SERVICE QUALITY IN REGULATED MONOPOLIES

by

Catherine Waddams Price

Centre for Competition and Regulation, University of East Anglia

Bitten Brigham and Lin Fitzgerald

Centre for Management under Regulation and Warwick Business School
University of Warwick

CCR Working Paper CCR 02-4

Abstract

Economic regulators provide incentives for good quality of service as well as

constraints on the prices or revenue which can be charged by firms with monopoly
power. Economic theory suggests that regulators should choose standards
according to consumers’ valuation and the marginal cost of quality improvements,
and that firms respond by equalising the marginal costs from not making
improvements (i.e. the regulatory penalty plus any loss in revenue) with the
marginal costs of improvement. This paper explores the evidence for such
economically rational behaviour by both regulators and regulatees. We use a
specially constructed data set on service quality targets and achievements across the
main UK utility sectors; documentary evidence from regulators; and interviews with
managers in companies subject to those regulators. There are inconsistencies both
within and between regulators. We conclude that regulators are motivated by
political as well as economic factors. And that companies may not respond
primarily to the regulator’s financial rewards or penalties for their quality targets,
with a consequent danger that regulated consumers pay for marketing in
unregulated markets; and the resulting level of service quality may be ‘too high’ in
the economic sense.

August 2002
Key words: economic regulation; service quality; performance measurement, customer
satisfaction

ISSN 1473-8473
We acknowledge funding from the Economic and Social Research Council, project reference
RO22250147; we are grateful to the staff of regulators and companies who were interviewed,
and to Vinh Sum Chau for his excellent and extensive research assistance. We also thank
participants at a seminar at the Centre for Law, Economics and Financial Institutions at
Copenhagen Business School for helpful comments and suggestions.

2
1. Introduction

The balance between quality of service and cost is a major public concern, both in the public
and in the private sectors. Higher quality is generally more expensive to provide, and results in
higher prices for consumers (or taxes for taxpayers). The form of regulation of UK monopolies
has exacerbated these concerns, and regulators have raised targets and penalties over time.
Does the combination of profit maximisation and price cap give companies incentives to save
‘too much’ on safety, environmental and service quality?

We abstract from inter-agency and political dimensions to examine the trade off between price
and quality where both are determined by a single (economic) regulator, namely for the quality
of consumer service which companies should deliver. All the economic regulators have some
such service measures, such as reliability of supply, speed of communication and response to
customers. We collect information on the targets and penalties which regulators have set, and
on the achievement of companies to these incentives, and assess these in the light of the
economic criteria. Moreover both the gas and energy regulator are introducing and
strengthening the explicit rewards and penalties for service quality within the price cap.

We use three sets of sources. The first is a data base, especially constructed for this study, of
the standards, targets and fines set and achieved in the four utility industries (gas, water,
electricity and telecoms) since privatisation. The second is documentation from water and
energy regulators about setting quality targets; and the third, interviews with a number of
senior managers in two regulatory offices and three companies.

In section 2 we discuss the economic arguments for determining quality criteria. Section 3
presents evidence on the regulators’ choice of standards, targets and fines, and section 4 traces
how service standards have developed across the four sectors. Section 5 reports the company
responses, both in terms of achievement and comment. Section 6 identifies some important
practical implications for choosing quality targets and penalties in the light of company
responses.

3
2. Quality of Service and regulated industries

2.1 Measuring performance

Measuring performance has become increasingly important in management of organisations,

both public and private, over the last two decades. Performance measurement has been used as
an instrument to measure the success of an organisation in meeting its own objectives, and as a
management tool for improvement (Merchant, 1985, Fitzgerald et al 1991, Euske et al 1993).
As performance measures are increasingly ‘politicised’, particularly in the public sector, there
is also growing concern that performance measures can have unintended dysfunctional
consequences, as managers respond to the incentives which they provide (Burchell et al, 1980).
In the economics literature, the issue has been presented as an agency problem, applied to both
private and public sectors. Holmstrom and Milgrom (1991) address problems of multi-tasking
in a private company; Rees (1984) discusses the objectives and incentives of managers in
nationalised industries. Several multi-dimensional frameworks have been proposed which
stress the interdependence of financial and non-financial performance measures (Johnson and
Kaplan, 1987, Fitzgerald et al, 1991, Nanni et al, 1992, Eccles and Pyburn, 1992, Kaplan and
Norton, 1993). However as Jensen, 2002, has shown, if robust scorecards cannot be found to
balance the interests of different stakeholders, this may have the perverse effect of making
managers accountable to no-one.

When industries are privatised, performance measurement takes on an additional dimension.

The original economic model of price cap regulation was developed as a temporary measure in
telecoms, an industry expected to become competitive very rapidly (Littlechild, 1983). While
it tried to surmount many of the recognised problems of traditional US style rate of return
regulation, the need to reset price caps introduced a strong rate of return element to their
implementation and practice. Littlechild acknowledged that “it would seem sensible to ensure
that quality of service did not deteriorate as a result of the tariff reduction scheme” (p. 35).

Literature on quality of service in regulated industries is relatively sparse. Spence (1975)

showed that an unregulated monopolist might either under or over provide quality, according
to the preferences of the marginal consumer relative to the average. Rate of return regulation is
likely to increase incentives for over provision of quality if its provision is capital intensive and

4
would boost the rate base (Sherman, 1989). The rate of return element inherent in price cap
regulation suggests that it, too, might stimulate over provision of quality. However price cap
could also lead to degradation of quality, as indicated in the Littlechild quotation above, if
higher quality incurs higher costs. Conversely, profitability may be positively related to
quality of service in some respects. Ogden (1997) identifies prompt bill payment, reduced
costs of handling complaints and reputation in other competitive markets as three incentives to
improve service cited by managers in water companies.

2.2 Determining optimal quality

There are a number of ways in which regulators can address quality issues. Baldwin and Cave
(1999) discuss the difficulties of determining both the optimal level of quality and whether it is
being delivered, and recognise the usual “tendency for the measurable to drive out the
unmeasurable” (p.254). Ideally, consumers can be given a choice of quality, and charged
according to the costs of providing it. But the possibility of choosing different service qualities
has been highly politicised, with public outcry at the possibility that those with more money
should be allowed to ensure a more reliable electricity or water supply, for example.

