FINANCING THE SCOTTISH WATER AND SEWAGE INDUSTRY
JEANETTE FINDLAY April 2004
Contact: Department of Economics Adam Smith Building University of Glasgow Glasgow G12 8RS
E: J.Findlay@socsci.gla.ac.uk T: +44 141 330 4974/4658 F: +44 141 330 4940
Introduction
The remit for this research, is outlined in the research brief, `Financing the Scottish
Water and Sewage Industry', December 2003. Specifically, the STUC is seeking to
develop policy on the following three areas:
· Alternatives to the current structure of water charges
· Investment and the Treatment of debt
· Regulatory approaches to efficiency.
In order to assist the discussion on these themes I have provided below a detailed,
critical analysis of the approach taken by the economic regulator of Scottish Water,
the Water Industry Commissioner for Scotland (the WIC) in his Strategic Review of
Charges 2002-6 and supplemented by the Cost and Improvement Report 2002-3 along
with the Investment and Asset Management Report 2002-3. I have also taken into
consideration a number of reports from the water quality regulator for the industry,
the Drinking Water Quality Regulator for Scotland, and other bodies such as the
Scottish Environmental Protection Agency and the Water Customer Consultation
Panels whose remit is also to monitor quality on behalf of customers served by
Scottish Water.
The Regulatory Regime
The conceptual and practical difficulties of regulation, both of public sector industries
and other (usually privatised) natural monopolies, are well understood by
economists.1 The main problems which arise under this heading are `how to induce
firms in non-competitive markets to act in a way that is compatible with social goals'
(Train, 1991). In other words to ensure that firms/industries act in the public interest
when their activities are not constrained by competitive pressures in such a way as to
increase efficiency and to keep prices at a socially acceptable level.
Productive efficiency in this regard can be understood as producing a greater level of
output/quality using the same inputs or, alternatively producing the same level of
output/quality using fewer inputs. It does not involve any consideration of costs since
these are related to input prices which are not entirely under the control of the
producer. Productive efficiency also takes on a dynamic character in the sense that
optimal regulation should induce cost-reducing investment by the firm. Allocative
efficiency is defined as circumstances where the product price is equal to marginal
cost or as near to marginal cost as can be achieved while still allowing the firm to
remain viable. In any event no excess profits should be made. The regulator's
problem in these circumstances is to gather cost information from the firm in order to
determine rates of price increase which will keep excess profits to a minimum,
maintain the viability of the firm/industry and encourage investment designed to
reduce costs in the future. For private or privatised industries regulation has always,
in the UK context, been intended as a temporary measure. The gradual introduction
1 See Train, 1991, Armstrong & Sappington, 2003, Crew & Kleindorfer, 2002
of competition was always intended to provide the ultimate replacement for the
regulator.2 In terms of the regulation in the public sector there is clearly no issue of
excess profits, and in some cases there are private sector competitors, but the
economic regulator still has a role in promoting dynamic efficiency and monitoring
prices for customers.
In most industries the issue of quality is important, but none more so than in the water
industry which has among its remits the production and supply of safe, clean drinking
water for all citizens and the safe removal and processing of sewage from homes and
businesses. The WIC has acknowledged this fundamental role of the water industry
in his Strategic Review where he states
cutting costs and makingefficiencies' are not the same thing even though they
are often understood to be synonymous. A true efficiency is achieved only when a
service or product of equal utility is delivered or created for less cost. It is not in
the customer interest to cut costs in any way that will have an adverse impact
upon the service that is provided to the customer. Nor is it acceptable to take
3
short cuts with safety, public health or the environment.'
In Scotland the quality of the water is overseen by the Drinking Quality Regulator for
Scotland (DWQRS), a post which was created by the Water Industry (Scotland) Act
2002 to `provide an independent check that Scottish Water is complying with the
drinking water quality regulations'4 In addition the Scottish Environmental Protection
Agency (SEPA), which was established by the Environment Act in 1995 is
responsible for regulating all activities that may affect the environment including
water abstraction, activities that may pollute water' and thestorage, transport and
2 OECD, The role of competition policy in regulatory reform, 2002, OECD 3 WIC, 2001, p39 4 www.dwqr.org.uk
disposal of waste'.5 Finally there are the five Water Customer Consultation Panels,
also created by the Water Industry (Scotland) Act 2002 to represent the interest of
domestic and non-domestic customers of Scottish Water. Their task is to
- represent the views and interests of the customers of Scottish Water, and each Panel must make
- reports and recommendations on any matter they consider relevant to the interests of customers6.
