
In February 2003, the FERC approved a 100 basis point adder to the
allowed ROE of International Transmission, as well as a favorable debt/equity
ratio. In November 2003, it approved the same adder to the allowed ROE of
METC and permitted a formulaic debt/equity ratio.104
In explaining these
decisions it stated that
the Commission has granted certain rate treatments to transmission
owners in consideration for the benefits achieved by establishing fully
independent ownership and operation of transmission.
In 2000, the FERC approved certain innovative pricing provisions for
American Transmission Company.105 The treatment of investment cost in COSR
was changed in two respects: the value of construction work in progress was
allowed to be included in the rate base (where it can earn a rate of return) and
pre-certification costs related to construction projects are now expensed. The
103
104
105
FERC, Proposed Pricing Policy for Efficient Operation and Expansion of the Power Grid,
Docket No. PL03-1-00, 102 FERC 61,032.
105 FERC 61,214 (2003).
107 FERC 61,117 (2004).
115
company was, additionally, permitted a formulaic 50/50 debt/equity capital
structure for purposes of calculating the allowed rate of return. The FERC notes
at 2 of its order that
ATC requested these modifications as alternative incentives to the ROE
basis point incentive adders outlined in the Commission’s Proposed
Pricing Policy Statement. ATC requested these alternative incentives to
facilitate the financing of approximately $2.3 to $2.8 billion in new
transmission facility construction over the next ten years.
The FERC also approved in 2005 a 50 basis point incentive adder to the
ROE component recovered in RTO-New England’s rates for regional network
service. In the same order it accepted subject to suspension, hearing, and the
application of any future and more definitive pricing policy statement a proposed
100 basis point adder for new transmission investment. In a follow up order106
the FERC clarified the kinds of new investments that would be eligible for this
adder.107 It also addressed the reasonableness of granting ROE premia, stating
at 67 that
A return on equity is not susceptible to precise calculation. It is based,
rather, on a range of reasonable returns, which take into account a
number of factors that may be both cost-related and policy related,
including business risk factors. In this context, it is appropriate for the
Commission to adjust the allowed return for Transmission Owners that
undertake commitments designed to enhance the overall competitiveness
and efficiency of the wholesale markets, so long as the resulting rate of
return is within the range of reasonable returns.
In June 2005, the Commission issued an order on remand concerning the
50-basis point adder to the allowed ROEs of TOs that participate in the Midwest
ISO. It vacated its prior decision to offer the adder, while observing that the ISO
106
107
109 FERC 61,147 (2004).
The Commission states at 66 that the adder would pertain to investments “that, among other
things: (1) are approved through the RTEP process; (ii) are capable of being installed
relatively quickly; (iii) include the use of improved materials that allow significant increases in
transfer capacity using existing rights-of-way and structures; (iv) utilize equipment that allows
greater control of energy flows, enabling greater use of existing facilities; (v) has
sophisticated monitoring and communications equipment that allows real-time rating of
transmission facilities, facilitating greater use of existing facilities; or (vi) is a new technology
or innovation that will increase regional transfer capacity.
116
or its TOs can make a filing to include an incentive adder. The Commission
stated at 3 that
we continue to believe that implementation of incentives to encourage
participation by transmission owners in a regional transmission
organization (RTO) such as the Midwest ISO is sound public policy.
In late June of 2005, the FERC issued a policy statement to clarify the passive
ownership structures for transmission utilities that could qualify them for ITC
status and thereby make them eligible for innovative rate treatments.
The
Commission stated that it would be willing to accept proposals from ITCs which
have market participants as passive minority equity holders.
This decision
indicated renewed interest in the Transco approach to industry organization.
Title XII of the Energy Policy Act of 2005, titled the Electricity
Modernization Act of 2005108, contains a section on transmission “rate reform”.
Section 219 of the Act gives the FERC one year to establish, by rule,
incentive-based (including performance-based) rate treatments for the
transmission of electric energy in interstate commerce by public utilities for
the purpose of benefiting consumers by ensuring reliability and reducing
the cost of delivered power by reducing transmission congestion.
The chief goal of the section seems to be the promotion
of transmission
investment. In Subsection (a), the promotion of investment, the provision of “a
return on equity that attracts new investment”, and the deployment of new
technologies (which presumably involve investment) are explicitly stated goals.
In subsection (c) it is stated that
the Commission shall, to the extent within its jurisdiction, provide for
incentives for each transmitting utility or electric utility that joins a
Transmission Organization.
The Act strengthens the ability of the FERC to offer innovative pricing benefits to
utilities that participate in RTOs and perhaps other kinds of “transmission
organizations”.
It may also push the FERC to change its policy and allow
passive TOs to operate under PBR.
117
On November 18, the FERC issued a notice of proposed rulemaking
entitled Promoting Transmission Investment Through Pricing Reform.109
It
proposes to amend its regulations concerning incentive-based rate treatments for
power transmission. A key feature of the NOPR is an expressed willingness to
reconsider the ratemaking treatment of new transmission investments by all
utilities in light of the restructuring of the industry.
This is an important
clarification in policy. Measures under consideration include
•
An ROE “sufficient to attract new investment”
•
Inclusion of prudently incurred CWIP in rate base
•
Expensing of pre-commercial operations costs
•
Reduced risk of stranded transmission cost
•
Accelerated depreciation
•
Deferred recovery of new facility costs for utilities subject to retail rate
moratoria
The FERC, additionally, proposed to offer an ROE premium for ITCs, for TOs
that participate in RTOs and other ISOs, and for investments in advanced
technologies.
RECENT DEVELOPMENTS: SERVICE QUALITY
The regulation of transmission service quality received increasing
attention in the States following a regional transmission outage in 2003 which,
apparently, originated in the system of a MISO utility.
The FERC has not
traditionally played a direct role in transmission service quality regulation.
Oversight has instead been exercised chiefly by the North American Electric
Reliability Council (“NERC”), a voluntary association that includes most U.S.
electric utilities.
The NERC established Operating Policies and Planning
Standards that provide voluntary guidelines for operating and planning the
transmission system.
108
109
In 2005 the NERC adopted a comprehensive set of
H.R.6, Title XII, 109th Congress (2005)
113 FERC 61,182 (November 2005).
118
measurable reliability standards. However, these standards are also voluntary
and are not coupled with mandatory enforcement penalties.
Section 1211 of the EPAct contains amendments to the Federal Power
Act that address reliability regulation. The most critical provisions are as follows.
•
The FERC is authorized to certify and regulate an Electric Reliability
Organization (“ERO”) “the purpose of which is to establish and enforce
reliability standards for its bulk power system”.
•
The term reliability standards “includes requirements for the operation of
the existing bulk power system facilities…and the design of planned
additions or modifications to such facilities to the extent necessary to
provide for reliable operation.” This language suggests that regulation
may focus on quality management practices as much or more than quality
management outcomes.
•
The Act defines the term ‘reliable operation’ as “operating the elements of
the bulk power system within equipment and electric system thermal,
voltage, and stability limits so that instability, uncontrolled separation, or
cascading failures of such system will not occur.” This language suggests
that the ERO will focus primarily and even exclusively on traditional
reliability concerns and not on other dimensions of transmission service
quality, such as congestion management and the promptness of
responses to service requests.
•
The ERO is authorized to “impose…a penalty on a user or owner or
operator of the bulk power system for a violation of a reliability standard
approved by the Commission”.
•
Reliability organizations established under the terms of the Act must be
independent of the users and owners and operators of the bulk power
system.
However,
they
cannot
be
departments,
agencies,
or
instrumentalities of the federal government. This language prevents the
FERC and RTOs from performing most duties of the ERO and, indeed,
makes RTO activities a subject for ERO scrutiny. The Act also seems, to
119
encourage the assignment of the job to an organization that evolves from
the present NERC.
•
The FERC is, however, given considerable power over the ERO including,
in addition to the power of certification, the power to approve or reject
reliability standards and penalties and to issue its own penalties.
•
The regulatory process provided for under the act is cumbersome.
Consider, for example, the guidelines for penalties. The Act states that
the ERO can make a penalty finding “after notice and an opportunity for a
hearing.” Such penalties are subject to review by the Commission on its
own motion or upon application by the party that is subject to the penalty.
A Commission review requires “notice and opportunity for a hearing
(which hearing may consist solely of the record before the ERO and
opportunity for the presentation of supporting reasons to affirm, modify, or
set aside a penalty)”.
The Act also provides for the establishment of
regional advisory bodies to council the ERO.
The FERC issued a set of proposed rules concerning the certification and
regulation of the ERO in a September 2005 notice.110 The proposed rules are,
for the most part, a straightforward implementation of the guidelines provided in
the act. The envisioned regulatory system is even more cumbersome than that
detailed in the Act. The Commission states at 19, for instance, that it “generally
anticipates that it will provide notice and opportunity for hearing of any proposed
Reliability Standard or a modification to a Reliability Standard”.