One characteristic of network industries is the common level of quality provision across at
least some groups of consumers (those supplied from the same water source or section of a
distribution system), so that one consumer’s quality of supply must be shared with that of his
neighbour. In such cases regulators have to determine which services they should set standards
for, and identify, appropriate target levels and penalties. In economic terms, the appropriate
incentives to the firm would involve a penalty for breaching the standards which is equal to the
total monetary valuation of the consumer losses incurred for each quality standard (e.g. the
reliability of a particular distribution wire). In each case the optimal quality is that at which the
willingness to pay for additional quality by all the consumers affected is exactly equal to the
cost of providing that extra quality. But this would clearly be very difficult operationally, both
in establishing what consumers wanted and in fine tuning the system (the costs of such fine
tuning would itself not be worth the benefits). Ascertaining consumers’ willingness to pay is
itself problematic. Consumers often give inconsistent responses to surveys about willingness
to pay, because of both conceptual difficulties and well known free rider problems; moreover
there is usually an asymmetry between willingness to pay and compensation demanded if
standards fall (willingness to accept), which is typically about twice as high. There are obvious

5
explanations of this (for example consumer expectations) but it does not ease the task of the
regulator in determining what targets should be set.

One crucial factor determining consumer valuation of quality is their previous experience.
Whether standards were above or below the optimum level in the regulated industries’
nationalised predecessors is a moot point. Standards were generally driven by the engineers
who operated and in most cases managed the industries, and were likely to be determined by
technical capacity rather than consumer choice, an argument that they may have been ‘gold
plated’. On the other hand the control regime included incentives to increase labour intensity
(Rees, 1984) and in the years just before privatisation investment by many of the industries
was severely restricted under public expenditure limits. Moreover consumer expectations have
increased across the economy, but particularly for these industries since they became private
(Ogden 1997). This phenomenon was illustrated by the greater demands made of water
companies in the drought of 1995, compared with that in 1976 when the industry was still
public. Furthermore, public consciousness and valuation is affected by media coverage and is
unlikely to be stable, making optimum quality very difficult to identify.

2.3 Rewards and penalties for quality

The economic regulator could incorporate quality concerns directly in the price cap, making
the price or revenue allowed directly dependent on quality delivered. If quality increased a
company would be allowed to raise its price, recouping some of the costs directly from the
consumers who had benefited, and a degradation of quality would similarly be reflected in
lower allowed prices. Alternatively, the regulator could impose a particular level of quality,
and make breach a matter for legal action. In this case the penalties are likely to depend on
legal arguments, but should conform to the same principles as if penalties are applied by the
regulator. Ideally any degradation away from the optimum level should confront the company
with costs which are equal to the value of the total consumer losses. In some cases high levels
of compliance may be achieved with lower penalties, but this generally requires some other
mechanism to be in force (see for example Harrington, 1988, Livernois and McKenna, 1999,
Rodrigues-Ibeas, 2002). There may also be direct or indirect loss of revenue from degraded
quality. For example a breakdown in the distribution system means less product is delivered
and paid for; where reliability is very poor, and/or very valuable, alternative supplies may be
ensured, as through private generators with subsequent loss in distribution revenue. But most

6
consumers do not have an alternative if a network fails, so revenue losses are small and costs
need to be reflected in a financial penalty for the firm. In terms of incentives it makes no
difference whether this is in the form of lower allowed revenue, fines or compensation, i.e.
whether the money is paid to a central regulator or government fund or to consumers through
lower prices or compensation. But there is clearly an attractive political and distributive
argument that consumers who have suffered poor service should receive some compensation.

Compensation may be discretionary, in the sense that it depends on consumers claiming

payment. To give appropriate incentives to companies against breaching the standards,
compensation rates need to be adjusted according to the proportion of consumers likely to
claim (so that if only one tenth claim, the payment should be ten times the actual damage
incurred). This of course raises issues of equity, particularly if it is high income consumers
who are most likely to complain and claim compensation. A counterargument is that those
most likely to complain are those who have suffered the greatest damage – but in monetary
terms the valuation of such damage itself is likely to depend on income and so be
distributionally determined. An alternative which is increasingly applied is to make
compensation mandatory and require that the company pays it to all consumers affected.

Quality may be determined, rewarded or penalised, either in absolute terms, or relative to other
providers, where there are several companies that can be compared (e.g. regional monopolies
as in water or electricity). Where such comparison is possible it is likely to be used across a
range of variables, including costs as well as quality.

3. Regulators’ approaches to targets, penalties and rewards

Economic regulators set both price caps and some quality standards in the industries which
they regulate. Since privatisation, price levels which industries were allowed to charge under
their price cap generally decreased while the quality standards were being raised1. (Markou
and Waddams Price (1999) provide summaries of both price cap and service quality targets,
and Chau, 2002, gives details of the targets themselves). We focus on the two regulators which

1
The only exception is the water industry, where average prices rose in the first ten years after privatisation,
reflecting both investment requirements and higher environmental standards, mostly imposed by the EC (such
environmental quality is excluded from this analysis).

7
have the longest and most consistent history in this respect: the Office of Water Services,
Ofwat, economic regulator of the water industry; and the Office of Gas and Electricity
Markets, Ofgem, which regulates the gas and electricity industries2. We use documentation to
identify arguments for selection of quality standards, target levels and fines. These refer both
to the individual measures of quality, and to more recent composite measures (Overall
Performance Assessments, OPA in water and the Information and Incentives Project (IIP) in
electricity).

The water regulator started with clear ideas of the role of consumers in determining standards
and targets. In preparing for the first review of the price cap Ofwat stated that
“… [C]ustomers should be sovereign and given the material on which they can make informed
judgements about the quality of the service they want and the price they are prepared to pay.”
[Ofwat 1992, p.6]

The electricity regulator has similarly recognised the importance of consumers’ views in
developing the IIP project
“…it is important to consider customer’s willingness to pay for improvements to quality
service.” [Ofgem, 2000a, p.16]

and in relation to regulation of Transco

“Consumers are interested in quality as well as prices. Therefore the price control review has
involved consideration of the appropriate guaranteed and overall standards of performance and
other quality indicators.” [Ofgem, 2001a, p.3]

“… [O]ver time, it may be necessary to incentivise additional services which are identified as
being of value to customers or where the incentives on Transco to deliver the appropriate
quality of service need to be improved.” [Ofgem, 2000b, p.27]

In the course of the second price review, the water regulator had a clearer idea of what
consumers wanted
“Experience to date suggests that there is limited willingness among customers to pay extra for
improvements in service.” [Ofwat, 1997a, p.13]

2
In telecoms, where consumers have some degree of choice, companies set and report on their own performance
targets; the rail and postal systems’ quality standards have been established too recently for us to be able to trace
their evolution.