A review of the publications and statements of the above bodies indicate that they
have a number of serious concerns about both the quality of the outputs which
Scottish Water produce (both in terms of water quality and regulatory information)
and the role of the WIC. Many of these concerns relate to the quality of the existing
assets, much of which is in need of refurbishment or replacement. In its submission
to the Scottish Parliament Finance Committee, in September 2003, the Water
Customer Consultation Panels' Convenor, Ian Smith, reported customer concerns
over `delays and difficulties in obtaining increased service capacity and infrastructure
to meet demands' and went on to raise the issue of the importance to customers of
`sufficient flexibility at a local level in its (Scottish Water) investment and operational
activities'. In his first Annual Report published in October 2003, Tim Hooton
(DWQRS) outlined a number of disturbing indicators in relation to water quality.
Although in 2002 the number of samples which failed to comply with the relevant
standard7 had fallen compared to 2001, the absolute number was described by Mr
Hooton as significant' which indicateda need for improvements'. Within these
figures there is evidence that the number of microbiological failures actually
5 www.sepa.org.uk 6 www.watercustomer.org 7 The Water Supply (Water Quality) (Scotland) Regulations 1990
increased. The presence of both coliforms and faecal coliforms in drinking water
samples increased throughout the year although in both cases this rise went against a
long-term downward trend. There were a number of incidents during the year in
North Ayrshire and Glasgow which led to bottled water being distributed and/or a boil
notice being issued. The DWQRS found it necessary to report Scottish Water to the
Procurator Fiscal (PF) in relation to the North Ayrshire incident although the PF
subsequently decided to take no action due to the remedial work which Scottish Water
undertook in the interim period. A number of the incidents throughout the year were
precipitated by extreme weather conditions ie heavy rainfall. The report highlights
the need for Scottish Water to `invest in robust equipment and processes that can cope
with the vagaries of the Scottish weather' a point I shall return to later. It is
important to understand that insufficient investment over a period of years is likely to
lead to deterioration in drinking water quality. However this deterioration will not
necessarily show up immediately. It is presumably for this reason that the DWQRS
found it necessary to refer to `increasing pressure.... to invest more in its
infrastructure..' and to issue a warning to the effect that
`Improvements in efficiency however, must not compromise drinking water quality and public health'
Scottish Water was also reported to the Procurator Fiscal at Hamilton by SEPA in
2002 in relation to a charge of causing pollution to the Kittoch Water and its parent
river, the White Cart Water by allowing effluent to enter the water from the Philipshill
Sewage Treatment Works. Following the case, in which Scottish Water were fined
£10,000, SEPA spoke in the press of the `need to ensure that their treatment works do
not fall below the required standards'. Scottish Water gave assurances that major
improvements in the facility would be carried out.
In conclusion, it is clear that there are no disagreements among the agencies,
including the WIC, in relation to the quality problems which are a feature of the water
industry in Scotland. In addition, there appears to be no major disagreement in
relation to a fundamental cause of these problems, that is, years of underinvestment in
the capital stock. All of this has to be seen in the context of increasingly stringent
water quality and environmental controls which are coming into force in the near
future. Without investment these changes will merely result in increasing the quality
gap.
Econometric Analysis
The office of Water Industry Commissioner for Scotland was created by Part II of the
1999 Water Industry Act and came into being on 1 November 1999.8 In taking up his
responsibilities as the economic regulator for the water industry in Scotland, Alan
Sutherland (WIC), made it clear that
`Ultimately the best and only way of promoting customer interests in a public sector model is for the customer regulator to improve the economic efficiency of the industry, and thereby the value for money generated' 9
In attempting to achieve these improvements in efficiency, the WIC looked to the
practices of the economic regulator of the privatised water companies in England and
Wales, OFWAT. In his own words he followed the lead of OFWAT in that he `relied
8 www.watercommissioner.co.uk
9 WIC, Commissioner's Corporate Plan, July 2002
heavily' on the incentive effects of `comparative competition.10 He further described
his reliance on comparisons with companies in England and Wales as the
`cornerstone' of his Review.11 While introducing competition is a recognised practice
in the UK to regulate utilities12, it is usually done in the context of privatised
industries. The WIC's reliance on comparisons with other companies to induce
increased efficiency from the management of Scottish Water, which is a public body,
has no basis in economic theory and his evidence13 that competition between the three
Scottish water authorities (prior to the merger which created Scottish Water) led to
improved standards, no longer holds in a single-authority context. In addition, his
assertion that
`Management should, after all, want to show all stakeholders how good they are'14
is not a behavioural assumption I have seen appear in any economic study in any
context. The absence of any convincing behavioural assumptions leave this part of
the WIC's analysis open to question.