The
Commission also proposes to “by order affirm, set aside, or modify” each and
every penalty.
110
112 FERC 61,239 (2005).
120
The Commission provides useful guidance as to the nature of penalties,
stating that
Any penalty imposed for a violation of a Reliability Standard shall bear a
reasonable relation to the seriousness of the violation and shall take into
consideration efforts of such user, owner, or operator of the bulk power
system to remedy violation in a timely manner. The imposition of
penalties is not limited to monetary penalties and may include, but is not
limited to, limitations on activities, functions, operations, and other
appropriate sanctions, including the establishment of a reliability watch list.
CONCLUSION
We conclude from this review of FERC policy and its relevance for
HQ TransÉnergie that there has to date been essentially no use PBR in the
regulation of United States power transmission. This outcome is the result of a
complex set of factors that includes the following.
•
Most TOs in the States are not independent of market participants. Given
the FERC’s policy over several years to use PBR as a reward for
independence, this has meant that the vast majority of TOs did not
qualified to operate under PBR.
•
Independent organizations such as ISO New England now control the
transmission systems serving most of the U.S. economy and these
organizations are eligible for PBR. However, all of these entities are of
non-profit character, are not owners of transmission systems, and could
experience substantial operating risk under PBR, and have not proposed
to operate under PBR.
•
The FERC has devoted immense time and effort to the development and
monitoring of organized power markets which could otherwise have gone
to the refinement of TO regulation.
Malfunction of an ISO-managed
market in California was a particular distraction.
•
In the case of transmission services for the retail loads of vertically
integrated utilities the FERC share jurisdiction with state regulators. It is
not clear what happens if the FERC opts for PBR but state regulators
prefer a continuation of COSR.
121
•
A major issue before the FERC is who should pay for new investments
that are needed to promote long distance trade but are not currently
needed for native load service.
It is not yet clear how existing PBR
mechanisms would be adapted to deal with this situation.
•
The ability of PBR to facilitate marketing flexibility is a potential advantage
in power transmission regulation. However, the FERC has developed a
pro forma tariff for unbundled power transmission that does a rough and
ready job of offering low rates for the chief price elastic use: point to point
services. Furthermore, the Commission has encouraged approaches to
congestion management (e.g. LMP) and transmission ratemaking (e.g.
incremental pricing and secondary capacity markets) that do not dovetail
easily with the traditional price cap approach to marketing flexibility and
are to some degree substitutes for it.
•
The FERC is clearly preoccupied right now with the stimulation of
transmission system investment. It has been slow to address how the
ROE needed to attract investment has been changed by its restructuring
initiative.
Attention to the establishment of an appropriate risk return
balance that calls forth needed investment is understandable. PBR is not
a remedy for an underinvestment problem given the strong incentives that
it provides for cost containment; the fact that PBR generally increases
operating risk; and the fact that “bunched” investments are difficult to
accommodate under the North American approach to index-based PBR.
•
United States transmission operators are in most cases IOUs that have
long operated under COSR.
Many and perhaps most have not been
involved in a rate case for transmission assets for many years. Recall
also, that O&M expenses account for only a small share of the cost of
power transmission.
These conditions reduce the potential for rapid
performance gains from transmission PBR that might exist with a state
enterprise or a recently privatized utility.
122
It is also important to note that the rate of return premia and other
inducements that the FERC discusses under the heading of incentive ratemaking
are not a customary form of PBR and are not applicable to HQ TransÉnergie.
HQ TransÉnergie is already an ITC, is not controlled by an independent system
operator, and does not face an unusual risk of cost recovery. PBR is generally
about improved performance that enables utilities to offer lower rates than would
be possible under COSR.
9.2 Canada
9.2.1 Jurisdiction
Regulation of the Canadian power industry occurs chiefly at the provincial
level.
The National Energy Board plays a much smaller role in power
transmission regulation than its U.S. counterpart. Important differences exist in
provincial approaches to regulation.
Power transmission service in Canada is provided chiefly by provinciallyowned utilities. Most or all of the provinces with these utilities have moved to
make them operate more like IOUs.
These utilities have been subject to
increasingly close oversight by provincial regulators. IOUs predominate in the
electric utility industries of Alberta, Nova Scotia, and Prince Edward Island and
continue to own most transmission facilities in these provinces.
9.2.2 Industry Structure
Policymakers in most Canadian provinces have in recent years required
utilities to offer unbundled transportation services.
These efforts have been
motivated in part by a desire to promote power market competition and in part by
a desire to facilitate exports to the United States by conforming to FERC
guidelines laid forth in Order 888 and other decisions. Prior to these unbundling
initiatives, all of the large transmission providers were also extensively involved
in power generation. This has inevitably lead to concerns about independence
123
when transmission services were unbundled.
different solutions to this challenge.
The provinces have pursued
The context in Québec is presented in
Section 11.
ALBERTA
Transmission services in Alberta were for many years provided by a trio of
vertically integrated IOUs.
After a provincial initiative to promote wholesale
market competition, all three of these companies continued to own and operate
generating plants in the province. However, one company spun off its sizable
transmission system and this became an independent transmission utility
(AltaLink). Another company established a specialized transmission subsidiary
(EPCOR Transmission). The province established a power pool and engaged a
for-profit entity (ESBI) for several years to be Alberta’s “transmission
administrator”.
Both functions have since been provided by an independent
entity, the Alberta Electric System Operator. The TOs continue to perform many
O&M functions.
BRITISH COLUMBIA
In British Columbia, transmission service was for many years provided
chiefly by a vertically integrated government-owned corporation, BC Hydro. A
separate corporation, BC Power Transmission, was established recently to
operate BC Hydro’s facilities. It owns control centers and certain other system
operation assets. However, BC Hydro still owns the grid and also provides many
O&M services to BCTC under contract.
MANITOBA
In Manitoba, generation and transmission service has for many years
been provided by a vertically integrated government-owned corporation,
Manitoba Hydro. An unbundled transmission tariff was first established in 1997.
The transmission function is separated from other business functions and bills
the other business units for the use of the transmission facilities.
124
NEW BRUNSWICK
In New Brunswick, transmission services were for many years provided by
a vertically integrated utility, New Brunswick Power. An OATT was approved for
the company in 2003. The transmission assets have since been transferred to a
specialized subsidiary, New Brunswick Power Transmission.
This company
performs many O&M functions, but tariff design and implementation are now
undertaken by a new independent entity, the New Brunswick System Operator.
NEWFOUNDLAND AND LABRADOR
Power transmission service in Newfoundland and Labrador is provided
chiefly by two vertically integrated utilities: Newfoundland and Labrador Hydro, a
government-owned corporation, and Newfoundand Power, an IOU.
These
companies do not trade with the United States and power transmission is not
separately regulated.
Nova Scotia
In Nova Scotia, power transmission services are still provided by a
vertically integrated IOU, Nova Scotia Power (d/b/a NSPI).
approved for the company in 2005.
An OATT was
The company operates the grid and
administers the tariff.
ONTARIO
In Ontario a government-owned corporation, Ontario Hydro, for many
years provided most generation and transmission services in the province. The
province undertook a radical restructuring that placed power transmission
operations in a specialized power delivery utility, Hydro One Networks. Hydro
One is unaffiliated with any generating company but the province, which owns
100% of Hydro One, still owns extensive generation capacity in the province..
The Ontario power grid is now operated by the Independent Market Operator.
PRINCE EDWARD ISLAND
Transmission service on Prince Edward Island is provided by Maritime
Electric, a vertically integrated IOU.
125
Saskatchewan
In Saskatchewan, generation and transmission has for many years been
provided by SaskPower, a government-owned corporation. The Company now
provides power transmission services under an OATT. Functional independence
was promoted in 2001 by the establishment of a subsidiary company, North
Point, to perform certain generation, load management, and marketing services.
9.2.3 Regulatory System
ALBERTA
Transmission utilities are regulated in Alberta by the Energy Utilities Board
(“EUB”). This Board must approve the transmission facility owner (TFO) tariffs
that TOs file with the transmission administrator. These are essentially revenue
requirement applications and do not concern the rates charged to transmission
system users.
The EUB has experimented with PBR in decisions concerning ATCO Gas
and NOVA Gas Transmission.
However, it has generally followed a cost of
service approach to the regulation of the power industry, and uses COSR to
regulate the TFO applications. ETI expressed an interest in PBR in its first TFO
application, and began monitoring a set of performance measures.
An
Alberta
Transmission
Reliability
Committee
chaired
by
the
transmission administrator and comprised of industry stakeholders has
considered the development of a set of performance standards for the
transmission system in Alberta.