8
The water regulator also recognised the difficulties of a common distribution system
“… [A] number of other features which characterise the water business … the difficulty of
differentiating quality for different customers. All customers must pay for improvements in
quality and service levels.” [Ofwat, 1993, p. 16]

And the trade-off between quality and price was clearly recognised at an early stage by Ofwat
“The strategic assessment shows how much higher bills could rise if all the quality
improvements currently being considered were implemented as new obligations … All the
figures in this assessment are subject to a wide degree of uncertainty.” [Ofwat, 1992, p.5]

“Some of the benefits could go to customers in the form of higher standards of service or lower
prices;” [Ofwat, 1993, p.7]

At about the same time the electricity regulator pointed out that the trade-off was not automatic
“High standards of customer service do not necessarily involve higher costs, though they may
do where investment is required.” [Offer, 1994, p.xiii]

The energy regulator recognised these tensions later in developing the IIP
“There may be a concern that in the future companies will attempt to outperform the cost
assumptions underlying price controls by reducing the quality of supply of the service
provided.” [Ofgem, 1999a]

And the relation to the price cap was clear in the 1999 price review of electricity distribution
“… while recognising that sufficient revenue must be raised to maintain an appropriate quality
of service, to finance required new investment and to allow an appropriate return to
shareholders.” [Ofgem, 1999b, p.5]

“The trade-off between higher (lower) prices and higher (lower) quality involves two decisions
that are taken at the time of the price control review. In effect a judgement is made on the new
value for money line and where on the line each company is positioned.” [Ofgem, 2000a, p.16]

Similar views were expressed by the water regulator

“…companies which are more efficient and which have higher standards of performance should
benefit at the price review compared with those that are less efficient or have lower standards.”
[Ofwat, 1998, p.9]

9
The first electricity regulator had been much more cautious
“… I do not believe it would be appropriate to make direct connections between performance
under the Standards of Service and the supply price control. I will, however, review the
potential for linking performance and price when I review the price controls on the RECs’
distribution businesses.” [Offer, 1993, p.20]

The energy regulator’s view of the role of standards was clarified

“Ofgem believes that is important to introduce guaranteed and overall standards of performance
in respect of certain non-contestable activities carried out by Transco. They will provide the
primary protection to Transco’s final consumers in relation to quality of performance, at least
until they are supplemented by financial incentives on interruptions under Transco’s price
controls …” [Ofgem, 2001c p.75]

Ofwat raised a particular issue about the timing of quality improvements and related price
increases
“Currently customers are paying for quality improvements through bills while the work is being
carried out before the improvements have been completed. This is contrary to normal
commercial practice where suppliers of new services only charge once the service is provided
… a new approach to the funding of quality obligations … Companies would only be able to
raise revenue from customers for quality obligations after they had delivered the planned
outputs. This would provide companies with stronger incentives to deliver schemes on time.
[Ofwat, 1997b, p.27]

In terms of how to ascertain consumer needs, Ofwat has most recently relied on consumer
surveys carried out by the companies, and the views of the Consumer Councils (CSCs).
Reporting these it notes
“There was less consistency in the precise balance between bill reductions and service
improvements. The companies have concluded that customers would prefer a balance in favour
of service improvements, whereas others, including the CSCs, take a different view.” [Ofwat,
1999, p.6]

This quotation also highlights the strategic role that consumers’ views can take in the debate
between the regulator and the companies. Here the regulator seems to suspect that companies

10
are interpreting consumer preferences to argue for more investment and higher prices, while
the consumer representative body disagrees.

4. Evolution of Service quality measures, target levels and penalties

Measures of service in UK regulation evolved in several directions. In some cases (e.g.

telecoms) comparative performance indicators are published at quarterly intervals, but the
regulator imposes no direct financial implications. This typically occurs where there is some
degree of competition, and the information can be used by consumers in choosing suppliers
(e.g. telecoms and energy supply). For monopoly sectors, regulators have typically developed
two sets of standards. One relates to performance for individual consumers and attracts
compensation (fixed by the regulator) for the consumer if it is not met (in gas and electricity
known as guaranteed standards). Such compensation may be automatic or in response to
consumer requests. The number of requests in the latter case depends on the company’s
diligence in informing customers of their entitlements. One of the electricity industry’s
guaranteed standards (GS10) is to notify consumers of their rights to compensation within a
specific time.

The second set of standards, overall standards, relates to areas where it is inappropriate to
provide individual guarantees, but where customers have a right to “expect” a minimum level
of service. These may cover similar areas, and are often focused on the system as a whole and
achievement of a particular percentage of a target standard. A third measure, combining
overall performance across several measures, has recently been introduced for water and
sewerage and electricity distribution, a form of yardstick competition as suggested by Shleifer
(1985) in which quality can be incorporated as a dimension of output. We compare these in
section 4.5 below. The penalty here is a lower position in a league table, with direct financial
reward or punishment through the allowed price cap. Legal instruments are generally used to
enforce standards set by non-economic (e.g. environmental or health and safety) rather than by
the economic regulators, and are not the direct subject of this paper (though see Baron, 1985
for the implications of separating quality and economic regulation).

We focus on the monopoly parts of the privatised industries where regulation and quality of
service measures are likely to persist. The current regulated monopoly networks are Transco

11
(gas transmission and distribution), the National Grid Company (high voltage national
electricity transmission), the Regional Distribution Companies (low voltage regional electricity
distribution), and the water companies (vertically integrated water only or water and sewerage
companies). We omit systematic analysis of performance measures in railways because of
their relatively short and troubled history in the private sector.

4.1 Water

In water the Director General of Water Services specifies a set of standards known as ‘DGs’.
These refer to a variety of service dimensions ranging from operational aspects such as
pressure of mains water and risk of sewer flooding, to consumer interface issues such as
response to billing enquiries and written complaints. Since the original eight standards were
established when the industry was privatised in 1989/90, some have been made more specific,
some have been abolished, and two new categories, referring to bills of metered consumers and
speed of telephone response, have been added.