For the purposes of measuring the `efficiency' of the Scottish industry the WIC used a
suite of eleven econometric models (equations), which were developed for OFWAT
by Professor Mark Stewart of the University of Warwick. These models, together
with an analysis of unit costs for certain activities, are designed to explain the main
determinants of various elements of operating and capital expenditure for the water
10 WIC, 2001, Section 2 Chapter 7 p72 11 ibid p71 12 OECD, 2002 op cit 13 WIC, 2001 op cit, p72 14 ibid p77
industry including distribution costs, resource and treatment, power, business
activities, sewerage network and infrastructure.
The data collected by OFWAT from the water and sewerage companies in England
and Wales are used to estimate the models and provide a method of comparing the
efficiency of the companies in various aspects of their activities. While it is not my
purpose here to question the usefulness of the models in the English and Welsh
context a number of comments can be made at the outset. First, the models are
designed to capture (or explain) the main elements of costs in a statistical sense. They
are not based on engineering principles and are therefore very specific to the set of
companies and market structure for which they were developed. Second, OFWAT
has acknowledged certain flaws in the models. In particular, it allows as a matter of
practice certain `special factors'15 to be accounted for in an ad hoc way which is an
sensible acknowledgement of the limits of any econometric exercise which is used for
policy reasons. Second, OFWAT acknowledges that limitations in the data which are
collected also explains some of the differences in the efficiency scores between the
English and Welsh companies.16 Third, OFWAT has accepted the need to make
changes to the original set of models on the basis of challenges from external bodies.
As a result of these challenges, OFWAT recently stated that is has
..made adjustments to water service operating and capital maintenance expenditure to ensure consistent treatment of leakage control costs between companies. We have done this to address concerns that costs used in our analysis are influenced by company specific accounting policies...we have also reduced the modelling residuals, that is the difference between actual costs and the costs predicted by the models, to take account of errors in the data and in our statistical
15 OFWAT, 2003 p 19 16 ibid, p 4
process. We have adjusted the water residuals by 10% and the sewerage residuals by 20%.17
This last point highlights another problem with the use of econometric models.
OFWAT has treated the residuals of the model (as explained above) as being a
measure of inefficiency. This is not a valid assumption unless one can be absolutely
sure that the data is correct and that the structural form of the model is as good as it
can be. Even under those criteria it is still a very strong assumption that there is no
other possible reason for the unexplained element of costs.
Fourth, it cannot be stated with any degree of conviction that the prime motivation for
the undoubted increases in efficiency which have taken place in England and Wales
(albeit with some considerable teething troubles and some remaining problems) was
the regulatory policies derived from these statistical models. It is at least possible, if
not likely, that it was, and continues to be, the influence of actual competition which
has led to the much reduced costs of the English and Welsh water companies. In
addition the debt write off and lengthy period of substantial investment has also led to
the differences that now exist between Scotland and England I will return to this
point later. This fourth point is not intended to be an argument in favour of
privatisation of water supplies. The arguments against such a move are many and
valid but they are beyond the scope of this report.
17 ibid p20
Application of the models to Scotland
However it is under this heading that the philosophy and practice of the WIC is most
vulnerable to criticism. While the suite of models may well have served a purpose in
the environment for which they were created, the wholesale transfer of them to an
entirely different set of circumstances in Scotland appears, at the very least,
questionable.
The models in question are what are known as cross-section models. That is they
rely on data from a number of different sources (ie companies) for a single point in
time. It would not be possible to construct such a model from scratch for Scotland
because there are not a variety of companies producing these services. Indeed,
OFWAT is forced into departing from the standard models in the case of sewerage
services in England and Wales because there are only 10 regulated sewerage
companies. To overcome this problem the model is not based on data from separate
companies but on data from individual large sewage treatment works (regardless of
ownership). OFWAT again admit that this is `not ideal'.18 In the case of Scotland,
even the pre-merger situation would not have provided sufficient valid data to
construct these models.