BRITISH COLUMBIA
Energy utility regulation in British Columbia is conducted by the
BC Utilities Commission.
This Commission has in the past regulated two
provincial utilities − Terasen Gas and West Kootenay Power using PBR.
BC Hydro has to date been reviewed using only conventional rate cases. BCTC
also operates under COSR and plans to continue doing so for the foreseeable
future. Since the rate base of the company is small relative to its operating cost,
126
PBR could involve extreme operating risk.
However, the Commission has
directed BCTC to begin monitoring a series of reliability and performance indices
and to report on these indices in its annual capital plan reports.
MANITOBA
Manitoba Hydro is regulated by the Manitoba Public Utilities Board. The
PUB uses a COSR approach to regulation.
However, Manitoba Hydro has
operated for extended periods without rate increases.
A rate case was
concluded most recently in 2004 that resulted in a 5% rate hike. The Company’s
OATT is based on the FERC pro-forma OATT.
NEW BRUNSWICK
The New Brunswick Board of Commissioner of Public Utilities has
traditionally used COSR to regulate electric utilities.
Its regulation of New
Brunswick Power Transmission does involve one PBR-style innovation, an ROE
range.
NEWFOUNDLAND AND LABRADOR
Newfoundland utilities are subject to the jurisdiction of the Newfoundland
and Labrador Board of Commissioners of Public Utilities. The Board uses a
largely traditional COSR approach to regulation.
Rate cases were recently
concluded for both of the transmission service providers. One innovation with a
PBR flavor was the use of an annual adjustment formula for the rate of return
which uses bond yields in Canadian capital markets..
NOVA SCOTIA
The Nova Scotia Utility and Review Board regulates NSPI using COSR.
The Company’s most recent rate case for NSPI was filed in 2004 and completed
this year.
ONTARIO
The Ontario Energy Board has in the past experimented with index-based
PBR for its jurisdictional power and gas distribution utilities. Its interest in PBR
for power distribution has been spurred in part by its responsibility to regulate
more than 100 distributors. Consumers Gas and Union Gas have also been
127
regulated using PBR. Their interest in PBR is spurred in part by the fact that
these companies have in recent years experienced slow growth in volumes per
customer. This has slowed productivity growth and induced them to file frequent
rate cases. The OEB has indicated a desire to continue with PBR regulation of
gas distribution. The transmission operations of Hydro One have never been
subject to PBR.
PRINCE EDWARD ISLAND
Maritime Electric is regulated by the Prince Edward Island Regulatory
and Appeals Commission. The company operated for several years under a
plan that set rates at 110% of the equivalent rates that New Brunswick Power
charged for similar service in New Brunswick.
In December 2003, the
Government of Prince Edward Island passed legislation returning Maritime
Electric to traditional cost of service regulation.
SASKATCHEWAN
Rate proposals of SaskPower are reviewed by the Saskatchewan Rate
Review Panel and must ultimately be approved by the provincial cabinet.
A
COSR approach to regulation is used. The most recent rate case was held in
2004.
ANALYSIS
COSR is used almost exclusively in the regulation of power transmission
in Canada.
The analysis developed in this paper provides some persuasive
theories as to why Canadian regulators have not as yet made extensive use of
PBR.
•
Most provincial regulators have jurisdiction over only a few utilities. This
sharply reduces the potential regulatory cost savings from transmission
PBR.
•
Most transmission owners have been subject to COSR for some time, and
several have operated for extended periods in the past decade without
rate hikes.
Considering as well the small share of O&M expenses in
128
transmission cost there is not much likelihood that PBR would trigger large
short-run gains in operating efficiency.
•
Major investments are expected in the transmission systems of several
provinces (e.g. Alberta & BC) in the next few years. We have seen that
these are difficult to accommodate under some popular forms of PBR.
•
The transmission systems in Alberta, Ontario, and New Brunswick are
now operated by independent, non-profit entities.
This precludes the
traditional use of PBR to afford greater system operators greater
marketing flexibility.
Utilities in many other provinces lack the
independence to be permitted extensive marketing flexibility.
•
The data required for the calculation of historical industry productivity
trends are not readily available.
9.3 Australia
9.3.1 Industry Structure
Power transmission in Australia was for many years provided by vertically
integrated utilities with monopolies on service in a particular state.
In some
states (e.g. Victoria, South Australia, Tasmania, and Western Australia) this utility
provided generation, transmission, and distribution services. In others (e.g. New
South Wales and Queensland) it provided only generation and transmission
services, and distribution was carried out by other utilities.
All of the vertically integrated utilities were state enterprises in the mid1990s. This facilitated a radical restructuring of the Australian power industry in
that decade under the terms of the National Electricity Law. Today, each state
has several competing generation businesses, several power distributors, and a
single monopoly transmission business. The transcos and their respective states
are as follows:
•
New South Wales
−
Transgrid
•
Queensland
−
Powerlink Queensland
129
•
South Australia
−
ElectraNet
•
Tasmania
−
Transend
•
Victoria
−
SPI Powernet
Only two states (e.g. Victoria and South Australia) have elected to privatize their
transcos. This means that in several states, the government is a common owner
of generation and transmission facilities.
A public enterprise, the National
Electricity Market Management Company (NEMMCO), has been established to
operate the interstate transmission system and a power pool called the National
Electricity Market.
These restructuring measures have encouraged regional trade flows for
which the transmission system was not designed.
This has given rise to
substantial transmission system investments in recent years.
9.3.2 Transmission Regulation
Prior to restructuring, regulation of Australian utilities was fairly informal.
Future rates were often agreed to in meetings between senior utility managers
and government officials.
After restructuring, the transmission utilities were
regulated for several years by state regulators. They were then transferred to the
jurisdiction of the Australia Competition and Consumer Commission (ACCC).
They are now regulated by a unit of the ACCC known as the Australian Energy
Regulator.
Because the transmission system is operated by an ISO, the regulation of
transmission utilities pertains chiefly to their revenue requirement for capital
ownership and routine O&M.
This sidesteps the issue of how accelerated
regional power flows should affect the rate trajectory.
employed to revenue requirement regulation.
A PBR approach is
Rate plans typically have a
duration of five years. British-style revenue cap indexing is commonly employed.
This means that the revenue cap index for each utility is specific to expectations
regard its capital investments and other costs during the plan period.
130
Extensive new transmission system investments are underway in Australia
to support expanded long distance trade. The expected investments of different
transmission companies vary greatly.
Since indexing is British-style, the X
factors of the revenue cap indexes vary greatly.
Here are the most recent
values.
Company
X-factor
Indexing Formula
Approval Date
Electranet
0.00
CPI – 0.00
December 2002
Energy Australia
1.30
CPI - 1.30
January 2000
Powerlink Queensland
-6.70
CPI + 6.70
November 2001
SPI Power Net
-0.77
CPI + 0.77
December 2002
Transgrid
-1.30
CPI + 1.30
January 2000
Notice that three of the five X-factors are negative so that index growth exceeds CPI growth.
The following considerations help to explain the use of PBR to regulate
power transmission utilities in Australia.
•
Regulators lacked long experience with COSR, and so incurred no special
start up costs in adopting the form of regulation prevalent in Britain.
•
The British approach to PBR, in any event, has a solid cost of service
foundation and is especially well suited to accommodating large scale
capital investments.
•
All of the transmission utilities had in the recent past been state
enterprises, which raised hopes concerning short run performance gains.
131
10.
PBR FOR HQ TRANSÉNERGIE
10.1 Features of Québec’s Transmission Industry
10.1.1 HQ TransÉnergie
HQ TransÉnergie as noted earlier provides power transmission services in
Québec. It operates and maintains the system and also owns the assets. Since,
additionally, the province is vast and its economy sizable, HQ TransÉnergie is
one of the largest transmission providers in North America.
10.1.2 Importance of Transmission
Power transmission plays an important role in the economy of Québec.
The province has an enormous capacity for low-cost hydroelectric generation.
Policy measures have ensured that the cost savings from this capacity are
passed along to retail customers in the form of low rates. Many consumers rely
on power for their space heating needs, which are extensive in Québec’s climate.
Industries, such as aluminum production, with power intensive technologies are
important in the province.
Most hydroelectric resources in Québec are located at sites that are
remote from the main centers of provincial power consumption.
It is also
noteworthy that Québec has a mounting interest in power production from wind
energy and most of the promising sites are also remote. Transmission thus plays
a key role in making low-cost and environmentally friendly power available to
Québec consumers. The province, effectively, employs a transmission-intensive
power supply technology. One manifestation of this is that transmission service
currently accounts for about a quarter of the revenue requirement of
HQ Distribution.