Almost all target levels have been raised since they were introduced (Chau, 2002). When the
scheme was first introduced, compensation payments for residential consumers depended on
consumers making a claim. In 1993 penalties were doubled and extended to business as well
as residential consumers. Take-up was low (few consumers knew of their rights to
compensation, or could not be bothered to claim) and from April 1993 automatic compensation
was introduced for some standards. This would increase the costs for the company quite
considerably, both in terms of compensation paid and because of the process costs of
identifying and reimbursing the consumers and demonstrating to the regulator that this had
been done. In 2001 penalties were again raised, and in most cases doubled to £20 per incident.

The regulator’s valuation of consumer loss has clearly increased considerably, both in terms of
individual consumer compensation and, even more, in terms of the costs of breach for the
companies (since both the amount and the extent of compensation has been increased).
However the standard penalties across the measures suggest that there was little fine tuning of
estimates of consumer valuation, despite recent increases in two of the penalties.

12
4.2 Electricity

The electricity regulator applied guaranteed standards for individual consumers, which should
be met in every case, or a penalty, whose minimum level was set by the regulator, was payable.
Overall standards for the system were also set, either as supplementary to guaranteed standards
or in different areas. No predetermined penalty attached to failing to meet these overall
standards. These standards were originally applied to the public electricity suppliers who
provided both distribution and supply services under a joint license; recent legislative changes
to accommodate the introduction of competition have involved the separation both of
businesses and licensing for these two functions.

The Director General originally set 10 guaranteed standards, and a further one was added in
1998. Required levels (i.e. those triggering compensation payments) have tightened
considerably across the board, particularly recently. Required response to consumer queries
was reduced from 10 to 5 working days in 1993-94; the same reduction was made in the time
to provide an estimate of charges in 1998-99, and more rapid responses to meter problems and
time to provide a meter were imposed. In 2000-01 the target for restoration of interrupted
supply was reduced from 24 to 18 hours.

The minimum compensation payable for most guaranteed standards has increased over time,
typically from an initial £10 per incident for each, to £20 to £50 by 1997/98. An exception is
the change in the penalty for failing to supply a meter. This rose from £20 to £40 for
residential consumers in 1993-94, and fell again to £20 in 1994-95 where it has stayed since
(though this partly reflects a tightening of the time allowed, noted above).

The overall standards were initially set individually for each company “after review of recent
performance, consultation with the companies, the Electricity Consumers Committees and
other customer representatives” (Chau, 2002). Such a procedure is consistent with balancing
consumer valuation against costs of increased quality, but there is no direct evidence of such an
approach, despite the quotations of the previous section. All the overall standards were raised
to 100% in 1998/99, and remained there, except for one. The changes in the standards are
summarised in table 1.

13
 Table 1: Changes in electricity guaranteed standards (source Chau, 2002)
Year GS-1 GS-2 GS-3 GS-4 GS-5 GS-6 GS-7 GS-8 GS-9 GS-10 GS-11
Within 3
Within 4 hours Within 10 Write to
working days All
of any working days Visit or reply Visit or reply A substantive customer
for domestic, 5 At least 2 days' appointments
1991/92 notification 24 hours (simple jobs), within 10 within 10 reply within 10 within 10 N/A
working days notice to visit on a day
during working 20 working working days working days working days working days of
for non- must be kept
hours days (others) failure
domestic
£20 domestic,
£50 non- £20 domestic, £10 domestic,
1991/92 £10 domestic, plus £50 non- £20 £20 non- £10 £10 £10 £10 £10 N/A
£10 per further domestic domestic
12 hrs
A substantive
1993/94 . . . . . . . reply within 5 . . N/A
working days

£40 domestic,
£100 non-
£20 domestic,
domestic within
1993/94 £20 £40-£100 £40 £40 non- £20 £20 £20 £20 £20 N/A
24 hrs, plus
domestic
£20 per further
12 hours

Offer and keep

a morning or
afternoon
1994/95 . . . . . . . . appointment, . N/A
or a more
precise time if
requested

£20 or £40
1994/95 . . domestic, £100 . . . . . . . N/A
non-domestic

Visit within 10
working days
1995/96 . . . . . . or substantive . . . N/A
reply within 5
working days
1995/96 . . £20 - £100 . . . . . . . N/A
1996/97 . . . . . . . . . . N/A
£40 (domestic)
£100 (non-
1996/97 . . . . . . . . . N/A
domestic) +
£25 per 24hrs

Respond within
3hrs weekdays Visit within 10
7am-7pm, and working days
1997/98 . . . . . . . . N/A
within 4 hrs at or substantive
weekends 9am- reply within five
5pm

£50 (domestic
customers) for
restoring
1997/98 . supplies within . . . . . . . . N/A
24 hours, plus
£25 per further
12 hours

Arrange an A morning or
appointment Visit within 7 Visit within 7 A substantive afternoon Write to the
Within 5 days Within 3 hours
within 2 Customers working days working days reply and appointment, customer and
for simple jobs, on weekdays
1998/99 . . working days must be given or substantive or substantive agreed refunds or a timed make payment
and 15 days for and 4 hours at
(domestic), 4 5 days' notice reply within 5 reply within 5 paid within 5 appointment if within 10
others weekends
working days days working days working days requested by working days
(non-domestic) the customer

£50
(domestic),
£100 (non-
1998/99 . . . . . . . . . £20
domestic), £25
for each 12
hour period
Must be
2000/01 . restored within . . . . . . . . .
18 hours
2000/01 . . . . . . . . . . .

14
Both the National Grid Company and Transco transmission have complex systems of
incentives for transmission services; in NGC’s case this is based on sliding scale regulation and
in Transco’s on auctions for entry and exit capacity, which will be extended to long term
capacity in the autumn of 2002. There is a simple supplementary set of performance targets for
NGC, agreed with Ofgem, for infrastructure to ensure there is sufficient capacity at peak. No
penalty or reward attaches to performance in this dimension, though the regulator does publish
comparative figures on the four Great Britain network systems (national grid, the two Scottish
networks and the interconnector to France).

4.3 Gas

Like the electricity distribution companies, Transco was privatised as part of a combined
distribution-supply business, and did not become a separate business entity until the mid 1990s
(ownership was separated in 1997). Transco has obligations as a system operator, and
incentives to provide adequate quality in this capacity have been sharpened from 2002 (Ofgem
2001b). In addition Transco’s quality of service to final consumers was converted from a
voluntary to a required basis as part of the recent price review (Ofgem 2001 c). Five
guaranteed standards, attracting different levels of payment, were introduced. These related to
supply interruptions exceeding 24 hours, reinstatement of premises after re-laying service
pipes, making and keeping appointments, providing adequate alternative heating and cooking
facilities for potentially vulnerable customers who were disconnected, and notification of
consumers or shippers of their rights to payments. Seven overall standards were imposed (but
at rather lower levels than the 100% provided for most of the electricity industry). Ofgem also
plans an incentive scheme for Local Distribution Zones (LDZs) similar to that introduced for
distribution companies in the IIP, although comparison within a company raises rather different
issues from those between companies.