Estimation of the models, using real data, produces estimated parameters for each
equation. These parameters are constants and serve as measures in some sense of the
effect of each explanatory variable on the dependent variable, in this case cost, which
is being estimated. The parameters represent the structure of the model and can be
18 OFWAT, 2003 op cit
used for predictive purposes as long as one can be sure that the underlying forces
which ultimately determine cost is unchanged. In fact, what the WIC has done is to
take the parameter values estimated from English and Welsh data and then use the
data from Scottish Water for the explanatory variables19. The assumption that this is
a valid process is a heroic one. Indeed, it has been pointed out before that the Scottish
industry is operating in such different circumstances that to make this assumption
seems almost foolhardy. The WIC indicates in various publications that he is aware
of the relevant differences and has taken them into account. However, my
examination of the WIC's owns publications does not bear this out.
The differences that have been referred to previously include geography, topography
and population density. I cannot find evidence that these have been adequately
accounted for in the models used by the WIC. Adjustments are made for `special
factors' identified by Scottish Water. These are applied at the end of the process and
in an aggregate sense ie adjustments are not made at the level of the individual
equations.
In relation to geography, topography and, in particular, the variety of sources used,
Scottish Water20 have argued persuasively that the form of the data demanded by the
WIC in the Annual Returns fails to account for the variety of sources which SW has
to extract water from. In particular the classification of burns does not enter into the
Water Resources and Treatment Model (which only uses the English and Welsh
classifications). This model is likely then to underestimate the true costs to Scottish
19 This procedure was departed from in the case of small sewage treatment works. In the case of these works, of which there is a large number in Scotland, unit cost estimates were derived. 20 Overview Document, Annual Return 2002-03, Scottish Water p24
Water and to overestimate the efficiency gaps between Scottish Water and any
English or Welsh comparator.
In the Water Distribution model, the explanatory variable (length of mains greater
than 300mm diameter divided by total mains length) is used as a proxy for
urbanisation. However, this is not an adequate proxy in the Scottish setting where the
distribution of the population is uneven. In fact, the Competition Commission in
reviewing a complaint from Mid Kent Water in 2000 mentions this factor and
suggests that more unevenly distributed population is likely to incur `significantly
higher costs'.21 If this is held to be true in the far more homogeneous English setting
then it is likely to be more so in the context of Scotland.
The other significant difference between Scotland and England and Wales lies in the
very different market structure. The water industry was privatised in 1985 and was at
that time given a substantial subsidy from general taxation in the form of a write-off
of all debt, a point emphasised in their evidence to the Scottish Parliament Finance
Committee, by the Water Customer Consultation Panels.22 This together with other
government grants totalled £6.6 billion. Since that time the number of companies
involved in the supply of water and/or sewerage services has risen markedly resulting
in a huge increase in competitive pressures23. A crucial point to note here is that the
time period over which any improvements in efficiency and quality came about in
England and Wales far exceeds the time period over which the WIC expects the same
21 Competition Commission, 2000
22 www.watercustomer.org 23 There are now 22 companies in England and Wales. 10 provide water and sewerage services and 12 are water only
improvements to arise in Scotland. It is clear from examination of the figures
produced by OFWAT that costs initially rose for some years after privatisation before
they began to fall. Total operating costs rose in the initial years, reaching a peak
around 1993 and falling fairly steadily since then.24
Almost twenty years of privatisation and competition with considerable borrowing to
fund increased investment has created a market and infrastructural landscape which
differs in a great many ways from Scotland. I can see no evidence that the WIC has
taken any of these factors into account in the adaptation of the expenditure models.
The issue of the actual comparators that are used also raises some questions. It
appears that in some cases the comparator is not an actual company but a hypothetical
company arrived at by combining the costs for distinct areas of activity from different
companies. Scottish Water is therefore being asked to meet standards not actually
met by any single company in England or Wales, albeit that the comparator in each
case is the one which generates the lowest gap.