132
10.1.3 Structural Considerations
Hydro-Québec is a public enterprise that provides most power generation,
transmission and distribution services in the province. Established in 1944, the
Company was converted by a 1981 act into a joint-stock business corporation,
with the provincial government as a 100% shareholder. HQ TransÉnergie was
established in 1997 as a division of Hydro-Québec which provides unbundled
transmission services.
creation
of
divisions
The restructuring initiative continued in 2001 with the
to
undertake
the
company’s
power
production
(HQ Production) and distribution (HQ Distribution). These two divisions are the
biggest customers of HQ TransÉnergie and account for almost all of its revenue.
HQ TransÉnergie provides transmission service to all of its other customers
under a non-discriminatory tariff. Furthermore, HQ TransÉnergie follows a Code
of conduct approved by the Régie governing transactions with affiliates.
10.1.4 Québec Regulation
Hydro-Québec when carrying on electric power transmission activities
through its division HQ TransÉnergie is subject to regulation by the Régie to the
extent provided by the Act respecting the Régie de l’énergie (the “Act”). Article
49 of the Act states that it must “favour measures or incentives to improve the
performance of the electric power carrier or a natural gas distributor and the
satisfaction of consumer needs”.
This language could be interpreted as
encouraging either or both of PBR and the use of internal benchmarks for
management purposes. The Régie has twice approved PBR plans for Québec’s
gas distributor, Société en commandite Gaz Métro. However, COSR has been
used to date to regulate HQ TransÉnergie.
The provincial government authorized the rates charged by the company
for many years. The first transmission rate case was filed at the Régie in 1998.
A final decision in this case was not reached until 2002. A new rate case was
filed in 2004. A decision in this case is not expected until 2006.
133
The approach to the design of HQ TransÉnergie’s Open Access
Transmission Tariff (OATT) is fairly common since it follows the FERC pro forma
tariff. HQ Distribution uses the tariffed native load service to secure delivery of
its power. HQ Production uses the tariffed point-to-point services to deliver the
power that it sells outside Québec. Point-to-point services are available on an
annual, monthly, weekly, daily, and hourly basis. The annual service is firm while
the hourly service is non-firm and the other services are available on both a firm
and non-firm basis.
The revenue requirement is divided amongst the transmission services on
the basis of their expected annual coincident peak demand during the test
period. System demand is winter peaking since HQ Distribution purchases most
power shipped on the system and has a winter peaking load. Most point to point
transactions are made in the summer, when there is excess generation and
transmission capacity and shipments can be made at the low hourly rates. Point
to point shipments thus account for only a small share of system revenue.
Several additional features of transmission ratemaking in Québec merit
note.
•
The native load revenue requirement is collected from HQ Distribution in
lump sum payments and not using a rate per unit of actual or expected
maximum system use. This is an important consideration in a period in
which HQ Distribution has accounted for a progressively larger share of
system use. HQ TransÉnergie must, effectively, file a new rate application
if it wishes to recover the revenue shortfall that results from changes in the
use of the transmission services.
•
According to a decision of the Régie, connection costs for new generation
or additions to the transmission system of up to CAN 522 $/kW are
included in the rate base.
Costs in excess of this must be paid by
customers (e.g. generators) who make connection requests. Customers
requesting connections to remote sites must therefore pay for a sizable
share of the incremental transmission system investment. This limits the
134
impact of system expansion on the unit cost of owning and operating rate
based assets.
•
The problem of congestion on the HQ TransÉnergie system is small in
comparison to the situation in the United States or Australia, where an
upturn in regional power trade over a system of balkanized ownership has
produced many bottlenecks. This reduces the need for complicated rate
designs to manage congestion.
•
Transmission tariffs must comply with the legal and regulatory framework
in Québec. Article 49 of the Act stipulates the requirement for uniform
rates throughout the territory served by the transmission system.
HQ
TransÉnergie currently has no authority to discount its rates or to offer
special contracts for the use of the system. Furthermore, the Régie has in
the past refused to grant HQ TransÉnergie full discretion to set discounts.
10.1.5 Québec’s Power Market
HQ Production has the legal obligation to supply up to 165 TWh of
heritage electricity yearly at a low fixed price to HQ Distribution. This “heritage
pool” at present roughly equals the needs of HQ Distribution. The Act directs HQ
Distribution to obtain its incremental power requirements via competitive bidding
at market-based rates. These requirements are expected to rise gradually in
coming years. HQ Production can participate in these bids or use its incremental
production capacity to sell power outside the province.
Policymakers have
intervened to influence the technologies that will be used to meet the incremental
power requirements of HQ Distribution. They have to date favored renewable
energy sources such as wind farms over thermal sources.
Québec is adjacent to Ontario in Canada, and to New York and New
England in the United States. Each of these regions has a sizable demand for
power and bulk power markets that are readily accessible to Québec producers.
The sizable MISO market in the Midwestern U.S. can be accessed through the
facilities of Hydro One.
135
Prices in these markets have generally been high in recent years. In
Ohio, for instance, dismay over bulk power prices has prompted regulators to
delay competitive bidding and to sanction the continuation of the high retail prices
charged by some utilities which prompted the restructuring of the state’s power
industry in the first place. Ontario is struggling to secure new power supplies that
are sufficient to permit the shuttering of several coal fired power plants. In the
future, concern about global warming may give hydropower an additional
advantage in these markets.
Under the twin circumstances of high prices for new power supplies and a
growing provincial need, the capacity to produce power in Québec is expected to
increase substantially in the next ten years. HQ Production is currently building
new hydro facilities such as Mercier, Eastmain, and Péribonka. Major additional
projects are in the planning stages in the Eastmain, Rupert, and Romaine
watersheds. The company also maintains an inventory of other projects that may
become economic under favorable conditions. In addition to the HQ Production
initiatives, independent power producers are planning major expansions of wind
generation capacity.
Note also that Québec has a natural advantage in having some areas of
relatively sparse population and/or slow economic growth that are located close
to some of the major power consumption centers of eastern North America.
State of the art environmentally friendly power plants constructed in such areas
could have real advantages over U.S. production from high-priced natural gas or
the depleting reserves of eastern low-sulfur coal or from high-sulfur Midwestern
coal. Such plants could materially stimulate local Québec economies since they
can involve extensive ongoing staffing in addition to sizable construction projects.
The power market situation just described has important implications for
HQ TransÉnergie. Most of these new generation projects are remotely sited and
will require new transmission plant construction.
Since HQ TransÉnergie,
additionally, needs to make extensive system refurbishments and investments to
serve growing demand, the investment plan of the company needs to be
136
ambitious in the next few years. In totality, the Hydro-Québec Strategic Plan
2004-2008
reports that investments of CAN 3.8 billion are anticipated. This
compares to a 2002 asset base of CAN 17.3 billion for HQ TransÉnergie.
The power market situation also poses some marketing challenges for
HQ TransÉnergie. For example, the economics of some potential generating
projects will be sensitive to charges for use of the common system. Long term
contracts for these terms of service could improve the likelihood that some of
these projects go forward.
10.2 Indicated Regulatory Strategy
In this section, we integrate our discussion of the Québec situation with
the analysis developed in previous sections to consider the proper role of PBR in
the regulation of HQ TransÉnergie. We do this by reviewing systematically the
conditions that determine whether PBR is advantageous.
10.2.1 Cost of Effective COSR
PBR is more advantageous to the extent that the cost of effective COSR is
unusually high. We review here each of the potential complications discussed in
Section 2.
NUMBER OF JURISDICTIONAL UTILITIES
The number of utilities under the jurisdiction of the Régie is quite small. Its
regulatory burden is, in fact, less than that of a typical state regulator in the
United States. Such regulators typically have jurisdiction over several telephone,
gas, and electric companies. We have seen that COSR is still the most common
form of energy utility regulation at the state level. This situation is markedly
different from that which induced the Ontario Energy Board to use PBR to
regulate power distributors.
INTRACTABLE REGULATORY ISSUES: COST
Regulation of the cost of power transmission is not especially
controversial. The construction of transmission facilities to remote sites could be
137
controversial if their costs were born by all system users. In Québec, however, a
sizable share of the costs of connecting remote sites is born by those who
benefit.
The cost of regulating transmission cost also depends on the growth trend
and volatility of input prices. Power transmission is, as we have seen, a capital
intensive business that makes only modest use of price volatile energy inputs.
Prices of capital and other transmission inputs have, to the best of my
knowledge, risen only gradually in Canada in recent years.
INTRACTABLE REGULATORY ISSUES: MARKETING
Potential difficulties in the design of rate and service offerings are, as we
have seen, one of the most common reasons to use PBR. In Section 10.2, we
noted that transmission system operators sometimes need a substantial amount
of marketing flexibility. In the case of HQ TransÉnergie, the need for marketing
flexibility is diminished by several circumstances.