4.4 Comparison within Ofgem

For energy, a single regulator determines quality standards and penalties in the gas and
electricity networks; any differences between the two should reflect either customer
preferences or differential costs of providing elements of quality or both. Although the
regulator is the result of merging two separate bodies, the gas and electricity regulators, who

15
may have had different priorities, these have been a single body since 1999, and so it is
reasonable to expect some consistency or at least convergence of approach. The following
table compares the most recently published guaranteed and overall standards in electricity and
gas (concentrating on those standards in electricity which are most relevant to the distribution
element).

16
Table 2: Gas and Electricity Guaranteed Standards 2001-2 (source Chau, 2002)
SERVICE Performance Level Penalty Payment (£) Difference
Electricity Gas Electricity Gas Electricity Gas

GS1: Failure of ~ 3 hrs weekday, ~ 20 ~ ~

Supply fuse 4hrs weekend

GS2: Restore GS1: restored in 18 hrs compensation 50 30 6 hrs & £20-

supply after Restoring made for each domestic, 70
faults supply after 24 interruption 100 difference
unplanned beyond 24 hrs nondom, +
interruption 25 each 12
hrs

GS5: Notice of ~ 5 days notice ~ 20 ~ ~

Supply domestic,
Interruption 40
nondomest

GS7: 7 days or 5 for 20

Responding to substantive reply
meter problems

GS6: ~ 7 days or 5 for ~ 20 ~ ~

Investigation of substantive reply
Voltage
Complaints

GS9: Making GS3: Making Must offer and Must offer and 20 20 SAME
& Keeping & Keeping keep keep
Appointments Appointments appointment appointment

GS10: GS5: 10 working days Where not 20 20 SAME

Notifying Notifying informed in 20 Payment,
Customers re Customers re working days, but 10 days
Standards Standards payment is due payment
difference

GS11: ~ respond in 3hrs ~ 20 ~ ~

Responding to weekdays, 4 hrs
Pre-payment weekend
Faults

~ GS2: ~ Premises to be ~ 50 ~
Reinstatement reinstated within domestic,
of Consumer's 10 working 100
Premises days; payment nondomest
for each 5 ic
working days
thereafter

~ GS4: Adequate ~ Payment if ~ 24 ~

Heating and supply
Cooking disconnected
Facilities without
alternatives for
disadvantaged

17
Table 3: Gas and Electricity Overall Standards 2001-2 (source Chau, 2002)
SERVICE Performance Level Achievement Level Difference
(%)
Electricity Gas Electricity Gas Electricit Gas
y

OS1a: Faults ~ 3 hrs ~ Various ~ ~

reconnected in 3
hrs

OS1b: Faults ~ 18 hrs ~ Various ~ ~

reconnected in 18
hrs

OS2: Voltage ~ 6 months ~ 100 ~ ~

faults corrected

OS8: Respond to OS4: 10 working 5 working 100 90 5 days,

LETTERS Acknowledge days days 10%
correspondence difference

~ OS1: Respond to ~ 30 seconds ~ 90 ~

telephone calls

~ OS2: Notify ~ A letter to be ~ 95 ~

planned supply provided 5
interruptions working
days in
advance

~ OS3: Informing ~ 12 hours of ~ 97 ~

customers when knowledge
due to be
reconnected

~ OS5: Visits after ~ 2 working ~ 93 ~

receipt of days after
complaint receipt of
corresponde
nce, visit 5
working
days after
contact

~ OS6: Response ~ Substantive ~ 90 ~

to response
COMPLAINTS within 10
working
days

~ OS7: Respond to ~ 1 hr ~ 97 ~
EMERGENCY uncontrolled
calls escapes, 2
hrs
controlled
escapes

18
We would expect differences both in consumer preferences and costs between the industries.
Consumers generally find supply interruption more inconvenient in electricity than in gas
(which has more substitutes); the costs of maintaining supplies are likely to be different, and
restoring interrupted gas supplies is much more complex and expensive than for electricity.
However it is unlikely that consumer preferences or costs vary much for service aspects such
as response to telephones, letters or complaints.

We observe three types of differences between the standards. The first is in the standards
themselves; the second is in the level of service expected (e.g. time expected for response); and
the third is in the penalties imposed. Electricity has guaranteed standards for notice of supply
interruption, and responding to meter problems and prepayment faults; and overall standards
for reconnection. All of these could be applied to gas but are not; gas does however have two
overall standards with respect to notifying consumers of disconnection and reconnection.
Conversely, gas has guaranteed standards to provide adequate heating and cooking facilities for
disadvantaged persons and overall standards for responding to telephone calls, visiting after
complaints are received, and response to complaints. Telephone response is now one of the
dimensions of the electricity distribution Information and Incentives Project3 (section 4.5).
Nevertheless substantial differences between the measures of service quality remain between
the industries which are difficult to attribute to differences in technology or consumer
preferences.

Four aspects of quality are subject to standards across both industries. Both have guaranteed
standards for restoring supply after unplanned interruption, making and keeping appointments
and notifying customers about the standards; and both have an overall standard for
acknowledging correspondence. Both must offer and keep a morning or afternoon appointment
and pay the same £20 compensation if they fail to do so. The target for restoring supply is
shorter in electricity (18 hours rather than 24), which may well reflect consumers’ greater
reliance on electricity. However the penalty for exceeding the target is greater in electricity
(minimum £50 as compared with £30 for gas), which is less easily explained. More surprising
is the difference in expected time for notifying consumers about the targets (10 working days
for electricity and 20 for gas), while failure to comply attracts the same compensation of £20.
The situation is reversed for the overall standard on acknowledging correspondence, where a

19
consumer can expect a reply within five working days for gas, but has to wait for ten for
responses to electricity letters. However the higher standard is balanced by a lower target level
(90% for gas compared with 100% for electricity) so they are not necessarily incompatible.
These differences emphasise the role of history and evolution in determining the levels rather
than of systematically balancing consumer benefit and cost.