Finally, despite the acknowledgement by the WIC in numerous statements to the
effect that efficiency measures have to be adjusted for quality25, I cannot find any
evidence in the methodology used that this has been explicitly adjusted for or taken
into account. There appears to be a general reliance on the evidence that quality
across all activities is, on average, better in England and Wales. As far as I can
determine from the published output of both the WIC and OFWAT there is a huge
variation in both practice and quality among the private water and sewerage
24 OFWAT, 2003 op cit 25 WIC, 2003, p1
companies and any genuine quality adjustment would require a detailed comparison
of quality measures between Scottish Water and any comparator company.
Investment and Treatment of Debt
Leaving aside the benefits of debt write-off enjoyed by the privatised industry in
England and Wales, there are issues arising from the projections by the WIC in
relation to debt funding of investment for the future in Scotland.
In the Strategic Review, the WIC referred to the fact that `The Scottish industry has
not invested sufficiently to meet environmental standards and to maintain its assets
properly' 26. The WIC also makes reference to the acknowledgement by the Scottish
Executive of historic underinvestment'27 and concludes that..levels of investment
in Scotland should be on balance higher than in England and Wales'.28 Yet in his
Investment and Asset Management Report 2002-03, published in 2004, the WIC
adjusts the investment figures for `relative efficiency' and concludes that, in fact,
investment (actual and planned) over the period 1996-2006 is marginally higher than
England and Wales and that there is no
` ..evidence to support the contention that there is a significant backlog of investment in Scotland relative to that in England and Wales..'29
26 WIC, 2001, op cit p6 27 ibid, p7 28 ibid p7 29 Invesment and Asset Management Report 2002-03, WIC, 2004
The WIC acknowledges that the data on which this conclusion is based is not directly
from the three authorities, but nevertheless concludes that the key issue is the
efficiency with which Scottish Water plans and delivers capital expenditure.
Another issue of contention between the view taken by the WIC and others relates to
the state of the assets which are currently in use in Scotland. In his Investment and
Asset Management Report 2002-03, the WIC concludes that
`..with the possible exception of water mains, the condition of assets in Scotland remains very similar to those in England and Wales.'
Scottish Water in their Overview Document for the Annual Return 2002-03 argue that
the way in which the data is collected `does not adequately reflect the true condition
and performance of the asset stock'.30 Reference is made to the level of leakage as an
indication of the poor state of the assets. Without a detailed independent analysis of
both the condition of the stock and whether or not it is `fit for purpose' it is
impossible to conclude which view is correct. However, it seems clear that the levels
of investment required to deliver improved quality and environmental standards may
differ substantially from those agreed in the Central Option of the Quality and
Standards II investment programme. Tim Hooton, the DWQRS, in his evidence to the
Finance Committee Enquiry, indicated that in his opinion
`..there is still a significant backlog of under-investment that needs to be addressed. The condition of Scottish Water's assets still varies hugely between sites. The poor performance of some assets arising from age or design continues to represent a high risk of failure to meet regulatory
30 Scottish Water, 2003, op cit Section 1.7, p6
standards compared with the best performing assets. Many projects aimed at levelling out such inequality were not included in the Quality and Standards II middle option.'
A further feature which appears not to have been incorporated into the Central or
middle option of the investment programme is the question of the degree of
automation which would be required to replace the amount of manpower at the rate
which has been proposed by Scottish Water in response to the WICs demands for
savings. Almost all of the cost savings made so far and commended by the WIC in
his Costs and Performance Reports relate to the reduction in staffing levels, much of
which arose as part of the merger of the three authorities. Further reductions in the
labour force, without adequate investment to introduce automation bring, without
question, a risk of compromising the quality of service already provided. The nature
of the skills and experience already lost to the industry, which involve a great deal of
tacit (ie non-codified) knowledge in operating particular equipment in particular
terrains, cannot be easily replaced. Concerns regarding quality must therefore be
raised by the lack of consideration of this issue in the investment plan.
A further problem in relation to investment is the level of debt that the WIC assumes
is necessary to carry out this programme. Again this is something that the Water
Customer Consultation Panels, representing both domestic and non-domestic
customers, has raised in its evidence to the Scottish Parliament Finance Committee.