•
The existing approach to rate design is fairly effective at meeting some of
the main marketing challenges. For example, costly uses of the system
are discouraged by the use of the annual coincident peak demand to
allocate the revenue requirement, as well as by the sizable extra charges
for remote connections.
Rates for point-to-point services and the unit
price of the native load service are reasonable under the market and
regulatory conditions of HQ TransÉnergie.
•
There are some potential new generation projects in Québec whose
economics would be sensitive to charges for use of the transmission
system. For example, a change in the rates for point-to-point services
could have major repercussions for point-to-point customers trading with
markets in neighboring regions. In this situation, PBR could be useful in
facilitating some marketing flexibility for HQ TransÉnergie. For example, it
could be authorized to enter into long term contracts governing the rates
for point-to-point service. The terms of such contracts could be subject to
a reasonable price floor.
On the other hand, the legal and regulatory
138
framework in Québec was noted above to discourage HQ TransÉnergie
from discriminating between potential new generators by adopting terms
and conditions for use of the system according to their specific
development needs. It is also noteworthy that transmission rates could be
fine-tuned under the current tariff design process.
•
Another challenge in the regulation of HQ TransÉnergie’s marketing is the
need, under the current regulatory system, for frequent reallocations of the
revenue requirement. As we have seen, this results from the fact that the
HQ Distribution bill does not adjust automatically with its use of the
system.
Rate design hearings raise the controversial issue of cost
allocation and can weaken performance incentives, especially to the
extent that the hearings also involve a reconsideration of the revenue
requirement as well as an examination of the expected demand.
Automatic adjustments to the native load bill would reduce the need for
frequent rate design hearings.
On the other hand, traditional COSR
remedies are also available to alleviate this situation. For example, the
native load bill could be converted from a lump sum payment to a rate per
unit of maximum demand, while treating as incentives the variations in the
revenue requirement.
This method has some benefits but its
implementation depends on the endorsement of stakeholders.
10.2.2 Cost of Effective PBR
The advantage of PBR is greater to the extent that the established
approaches are readily and effectively implemented.
We consider here the
extent to which established approaches to PBR can be readily and effectively
implemented for HQ TransÉnergie.
DATA AVAILABILITY
Data are not readily available for the accurate computation of the historical
productivity trend of the power transmission industry of eastern Canada or of
Canada as a whole. Indeed, the requisite data are not fully available even for
139
HQ TransÉnergie.
Utilities are not easily drawn to provide or share detailed
operating data and such data are not available in comparable formats.
Furthermore, the COPE data set of the Canadian Electricity Association , while
useful for benchmarking, is not in my opinion suitable for the calculation of longrun productivity trends. The required data are available for some utilities in the
United States. However, the suitability of these data has been complicated by
the ongoing restructuring of the industry. One of several important complications
is the many transfers of assets that have occurred between the transmission and
distribution categories.
RELEVANCE OF HISTORICAL PRODUCTIVITY TRENDS
The use of the historical productivity trends that are required in the North
American approach to indexing was noted above to be generally problematic in a
power transmission application.
In particular, it can be difficult to identify a
historical productivity trend that is relevant to a transmission utility’s particular
output and capital spending outlook.
ADJUSTMENTS FOR INPUT PRICE INFLATION
The appropriate compensation for input price inflation was shown in
Section 5 to be a subject of considerable controversy in many PBR proceedings.
The growth trend in the price of capital has been an area of particular dispute. In
a capital intensive business like power transmission, the importance of this issue
is amplified. The Ontario proceedings to develop price cap plans for Union gas
and provincial power distributors are good examples. If the Régie pursues an
index-based approach to PBR, it will find itself considering the right form of an
industry-specific input price index, including the right way to measure capital
prices.
On the other hand, HQ TransÉnergie does not have marked system
congestion problems or face price-volatile local power markets and this makes it
somewhat easier to design an inflation measure that effectively compensates the
company for changes in the unit cost of system operations.
140
DIFFICULTY OF IMPLEMENTING SERVICE QUALITY PROVISIONS
Service quality provisions were noted above to be a vitally important
component of transmission PBR. The implementation of comprehensive PBR or
PBR for O&M expenses could, by strengthening incentives for cost containment,
incent an undesirable decline in service quality if the service quality provisions of
the plan are improperly designed.
Important design issues include the
appropriate performance indicators, benchmarks, and penalty/award rates. Year
to year fluctuations in quality levels often reflect changes in weather and other
external business conditions rather than changes in quality effort.
Quality
benchmarks based on a company’s historic quality should therefore be based on
several (e.g. 5) years of data.
Quality benchmarks based on data for other
companies are problematic due primarily to differences in data collection
methods and in external business conditions such as forestation that influence
quality.
In the case of HQ TransÉnergie, work to develop appropriate quality
provisions still has some ways to go. The development of indicators covering the
main quality components has to be pursued allowing afterwards the gathering of
corresponding historical data. Once underway, such a program will improve the
ability of the Régie to monitor the efficiency of HQ TransÉnergie and reduce to
some extent the need to implement PBR. Research to develop appropriate
award-penalty mechanisms for transmission service quality is not well-advanced.
In the U.S.A. for example, consideration of appropriate mechanisms is just
beginning. Absent considerable additional work, it follows that it may be prudent
to delay the implementation of forms of PBR that greatly alter cost containment
incentives.
COST PERFORMANCE INDICATORS
The Régie has also expressed an interest in cost performance indicators. These
are commonly used in internal benchmarking but are not that widely used in
transmission PBR or in North American PBR generally.111
111
Benchmarking is, however, used as an input to rate cases, especially overseas.
Accurate
141
benchmarking of utility cost is generally challenging.
The operating scale,
services performed, and other business conditions facing utilities vary and their
effects on utility cost are complex and poorly understood. Pacific Economics
Group is a world leader in statistical cost benchmarking and has benchmarked
transmission cost and many other categories of energy utility cost. Based on this
experience, I can say that the state of the art in power transmission
benchmarking is well behind that for power distribution, gas distribution, or
bundled power service. Utility cost benchmarking in Canada is hindered by the
lack of standardized and publicly available data on utility operations. The main
current source of data, the COPE data set of the Canadian Electricity
Association, is not publicly available and the sample is not large. The simple
cost indicators discussed by the Régie do not have an accuracy sufficient to
provide incentives.
If the Régie expects to develop transmission cost
benchmarks that are suitable for incentives it should expect the process to take
several years.
PBR RISK
An important limitation of PBR is its tendency to increase utility operating
risk. The recovery of capital cost is a particular concern. HQ TransÉnergie is in
the midst of a program of accelerated investment that is occasioned chiefly by
the accelerated construction of new generation capacity in Québec and additions
to the transmission system to respond to growing demand.
In this context,
HQ TransÉnergie has an understandable interest in receiving payment for the
resultant cost promptly. Moreover, it is appropriate for the Régie to allow the
recovery of the revenue requirement under the statutes of the Act and
fundamental regulatory and economic principles COSR is well suited for ensuring
this cost recovery.
On the other hand, the current connection policies of the Régie have the
effect that a sizable portion of the potential new investments as well as parts of
the ongoing investments could be excluded from the rate base. It is possible that
PBR for the rate base could coincide with COSR for major capital investments,
142
an approach that has been pursued for other utilities in Canada (e.g. NOVA Gas
Transmission).
However, this could raise some legitimate concerns about
uneven performance incentives.
Namely, the company might seek ways to
reduce the cost of owning and operating rate based assets via larger
expenditures on major capital additions.
10.2.3 Prospects for Performance Gains
The benefits of PBR are greater to the extent that it can trigger early and
sizable gains in operating performance.
We review here some of the major
considerations that go into an assessment of HQ TransÉnergie’s prospects for
performance improvement.
•
O&M expenses are the most important category of controllable cost in the
short run. These were noted in Section 9 to account for an unusually
small share of the cost of power transmission, which limits the opportunity
for quick performance gains
•
PBR can in some cases lead to sizable short term gains in utility
marketing performance. This is not true in the case for HQ TransÉnergie,
however. As we have seen, its current tariff design does an adequate job
of addressing the major short-run marketing challenges. The company
does not face major congestion challenges.
Its current approach to
ratemaking involves, as it should, extra charges for remote connections
and does not discourage price elastic system uses.
•
HQ TransÉnergie has operated for some time under COSR, with periodic
reviews of operating prudence. Transmission rates were unchanged from
1997 to 2001. The company operated under a self-imposed expenditure
freeze from 2003 to 2006.
As for marketing performance, we have
already noted that the current approach to rate design is fairly sound.
According to the Act, the Régie has to hold a public hearing to review
each rate case that HQ TransÉnergie files. Under these circumstances,
the Régie clearly cannot have the same hopes for improved operating
143
efficiency that regulators in Australia and Britain had when they
implemented transmission PBR for current and recently privatized state
enterprises.