4.5 Use of comparative performance by Ofgem and Ofwat

Comparative regulation measures are an important tool where there are several companies, and
Ofwat is further down this path than Ofgem. Ofwat has developed an overall performance
measure with weights for each of the quality supply performance measures (the DGs) and
measures of water quality and environmental performance. The companies’ performance on
this overall performance measure affects their prices, though so far only for companies with
exceptionally good or poor performance. In electricity distribution there was considerable
criticism of the way the 1999 review of distribution prices was conducted (e.g. Tilley and
Weyman-Jones, 1999), and a new Information and Incentives Project (IIP) was established as
part of that review to come into operation in April 2002. The IIP aimed to address criticism
that there was insufficient incentive for service improvement, and to introduce direct incentives
for such improvement, and is discussed in section 4.5 below.

The comparative assessments of the companies has different impact in the water and energy
industries. For water, the overall performance measure has a maximum of half a percent
impact on revenue, with an increase likely in the future if the data improve. In electricity, up to
two percent of revenue is affected in the first two years with a subsequent increase to four
percent.

We see that all regulators have successively tightened standards of service expected and the
compensation to be paid to consumers (i.e. the cost of breach). While their motivation is
similar to that which economic models predict, it is clear that they do not ‘fine tune’ their target
levels and fines in the way that pure theory might predict, and that they are acutely conscious
of the limits of data which they face. Much of their increasing pragmatism is evident from the

3
Other discrepancies are more obviously explicable by industry differences, for example voltage excursions in
electricity, reinstatement of consumers’ premises and emergency calls for gas.

20
quotations in section 3. In the next section we show company responses to the incentives with
which regulators have presented them.

5. Response to performance measures by UK companies.

5.1 Company service quality performance

Most measured standards have shown marked improvement in performance since they were
introduced. Of course this begs the question of performance in other areas where performance
was not measured, and which (by definition) cannot be answered.

5.1.1 Water

The performance of the water and sewerage companies against the various DG measures shows
a pattern of consistent improvement across all companies and measures (apart from occasional
divergences in particular companies) – see figure 1 for overall performance and figure 2 for an
example of a typical measure. The improvement is particularly noticeable and steady in those
measures over which the companies have most control, i.e. response to written, billing and
telephone contact, and bills for metered consumers. It is a little more erratic for the two
measures which are dependent on climate, viz. sewer flooding and water pressure. But even
here we see that variability tends to come from two companies rather than be widespread
across companies, and that the general level of performance is improving over time. In
particular, average performance improves because the worst performing companies improve to
levels close to those of the best performing companies at the beginning of the period. Thus the
change comes not from moving the frontier of ‘best practice’ but from ‘catching up’ towards
the frontier by those who were some distance from it This no doubt reflects the impact of the
comparative system itself as the league tables were developed.

21
 Figure 1: Water Industry Performance across
DG
40 DG2: Property at risk of
low pressure

35
DG3: Properties subject
to unplanned supply
interruptions of 12+
30
DG4: Population subject
to hosepipe bans
Performance (%)

25
DG5: Properties subject
to sewer flooding
incidents (overloading
20
DG6: Billing contacts
not responded to (within
15 5 working days)

DG7: Written contacts

not responded to (within
10 10 working days)

DG8: Bills not based on

meter readings
5

DG9: Received
telephone calls not
0 answered within 30
seconds

1990-19911991-19921992-19931993-19941994-19951995-19961996-19971997-19981998-19991999-20002000-2001

Time (Years)

Figure 2: Flooding
f S
180

160
Performance (No. Connected Properties)

140
Anglian

120 Dwr Cymru

Northumbrian

Severn Trent
100
South West

Southern
80
Thames

United Utilities
60 Wessex

Yorkshire

40

20

1989-19901990-19911991-19921992-19931993-19941994-19951995-19961996-19971997-19981998-19991999-20002000-2001

Time (Year)

5.1.2 Electricity and Gas

A similar pattern emerges for the electricity distribution companies. There are steady and
consistent improvements, and the average performance has improved primarily because the
worse performing companies have brought their quality closer to that of the best (see figure 3
for an example). Among those who were performing well at the beginning, there has been

22
little improvement in many measures. This may represent a technical limit to improvement, or
the ‘optimum’ beyond which further quality improvements are not worthwhile, or, again, the
lack of pressure which comparative performance places on the best performers, except to retain
their position at the head of the table.

Figure 3: Electricity distribution AVAILABILITY: Minutes Lost

400

350
Minutes Lost (per connected customer)

Eastern
300 East Midlands
London
Manweb
250 Midlands
Northern
NORWEB
200
SEEBOARD
Southern
SWALEC
150
South Western
Yorkshire
100 Hydro-Electric
ScottishPower
AVERAGE
50

90/91 91/92 92/93 93/94 94/95 95/96 96/97 97/98 98/99 99/00 00/01

Time (Year)

NGC’s transmission services are subject to a complex incentive regime, largely based on
sliding scale regulation. Quality provision in the infrastructure relates mainly to providing
adequate capacity to meet peak demand. Some quality measures are chosen jointly by the
company and the regulator, and these indicate a consistently good and improving performance.
The National Audit Office notes that “In the past, Ofgem have not provided any detailed
incentive schemes for capacity because NGC’s basic quality performance has been very good.”
(NAO, 2002, p.33).

5.2 Manager’s attitudes to service quality measures

We asked three senior managers in different regulated monopoly companies about achieving
the regulator’s performance measures, in the light of our results reported earlier in this paper.
We focus here on their attitudes to quality measures set by the regulator, rather than the means
by which they do it, which are discussed in Brigham and Fitzgerald, 2001.

23
All three respondents perceived conflicting objectives between government and regulator, and
sometimes between different regulators (notably economic and health and safety or
environmental regulators).

Concerns were expressed both over the competence of the regulator, and over the tendency for
‘inflation’ or ratcheting up of standards. All three managers expressed doubt about how the
regulator had chosen standards, and the consequent levels which were set. Two companies
conducted their own consumer surveys to assess not only what mattered to consumers but also
what they would be willing to pay, suggesting in one case that the “appropriate customer
standards were below the current standards”. Concern was expressed by two companies
about inconsistency between short term and long term performance, particularly where
networks needed to be taken out of use temporarily in order to improve them for long term
service.