For investments which are likely to bring efficiency and quality improvements over
lengthy periods it does not seem unreasonable for the cost of that investment to be
spread over current and future customers in the form of long term debt. Indeed the
amount of debt-funding of investment in England and Wales currently stands at
around 35%. The WIC's plans, outlined in his review, to fund investment entirely out
of charges by 2005/6 appear to have no economic justification and also raises
questions about the desirability of raising some investment funding from general
taxation given the economic and social externalities involved in having a safe, clean
water provision and sewerage system. However, subsequent statements from the
WIC to the Finance Committee appear to suggest that current debt levels will be
maintained in real terms over this period. Notwithstanding this reversal however, the
method which the WIC used to calculate the amount of debt which Scottish Water
could sustain has been questioned in a paper again presented to the Finance
Committee in its recent enquiry.31 The technical merits of the paper are beyond my
expertise but it should be noted that whilst the Finance Committee ultimately rejected
the argument, a minority report by three of the MSPs on the committee found their
arguments to be valid.
Charging
In addition to the debate about what proportion of investment funding should come
from charging there has been a lengthy debate about the merits of charging in its own
right.
In England and Wales there are a set of principles which are supposed to underpin any
tariff and these are:
31 Cuthbert & Cuthbert, 2003
· Fairness and Equity
· Sensible Incentives
· Simplicity and Comprehensibility
The principles involved in a sensible and fair charging system, not to mention the
issue of affordability, merit research projects in their own right but it is possible to
make some comments here regarding some of the key issues.
It appears fair to state at the outset that all fixed costs which relate to assets with a
long life should not be borne by existing customers only. It would also be socially
divisive and politically unacceptable to charge domestic customers in different parts
of the country at different rates in relation to costs. This would make some rural
communities unviable and give incentives for migration patterns which are not
warranted for other economic and social reasons.
In terms of non-domestic customers, it does seem sensible to set in place a system of
charging which signals to heavy users of water and sewerage services to locate in
areas of low cost or to incorporate higher costs into their own charging system. A
harmonisation of charges for non-domestic customers would offer a perverse
signalling mechanism which would lead to economic in other sectors of the economy.
Conclusion
In summary, my review of the methods and policies of the Water Industry
Commissioner for Scotland suggest that he is open to the charge of placing undue
emphasis on the history and practice of the water industry in England and Wales; that
the behavioural assumptions underlying his approach are questionable; that his
methodology in assessing efficiency is flawed; that the timescale over which the
industry is being asked to introduce fundamental change is too short and that the
investment programme and job losses that are proposed are likely to lead to severe
problems in the provision of a quality water and sewerage service in Scotland.
In relation to the methodology for assessing efficiency, the information upon which
the view expressed here is based is taken entirely from sources published by the WIC
and OFWAT. A longer and more detailed study than was permissible in the present
case would be necessary to make more detailed criticisms and to suggest alternative
ways in which the efficiency of Scottish Water might be assessed.
REFERENCES
Armstrong and Sappington (2003) `Recent Developments in the theory of regulation' http://www.econ.ucl.ac.uk/downloads/Armstrong/reg.pdf
Competition Commission, (2000) Report into Mid-Kent Water plc, ISBN 0-11- 702527-5
Crew, M.A. and Kleindorfer (2002) "Regulatory Economics: Twenty Years of progress?", Journal of regulatory Economics, 21(1) 5-22
Cuthbert J and Cuthbert, M, (2003) Did Flaws in the Application of Resource Accounting and Budgeting Distort the Strategic Review of Water Charges in Scotland, Evidence to the Finance Committee of the Scottish Parliament
Lewis, T.R. and D.E.M. Sappington) (1989) "Countervailing Incentives in Agency Problems", Journal of Economic Theory 49(2) 294-313
OECD, The role of competition policy in regulatory reform http://www.oecd.org/dataoecd/46/37/2766143.pdf
OECD "Regulatory Reform in the Telecommunications Industry" http://www.oecd.org/dataoecd/46/30/2766201.pdf
OFWAT, (2003) Water and sewerage unit costs and relative efficiency 2002-3, p 19
Train, K E, (1991), Optimal Regulation, MIT Press
Vogelsang, I. (2002) "Incentive Regulation and Competition in Public Utility Markets: A 20-year Perspective", Journal of Regulatory Economics, 22(1) 5-27
Water Industry Commissioner for Scotland, (WIC),(2001) Strategic Review of Charges 2002-6
WIC, (2003), Cost and Improvement Report 2002-3 and
WIC, (2003) Investment and Asset Management Report 2002-3 www.dwqr.org.uk
www.sepa.org.uk