Since
the
energy
market
is
changing
rapidly,
HQ TransÉnergie and the Régie should nonetheless remain vigilant about
prevailing business conditions, and be flexible and open to challenges
prompted by customers' needs and stakeholders' objectives.
•
Effective PBR involves a delicate balance of incentives for cost
containment and the maintenance or improvement of service quality. The
Régie must recognize that a poorly designed PBR plan could get this
balance wrong and result in an undesirable decline in quality. PBR may,
relatedly, encourage HQ TransÉnergie to scale back other activities that
benefit the public but don’t contribute to its bottom line. These include
improvements in customer services and in the transmission system, in
conjunction with environmentally friendly approaches to system expansion
and approaches to input procurement that have more benefits for the
occupants of remote areas.
We may conclude from this discussion that PBR is unlikely to produce
sizable benefits for the customers of HQ TransÉnergie in the short run.
On the
other hand, given the current regulatory system and expected changes in
business conditions in the energy market inside and outside Québec, HQ
TransÉnergie is likely to file rate applications more frequently in the coming
years. This will weaken its performance incentives. For example, the company
could have weaker incentives to contain the cost of O&M expenses and capital
refurbishments under a two year rate case cycle than under a four or five year
rate case cycle.
It should also be noted that in a capital intensive business such as power
transmission, the containment of capital cost is, in the long run, a critically
important dimension of operating efficiency. HQ TransÉnergie has some control
over its capital spending but not necessarily over the cost of funds in financial
144
markets. Extending some form of PBR to capital spending yields benefits only in
the long run.
10.2.4 Conclusion
Our review suggests that COSR works reasonably well in an application to
power transmission in Québec. HQ TransÉnergie filed so far only two rate cases
before the Régie. Its unit transmission rates have been stable since 2001 and the
revenue requirement decreased slightly over the same period. It is currently
operating under an expenditure freeze until 2006 while the quality of service is
under control. It seems, therefore, that COSR has worked quite well thus far.
The eventual implementation of some form of PBR for HQ TransÉnergie
may yield some additional benefits.
However, there is not a strong case to
implement PBR at this time. Such an initiative is not likely to trigger sizable and
rapid improvements in operating performance or to substantially reduce the
regulatory cost of the Régie, not the least reason being that approval for all
investments must be obtained from the Régie according to regulations.
The
implementation of common forms of PBR such as North American style indexing
is, moreover, problematic for the reasons discussed above.
Under these circumstances, I recommend that the Régie proceed with
caution and prudence in implementing PBR for HQ TransEnergie. Meanwhile,
the Régie should continue cost-of-service regulation while gaining more insight
on its operating performance. Even after pursuing cost-of-service regulation on a
regular basis for a number of years, a transitional step involving close monitoring
makes sense for HQ TransÉnergie before moving toward a broad-based
incentive regulatory structure.
A sensible near term goal would be to focus on the development of an
appropriate set of quality indicators and to start monitoring these indicators.
Attention should be paid to similar initiatives in other Canadian provinces (e.g.
Alberta and British Columbia) and in the U.S.A. Quality may be broadly defined
to include issues of customer service as well as reliability. If the Régie intends to
145
consider the development of cost benchmarking, it should expect real progress in
this area to be slow and challenging, as has been the case for other utilities.
Alternatively, the Régie can rely on the historic performance of HQ TransÉnergie
to develop indicators that are traceable and reliable over time.
I believe, in conclusion, that the Régie is well advised to seize the
opportunity and continue work to develop transmission service quality and
prospective cost indicators.
The main purpose would be to provide further
regulatory insight and monitoring of HQ TransÉnergie’s performance, while
building a foundation for future PBR.
RESUME OF
MARK NEWTON LOWRY
December 2005
Home Address:
1511 Sumac Drive
Madison, WI 53705
(608) 233-4822
Business Address:
22 E. Mifflin St., Suite 302
Madison, WI 53703
(608) 257-1522 Ext. 23
Date of Birth:
August 7, 1952
Education:
High School: Hawken School, Gates Mills, Ohio, 1970
BA: Ibero-American Studies, University of Wisconsin-Madison, May 1977
Ph.D.: Agricultural and Resource Economics, University of Wisconsin
-Madison, May 1984
Relevant Work Experience, Primary Positions:
October 1998-Present
Partner, Pacific Economics Group, Madison, WI
Manages PEG’s Madison office. Specific duties include project management and research,
written reports, public presentations, expert witness testimony, personnel management, and
marketing. Research specialties include: performance-based regulation, statistical benchmarking;
utility industry restructuring, and codes of competitive conduct.
January 1993-October 1998
January 1989-December 1992
Vice President
Senior Economist, Christensen Associates, Madison, WI
Directed the company's Regulatory Strategy group. Participated in all Christensen Associates
testimony on energy utility PBR and statistical benchmarking during these years.
August 1984-December 1988
Assistant Professor, Department of Mineral Economics,
The Pennsylvania State University, University Park, PA
Responsibilities included research and graduate and undergraduate teaching and advising.
Courses taught: Min Ec 387 (Introduction to Mineral Economics); 390 (Mineral Market Modeling);
484 (Political Economy of Energy and the Environment) and 506 (Applied Econometrics).
Teaching and research specialty: analysis of markets for energy products and metals.
Mark Newton Lowry
August 1983-July 1984
Page 2
Instructor, Department of Mineral Economics, The
Pennsylvania State University, University Park, PA
Taught courses in Mineral Economics (noted above) while completing Ph.D. thesis.
April 1982-August 1983
Research Assistant to Dr. Peter Helmberger, Department
of Agricultural and Resource Economics, University of
Wisconsin-Madison
Dissertation research on the role of speculative storage in markets for field crops. Work included
the development of a quarterly econometric model of the U.S. soybean market.
March 1981-March 1982
Natural Gas Industry Analyst, Madison Consulting
Group, Madison, Wisconsin
Research under Dr. Charles Cicchetti in two areas:
–
Impact of the Natural Gas Policy Act on the production and average wellhead price of
natural gas in the United States. An original model was developed for forecasting these
variables through 1985.
–
Research supporting litigation testimony in an antitrust suit involving natural gas
producers and pipelines in the San Juan Basin of New Mexico. This research was
occasioned by an antitrust case involving Public Service Co. of New Mexico, Southern
Union Gas Pipeline Co., and Conoco and other natural gas producers.
Relevant Work Experience, Visiting Positions:
May-August 1985
Professeur Visiteur, Centre for International Business Studies,
Ecole des Hautes Etudes Commerciales, Montreal, Quebec.
Research on the behavior of inventories in metal markets.
Major Consulting Projects:
1.
2.
3.
4.
5.
6.
Research on Gas Market Competition for a Western Electric Utility. 1981.
Research on the Natural Gas Policy Act for a Northeast Trade Association. 1981
Interruptible Service Research for an Industry Research Institute. 1989.
Research on Load Relief from Interruptible Services for a Northeast Electric Utility. 1989.
Design of Time-of-Use Rates for a Midwest Electric Utility. 1989.
PBR Consultation for a Southeast Gas Transmission Company. 1989.
Mark Newton Lowry
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
Page 3
Gas Transmission Productivity Research for a U.S. Trade Association. 1990.
Productivity Research for a Northeast Gas and Electric Utility. 1990-91.
Comprehensive Performance Indexes for a Northeast Gas and Electric Utility. 1990-1991.
PBR Consultation for a Southeast Electric Utility. 1991.
Research on Electric Revenue Adjustment Mechanisms for a Northeast Electric Utility. 1991.
Productivity Research for a Western Gas Distributor. 1991.
Cost Performance Indexes for a Northeast Gas and Electric Utility. 1991.
Gas Transmission Rate Design for a Western Electric Utility. 1991.
Gas Supply Cost Indexing for a Western Gas Distributor. 1992.
Gas Transmission Strategy for a Western Electric Utility. 1992.
Design and Negotiation of Comprehensive Benchmark Incentive Plans for a Northeast Gas
and Electric Utility. 1992.
Bundled Power Service Productivity Research for a Western Electric Utility. 1993-96.
Development of PBR Options for a Western Electric Utility. 1993.
Review of the Regional Gas Transmission Market for a Western Electric Utility. 1993.
Productivity and PBR Research and Testimony for a Northeast Electric Utility. 1993.
Productivity and PBR Research and Testimony for a Northeast Electric Utility. 1994.
Productivity Research for a Western Gas Distributor. 1994.
White Paper on Price Cap Regulation for a U.S. Trade Association. 1994.
Bundled Power Service Benchmarking for a Western Electric Utility. 1994.
White Paper on PBR for a U.S. Trade Association. 1995.