Two of the managers commented on the form of incentives. One felt it was appropriate that
credit for out-performance or pain of under-performance should be shared between company
and customers. A second was concerned that over reliance on market mechanisms might
distort incentives, particularly if other market participants indulged in ‘gaming the system’.
Another stated that an optimal system would not necessarily mean 100% performance

All three managers thought that incentives were very powerful, and in one case a lot of work
was put into maximising revenue against the performance indicators. However one manager
referred to the standards as “constraints which aren’t always binding”, raising the possibility
of choosing to breach the standards. Another respondent made it clear that response to quality
standards was closely related to the image of the company; they would seek compliance, but
not necessarily a good comparative rank, in a measure which they felt did not have high
importance to consumers and was not a prime corporate objective. This suggests that
companies do not respond ‘blindly’ to the regulators’ incentives. One manager said that the
company would sometimes take the reputational risk and pay compensation rather than have
sufficient reserves to meet the standards.

All three companies are analysing the effect of not meeting some of the targets and responding
to the incentives. Companies are clearly sensitive to the standards set, but the direct financial
penalty is often less important than reputational effects. Preserving reputation emerged as an

24
extremely strong incentive in all the industries, making response to these mixed and
inconsistent messages difficult for the companies. Such reputation might have effects in other
markets, or more direct financial effects. This was linked directly to privatisation, because
private companies were more exposed than those in the public sector. Another respondent felt
that the main effect of privatisation had been the freedom from public borrowing constraints
and the ability to increase investment, and so quality levels. While the third saw a phased
attitude to quality, moving from disinterest, through an initial emphasis on cost cutting to a
more accepting and balanced view of quality issues. (This pattern fits well with the stages of
privatised companies described by Coen et al, 1998).

Performance on measures of service quality, especially where these are comparative, may have
financial consequences either in other markets (particularly overseas) or directly through a
deteriorating relationship with the regulator. There is serious doubt amongst managers about
whether the current standards reflect a balance between the costs and value of producing higher
quality, and that the standards themselves are set higher than customer preferences would
suggest.

6. Conclusions

In this paper we have explored service quality standards set by the economic regulators, and
the response of the companies, in areas where common distribution systems mean that
individual customers cannot choose a service level and price they will pay.

Our research shows that while the economic model is clearly useful in guiding decisions of
regulators and responses of companies, the regulatory arena incorporates much broader
considerations for both parties. The regulators operate within a much broader political context,
while the companies are concerned not only with the politics of regulation, but the spin-off in
terms of profitability in other (unregulated) markets. Regulators need to carry both public and
political opinion with them in determining service quality levels, even when they suspect that
this would not fit the strict economic model of balancing willingness to pay against
improvements in service.

25
Both energy and water regulators are now implementing comparative performance measures
which automatically reward or penalise companies according to their performance on a pre-
ordained range of measures. Both are still in the ‘caution’ stage with respect to data,
recognising the problems of accurate measurement, and with ambitions to double the financial
impact of their schemes if more reliable data can be guaranteed. This suggests an underlying
faith in comparative statistics where these can be used.

The standards have proved a powerful incentive for the companies concerned, who have
responded by improving their performance in almost every measure, particularly those which
are not subject to the vagaries of the physical climate. Improvement has been achieved largely
by those who were relatively less good to start with, i.e. by them moving towards the ‘best
practice’ frontier, rather than by movements in the frontier itself. This is despite companies’
own views that some target levels (e.g. speed of telephone response) are ‘too high’ in the
economic sense that its cost (ultimately to consumers) exceeds their valuation of the
improvement. Such ‘over-provision’ of quality is the same result as Baron (1985) found with
separate regulators, but occurs here in the case of a single regulator. The incentives for the
companies are not merely the immediate financial consequences in the regulated context, but
their implications in other markets, particularly those in the overseas utility markets, where
privatisation is offering substantial opportunities for regulated UK firms. Such spill-over
effects are similar to those found in procurement by Collie and Hviid (2001).

This ‘multiplier’ effect suggests that regulators should be cautious in setting targets, since it is
not necessarily their own rewards which will dominate the effect on companies. In particular,
the questions about the robustness of data in the overall performance assessment in water and
the Information and Incentives Project in electricity distribution are somewhat worrying. The
evidence presented in this paper suggests that it is the extraneous rewards, as much as the
direct ones, which will motivate companies. The total penalty for poor performance is greater
than that imposed by the regulator, so the company will ‘over-perform’. If the costs of this
‘super quality’ are recouped within the price cap, consumers in the regulated market are
effectively paying for the companies’ marketing activity elsewhere. This suggests the
possibility of perverse incentives from such measures, and the need for vigilance by regulators
in passing on higher costs from better service in the prices allowed within the regulated market.

26
Particular instances of such perverse incentives emerged from the research. Both regulators
and companies are concerned about incentives which reward current availability, especially
where there was a need for renewal of infrastructure which will take it temporarily out of
service in the interests of its long-term reliability. Similarly, using indicators of relative
performance might discourage co-operation, for example by one company helping to restore
power supplies for another in the case of localised adverse weather conditions. The regulators
are well aware of these issues, but it does suggest that the scope for ‘unintended consequences’
goes well beyond the direct impact of any scheme which they devise.

More generally this shows the difficulty of pinpointing particular measures which are designed
to capture quality of consumer service, even within a single regulator. Reconciling other
regulators’ objectives, for example health and safety and environmental, with the economic
regulator’s drive for efficiency and low prices, poses different but related issues. These have
been particularly sensitive in the light of recent fatal railway and gas accidents in the UK. The
research also highlights the power (for good or ill) of political influences on regulated
industries, a particular irony since removing business from such influences was one of the
justifications for the original privatisation process. Regulation is now entering a new stage in
the UK, following an initial phase of privatisation, and a second of establishing the immediate
post privatisation regulatory adjustments. Many commentators believe that the early squeeze
of profits and downward pressure on costs, which yielded more efficiencies than many had
anticipated, is now exhausted, so that increasing attention will be paid to service and other
measures of quality and output. While regulation is built on a fairly straightforward economic
model, this research shows that the process is more complex on both the regulator and
company side because of the broader political and world market context. This complicates the
regulator’s task in setting targets and penalties, where they need to identify not only consumer
trade-offs but also the full incentives which influence companies.