Productivity Research and PBR Plan Design for a Northeast Gas and Electric Company. 1995.
Regulatory Strategy for a Restructuring Canadian Electric Utility. 1995.
PBR Consultation for a Japanese Electric Utility. 1995.
Regulatory Strategy for a Restructuring Northeast Electric Utility. 1995.
Productivity Research and Plan Design Testimony for a Western Gas Distributor. 1995.
Productivity Testimony for a Northeast Gas Distributor. 1995.
Speech on PBR for a Western Electric Utility. 1995.
Development of a PBR Plan for a Midwest Gas Distributor. 1996.
Stranded Cost Recovery and Power Distribution PBR for a Northeast Electric Utility. 1996.
Benchmarking and Productivity Research and Testimony for a Northeast Gas Distributor.
1996.
Consultation on Gas Production, Transmission, and Distribution PBR for a Latin American
Regulator. 1996.
Power Distribution Benchmarking for a Northeast Electric Utility. 1996.
Testimony on PBR for a Northeast Power Distributor. 1996.
Bundled Power Service Benchmarking for a Northeast Electric Utility. 1996.
Design of Gas Distributor Service Territories for a Latin American Regulator. 1996.
Bundled Power Service Benchmarking for a Northeast Electric Utility. 1996.
Service Quality PBR for a Canadian Gas Distributor. 1996.
Productivity and PBR Research and Testimony for a Canadian Gas Distributor. 1997.
Bundled Power Service Benchmarking for a Northeast Electric Utility. 1997.
Mark Newton Lowry
46.
47.
48.
49.
50.
51.
52.
53.
54.
Page 4
Design of a Price Cap Plan for a South American Regulator. 1997.
White Paper on Utility Brand Name Policy for a U.S. Trade Association. 1997.
Bundled Power Service Benchmarking and Testimony for a Western Electric Utility. 1997.
Review of a Power Purchase Contract Dispute for a Midwest City. 1997.
Research on Benchmarking and Stranded Cost Recovery for a U.S. Trade Association. 1997.
Research and Testimony on Productivity Trends for a Northeast Gas Distributor. 1997.
PBR Plan Design, Benchmarking, and Testimony for a Southeast Gas Distributor. 1997.
White Paper on Power Distribution PBR for a U.S. Trade Association. 1997-99.
White Paper and Public Appearances on PBR Options for Australian Power Distributors.
1997-98.
55. Gas and Power Distribution PBR Research and Testimony for a Western Energy Utility. 199798.
56. Research on the Cost Structure of Power Distribution for a U.S. Trade Association. 1998.
57. Research on Cross-Subsidization for a U.S. Trade Association. 1998.
64. Testimony on Brand Names for a U.S. Trade Association. 1998.
65. Research and Testimony on Economies of Scale in Power Supply for a Western Electric
Utility. 1998.
66. PBR Plan Design and Testimony for a Western Electric Utility. 1998-99.
67. PBR and Bundled Power Service Testimony and Testimony for a Southeast Electric Utility.
1998-99.
68. Statistical Benchmarking for an Australian Power Distributor. 1998-9.
69. Testimony on Functional Separation of Power Generation and Delivery for a U.S. Trade
Association. 1998.
70. Design of a Stranded Benefit Passthrough Mechanism for a Restructuring Electric Utility.
1998.
71. Consultation on PBR and Code of Conduct Issues for a Western Electric Utility. 1999.
72. PBR and Bundled Power Service Benchmarking Research and Testimony for a Southwest
Electric Utility. 1999.
73. Power Transmission and Distribution Cost Benchmarking for a Western Electric Utility. 1999.
74. Cost Benchmarking for Three Australian Power Distributors. 1999.
75. Bundled Power Service Benchmarking for a Northeast Electric Utility. 1999.
76. Benchmarking Research for an Australian Power Distributor. 2000.
77. Critique of a Commission-Sponsored Benchmarking Study for Three Australian Power
Distributors. 2000.
78. Statistical Benchmarking for an Australian Power Transco. 2000.
79. PBR Testimony for a Southwest Electric Utility. 2000.
80. PBR Workshop (for Regulators) for a Northeast Gas and Electric Utility. 2000.
81. Research on Economies of Scale and Scope for an Australian Electric Utility. 2000.
82. Research and Testimony on Economies of Scale in Power Delivery, Metering, and Billing for a
Consortium of Northeast Electric Utilities. 2000.
83. Research and Testimony on Service Quality PBR for a consortium of Northeast Energy
Utilities. 2000.
Mark Newton Lowry
84.
85.
86.
87.
88.
89.
90.
91.
92.
93.
Page 5
Power and Natural Gas Procurement PBR for a Western Electric Utility. 2000.
PBR Plan Design for a Canadian Natural Gas Distributor. 2000.
TFP and Benchmarking Research for a Western Gas and Electric Utility. 2000.
E-Forum on PBR for Power Procurement for a U.S. Trade Association. 2001.
PBR Presentation to Florida’s Energy 2000 Commission for a U.S. Trade Association. 2001.
Research on Power Market Competition for an Australian Electric Utility. 2001.
TFP and Other PBR Research and Testimony for a Northeast Power Distributor. 2000.
PBR and Productivity for a Canadian Electric Utility. 2002
Statistical Benchmarking for an Australian Power Transco. 2002.
PBR and Bundled Power Service Benchmarking Research and Testimony for a Midwest
Energy Utility. 2002.
94. Consultation on the Future of Power Transmission and Distribution Regulation for a Western
Electric Utility. 2002.
95. Benchmarking and Productivity Research and Testimony for a Western Gas and Electric
Company. 2002.
96. Workshop on PBR (for Regulators) for a Canadian Trade Association. 2003.
97. PBR, Productivity, and Benchmarking Research for a Mid-Atlantic Gas and Electric Utility.
2003.
98. Workshop on PBR (for Regulators) for a Southeast Electric Utility. 2003.
99. Strategic Advice for a Midwest Power Transmission Company. 2003.
100.
PBR Research for a Canadian Gas Distributor. 2003.
101.
Benchmarking Research and Testimony for a Canadian Gas Distributor. 2003-2004.
102.
Consultation on Benchmarking and Productivity Issues for Two British Power
Distributors. 2003.
103.
Productivity and Benchmarking Research for a South American Energy Regulator. 20032004.
104.
Statistical Benchmarking of Power Transmission for a Japanese Research Institute. 2003-4.
105.
Consultation on PBR for a Western Gas Distributor. 2003-4.
106.
Research and Advice on PBR for Gas Distribution for a Western Gas Distributor. 2004.
107.
PBR, Benchmarking and Productivity Research and Testimony for a Western Gas and
Electric Distributor. 2004.
108.
Advice on Productivity for Two British Power Distributors. 2004.
109.
Workshop on Service Quality Regulation for a Canadian Trade Association. 2004.
Strategic Advice for a Canadian Trade Association. 2004.
110.
111.
White Paper on Unbundled Storage and Local Gas Markets for a Midwestern Gas
Distributor. 2004.
112.
Statistical Benchmarking Research for a British Power Distributor. 2004.
113.
Statistical Benchmarking Research for Three British Power Distributors. 2004.
114.
Benchmarking Testimony for Three Ontario Power Distributors. 2004.
115.
Indexation of O&M Expenses for an Australian Power Distributor. 2004.
116.
Statistical benchmarking of O&M Expenses for a Canadian Power Distributor. 2004.
117.
Benchmarking Testimony for a Canadian Power Distributor. 2005.
Mark Newton Lowry
118.
119.
120.
121.
122.
123.
124.
125.
126.
Page 6
Statistical Benchmarking for a Canadian Power Distributor. 2005
White Paper on Benchmarking for a Canadian Trade Association. 2005.
Statistical Benchmarking for a Southeast Bundled Power Utility. 2005
Statistical Benchmarking of a Nuclear Power Plant and Testimony. 2005.
White Paper on Utility Rate Trends for a U.S. Trade Association. 2005.
TFP Research for a Northeast Power Distribution Utility, 2005.
Seminars PBR and Statistical Benchmarking for a Northeast Electric Utility, 2005.
Statistical Benchmarking for a Northeast Power Distribution Utility, 2005.
White Paper on Power Transmission PBR for a Canadian Electric Utility, 2005.
Publications:
1. Public vs. Private Management of Mineral Inventories: A Statement of the Issues. Earth and
Mineral Sciences 53, (3) Spring 1984.
2. Review of Energy, Foresight, and Strategy, Thomas Sargent, ed. (Baltimore: Resources for
the Future, 1985). Energy Journal 6 (4), 1986.
3. The Changing Role of the United States in World Mineral Trade in W.R. Bush, editor, The
Economics of Internationally Traded Minerals. (Littleton, CO: Society of Mining Engineers,
1986).