27
 References
Baldwin, R and M. Cave, 1999, Understanding Regulation: Theory, Strategy and Practice,
Oxford University Press

Baron, D.P., 1985, Non co-operative Regulation of a non localised externality, RAND Journal
of Economics, 16

Brigham, B. and L. Fitzgerald, 2001, Controlling managers and organisations: The case of
performance measurement in a regulated water company, Centre for Management under
Regulation Research Paper 01/01, p22, University of Warwick

Burchell S., Clubb C., Hopwood A., Hughes J. and Nahapiet J. (1980). "The roles of
accounting in organisations and society", Accounting, Organisations and Society, Vol.5, No.1,
pp.5-27

Chau, V.S., 2002, Squeezing Hard to Improve Quality: Evolution of Customer Service
Performance Measures in UK Network Industries, Centre for Competition and Regulation
Working Paper 02-03, University of East Anglia, Norwich

Coen, D., D. Currie, M. Siner and P. Willman, 1998, Regulatory Organisations and Regulatory
Effectiveness in Privatised Companies, LBS Regulation Initiative discussion paper

Collie, D and M Hviid, 2001, International Procurement as a Signal of Export Quality,

Economic Journal, 111, 2001, 374-390.

Eccles, R.G. and P. Pyburn, 1992, Creating a Comprehensive System to Measure Performance,
Management Accounting, October, pp 28-33

Euske K.J., Lebas M.J. and McNair C.J. 1993. "Performance management in an international
setting, Management Accounting Research, Vol.4.

Fitzgerald L., Johnston R., Brignall S., Silvestro R. and Voss C. 1991. Performance
measurement in service businesses, London: CIMA

Harrington, W., 1998, Enforcement leverage when penalties are restricted, Journal of Public
Economics, 37, 29-53

Holmstrom, B and P. Milgrom, 1991, Multitask principal-agent analyses: Incentive contracts,

asset ownership and job design, Journal of Law, Economics and Organization 7:24-52

Jensen, M., 2002, Value Maximization and the Corporate Objective Function, forthcoming in
Unfolding Stakeholder Thinking, edited by Joerg Andriof, Sandra Waddock, Sandra Rahman
and Bryan Husted, Greenleaf Publishing.

Johnson, H. and R.S. Kaplan, 1987, Relevance Lost: The Rise and Fall of Management
Accounting, Harvard Business School Press, pp 253-263

Kaplan R.S. and D.P. Norton 1993. Putting the balanced scorecard to work, Harvard Business
Review, September-October, pp.134-147

28
Littlechild, S.C., 1983, Regulation of British Telecommunications Profitability, London:
HMSO

Livernois, J and C.J. McKenna, 1999, Truth or consequences Enforcing pollution standards
with self-reporting, Journal of Public Economics, 71, 415-440

Markou, E. and C. Waddams Price, 1999, UK utilities: past reform and current proposals,
Annals of Public and Co-operative Economics, 70, 3, pp 371-416

Merchant, K.A., 1985, Budgeting and the propensity to create budget slack, Accounting
Organizations and Society, 10, (2), 201-10

Nanni A.J., Dixon J.R., and Vollmann T.E., 1992. Integrated performance measurement:
Management accounting to support the new manufacturing realities, Journal of Management
Accounting Research, Vol 4, Fall

NAO, 2002, Pipes and Wires, April, London

OFFER, 1993, “The Supply Price Control: Proposals”, Office of Electricity Regulation, July

OFFER, 1994, “The Distribution Price Control: Proposals”, Office of Electricity Regulation,
August

OFFER, 1999a, "Quality of Supply: Attitudes of Business and Domestic Customers"

OFGEM, 1999b, “Review of Public Electricity Suppliers 1998-2000 – Distribution Price

Control Review: Consultation Paper”, Office of Gas and Electricity Markets, May

OFGEM, 2000a, “Information and Incentives Project – Output Measures and Monitoring
Delivery Between Reviews: Initial Proposals”, Office of Gas and Electricity Markets, June

OFGEM, 2000b, “Review of Transco’s Price Control from 2002: Update Paper”, Office of Gas
and Electricity Markets, November

OFGEM, 2001a, Review of Transco’s Price Control from 2002, Final Proposals Office of Gas
and Electricity Markets, September

OFGEM, 2001b, Transco’s National Transmission System Operator incentives 2002-7, Final
Proposals, Office of Gas and Electricity Markets, December

OFGEM, 2001c, “Review of Transco’s Price Control form 2002: Draft Proposals”, Office of
Gas and Electricity Markets, June

OFWAT, 1992, “The Cost of Quality: A Strategic Assessment of the Prospects for Future
Water Bills”, Office of Water Services, August

OFWAT, 1993, “Setting Price Limits for Water and Sewerage Services: The Framework and
Approach to the 1994 Periodic Review”, Office of Water Services, November

29
OFWAT, 1997a, “The Business Planning Process, Customer Consultation and Information
Requirements for the 1999 Periodic Review: A Consultation Paper”, Office of Water Services,
July

OFWAT, 1997b, “The Proposed Framework and Approach to the 1999 Periodic Review: A
Consultation Paper”, Office of Water Services, June

OFWAT, 1998, “Prospects for Prices: A Consultation Paper on Strategic Issues affecting
Future Water Bills”, Office of Water Services, October

OFWAT, 1999, Final Determinations: Future water and sewerage charges 2000-05, Office of
Water Services, November

Ogden, S., 1997, Accounting for Organisational Performance: the construction of the customer
in the privatized water industry, Accounting, Organisations and Society, 6, 529-556

Rees, R., 1984, A Positive theory of the Public Enterprise in M. Marchand, P. Pestieau and H.
Tulkens (eds) The Performance of Public enterprises, Amsterdam: North-Holland

Rodriguez-Ibeas, Roberto, 2002, Regulatory Enforcement with Discretionary Fining and

Litigation, Bulletin of Economic Research, 54:2, 105-118

Sherman, Roger, 1989, The Regulation of Monopoly, New York: Cambridge University Press

Shleifer, A., 1985, A theory of yardstick competition, Rand Journal of Economics 16: 319-27

Spence, M., 1975, Monopoly, Quality and Regulation, Bell Journal of Economics, 16, 417-29

Tilley, B., and T Weyman-Jones, 1999, Productivity Growth and Efficiency Change in UK
Electricity Distribution, A New Era for Energy: Price Signals, Industry Structure and
Environment , 1999 British Institute of Energy Economics Conference , St John's College,
Oxford, 1999, pp 1-10

Waddams Price, 2000, Efficiency and Productivity Studies in Incentive Regulation of UK

Utilities, Revista del Rosario (Colombia), 3,2, pp 11-24

30