4. Assessing Metals Demand in Less Developed Countries: Another Look at the Leapfrog
Effect. Materials and Society 10 (3), 1986.
5. Modeling the Convenience Yield from Precautionary Storage of Refined Oil Products (with
junior author Bok Jae Lee) in John Rowse, ed. World Energy Markets: Coping with Instability
(Calgary, AL: Friesen Printers, 1987).
6. Pricing and Storage of Field Crops: A Quarterly Model Applied to Soybeans (with junior
authors Joseph Glauber, Mario Miranda, and Peter Helmberger). American Journal of
Agricultural Economics 69 (4), November, 1987.
7. Storage, Monopoly Power, and Sticky Prices. les Cahiers du CETAI no. 87-03 March 1987.
8. Monopoly Power, Rigid Prices, and the Management of Inventories by Metals Producers.
Materials and Society 12 (1) 1988.
9. Review of Oil Prices, Market Response, and Contingency Planning, by George Horwich and
David Leo Weimer, (Washington, American Enterprise Institute, 1984), Energy Journal 8 (3)
1988.
10. A Competitive Model of Primary Sector Storage of Refined Oil Products. July 1987,
Resources and Energy 10 (2) 1988.
11. Modeling the Convenience Yield from Precautionary Storage: The Case of Distillate Fuel Oil.
Energy Economics 10 (4) 1988.
12. Speculative Stocks and Working Stocks. Economic Letters 28 1988.
13. Theory of Pricing and Storage of Field Crops With an Application to Soybeans [with Joseph
Glauber (senior author), Mario Miranda, and Peter Helmberger]. University of
Mark Newton Lowry
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
Page 7
Wisconsin-Madison College of Agricultural and Life Sciences Research Report no. R3421,
1988.
Competitive Speculative Storage and the Cost of Petroleum Supply. The Energy Journal 10
(1) 1989.
Evaluating Alternative Measures of Credited Load Relief: Results From a Recent Study For
New England Electric. In Demand Side Management: Partnerships in Planning for the Next
Decade (Palo Alto: Electric Power Research Institute,1991).
Futures Prices and Hidden Stocks of Refined Oil Products. In O. Guvanen, W.C. Labys, and
J.B. Lesourd, editors, International Commodity Market Models: Advances in Methodology
and Applications (London: Chapman and Hall, 1991).
Indexed Price Caps for U.S. Electric Utilities. The Electricity Journal, September-October
1991.
Gas Supply Cost Incentive Plans for Local Distribution Companies. Proceedings of the Eight
NARUC Biennial Regulatory Information Conference (Columbus: National Regulatory
Research Institute, 1993).
TFP Trends of U.S. Electric Utilities, 1975-92 (with Herb Thompson). Proceedings of the
Ninth NARUC Biennial Regulatory Information Conference, (Columbus: National Regulatory
Research Institute, 1994).
A Price Cap Designers Handbook (with Lawrence Kaufmann). (Washington: Edison Electric
Institute, 1995.)
The Treatment of Z Factors in Price Cap Plans (with Lawrence Kaufmann), Applied
Economics Letters 2 1995.
Performance-Based Regulation of U.S. Electric Utilities: The State of the Art and Directions for
Further Research (with Lawrence Kaufmann). Palo Alto: Electric Power Research Institute,
December 1995.
Forecasting the Productivity Growth of Natural Gas Distributors (with Lawrence Kaufmann).
AGA Forecasting Review, Vol. 5, March 1996.
Branding Electric Utility Products: Analysis and Experience in Regulated Industries (with
Lawrence Kaufmann), Washington: Edison Electric Institute, 1997.
Price Cap Regulation for Power Distribution (with Larry Kaufmann), Washington: Edison
Electric Institute, 1998.
Controlling for Cross-Subsidization in Electric Utility Regulation (with Lawrence Kaufmann),
Washington: Edison Electric Institute, 1998.
The Cost Structure of Power Distribution with Implications for Public Policy (with Lawrence
Kaufmann), Washington: Edison Electric Institute 1999.
Price Caps for Distribution Service: Do They Make Sense? (with Eric Ackerman and Lawrence
Kaufmann), Edison Times, 1999.
Performance-Based Regulation of Utilities (with Lawrence Kaufmann), Energy Law Journal,
2002.
“Performance-Based Regulation and Business Strategy” (with Lawrence Kaufmann), Natural
Gas, February 2003
Mark Newton Lowry
Page 8
31. “Performance-Based Regulation and Energy Utility Business Strategy (With Lawrence
Kaufmann), in Natural Gas and Electric Power Industries Analysis 2003, Houston: Financial
Communications, 2003.
32. “Price Control Regulation in North America: The Role of Indexing and Benchmarking”,
Methods to Regulate Unbundled Transmission and Distribution Business on Electricity
Markets: Proceedings,
Stockholm: Elforsk, 2003.
33. “Performance-Based Regulation Developments for Gas Utilities (with Lawrence Kaufmann),
Natural Gas and Electricity, April 2004.
35. “Econometric Cost Benchmarking of Power Distribution Cost” ” (with Lullit Getachew and
David Hovde), Energy Journal, July 2005.
Professional Presentations:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
American Institute of Mining Engineering, New Orleans, LA, March 1986
International Association of Energy Economists, Calgary, AL, July 1987
American Agricultural Economics Association, Knoxville, TN, August 1988
Association d'Econometrie Appliqué, Washington, DC, October 1988
Electric Council of New England, Boston, MA, November 1989
Electric Power Research Institute, Milwaukee, WI, May 1990
New York State Energy Office, Saratoga Springs, NY, October 1990
National Association of Regulatory Utility Commissioners, Columbus, OH, September 1992
Midwest Gas Association, Aspen, CO, October 1993
National Association of Regulatory Utility Commissioners, Williamsburg, VA, January 1994
National Association of Regulatory Utility Commissioners, Kalispell, MT, May 1994
Edison Electric Institute, Washington, DC, March 1995
National Association of Regulatory Utility Commissioners, Orlando, FL, March 1995
Illinois Commerce Commission, St. Charles, IL, June 1995
Michigan State University Public Utilities Institute, Williamsburg, VA, December 1996
Edison Electric Institute, Washington DC, December 1995
IBC Conferences, San Francisco, CA, April 1996
AIC Conferences, Orlando, FL, April 1996
IBC Conferences, San Antonio, TX, June 1996
American Gas Association, Arlington, VA, July 1996
IBC Conferences, Washington, DC, October 1996
Center for Regulatory Studies, Springfield, IL, December 1996
Michigan State University Public Utilities Institute, Williamsburg, VA, December 1996
IBC Conferences, Houston TX, January 1997
Michigan State University Public Utilities Institute, Edmonton, AL, July 1997
American Gas Association, Edison Electric Institute, Advanced Public Utility Accounting
School, Irving, TX, Sept. 1997
Mark Newton Lowry
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
American Gas Association, Washington, DC [national telecast], September 1997
Infocast, Miami Beach, FL, Oct. 1997
Edison Electric Institute, Arlington, VA, March 1998
Electric Utility Consultants, Denver, CO, April 1998
University of Indiana, Indianapolis, IN, August 1998
Edison Electric Institute, Newport, RI, September 1998
University of Southern California, Los Angeles, CA, April 1999
Edison Electric Institute, Indianapolis, IN, August 1999
IBC Conferences, Washington, DC, February 2000
Center for Business Intelligence, Miami, FL, March 2000
Edison Electric Institute, San Antonio, TX, April 2000
Infocast, Chicago, IL, July 2000
Edison Electric Institute, July 2000
IOU-EDA, Brewster, MA, July 2000
Infocast, Washington, DC, October 2000
Wisconsin Public Utility Institute, Madison, WI, November 2000
Infocast, Boston, MA, March 2001
Florida 2000 Commission, Tampa, FL, August 2001
Infocast, Washington, DC, December 2001
Canadian Gas Association, Toronto, ON, March 2002
Canadian Electricity Association, Whistler, BC, May 2002
Canadian Electricity Association, Montreal, PQ, September 2002
Ontario Energy Association, Toronto, ON, November 2002
Canadian Gas Association, Toronto, ON, February 2003
Louisiana Public Service Commission, Baton Rouge, LA, February 2003
CAMPUT, Banff, ALTA, May 2003
Elforsk, Stockholm, Sweden, June 2003
Edison Electric Institute, national e forum, June 2003
Eurelectric, Brussels, Belgium, October 2003
CAMPUT, Halifax, May 2004
Edison Electric Institute, national eforum, March 2005
Edison Electric Institute, Madison, August 2005
Edison Electric Institute, national e forum, August 2005
Journal Referee:
Agribusiness
American Journal of Agricultural Economics
Energy Journal
Journal of Economic Dynamics and Control
Materials and Society
Page 9