Further supplementary memorandum from
Water UK
CREATING AN AGENDA FOR ACTION ON CAPITAL
MAINTENANCE
A report of the Water UK think-tank to share
and develop strategies
SUMMARY
The outcome of the last periodic review has
left a sizeable gap between the positions of companies and the
regulator. Companies had argued that a spend of around £8
billion was needed to ensure that the asset base delivered the
level of service required for the longer term. Ofwat felt that
a spend nearer £6.3 billion would achieve that.
The debate in the think-tank exposed differing
views as to how to measure customer serviceability (ie what the
customer is happy with) and what proxies can effectively determine
the relationship between spending money on maintenance and the
achievement of customer serviceability.
Without adopting more rigorous techniques, such
as risk assessment and econometric modelling, water companies
could lay themselves open to scrutiny by their shareholders and
their customers in terms of failing to manage their assets effectively.
Ofwat would obviously have something to say about such a situation.
However, sophisticated methods require sophisticated
data which can be expensive to collect and analyse. To embark
on a major exercise to put in place such techniques across the
whole industry capital asset base would take a significant amount
of time and could be expensive. Indeed, part of the initial analysis
has to be the determination of which assets lend themselves to
giving an economically viable answer when subjected to these techniques
and some assets just don't warrant such attention.
Secondly, there are some areas which are already
ripe for attention—these could be tackled first and there
are some pragmatic approaches which could be introduced quickly.
Even if they do not provide the best longer-term solution, these
approaches could certainly improve the situation in terms of a
better focused pattern of spend on maintaining the capital assets
and ensuring high levels of customer serviceability.
Ofwat has suggested that it is in the companies'
own business interest to adopt clear well-proven methods of assessing
the expenditure required and provided that is done, they are happy
to discuss, on a company by company basis, the appropriateness
of the methods for identifying the impact on serviceability to
customers.
The industry feels that it would be more valuable
to have industry-wide agreed methods which it is confident would
not only satisfy business and customer requirements as they see
it, but would also be acceptable to Ofwat in the next periodic
review. The draft agenda for action which has been developed as
a result of the day's deliberations is designed to help achieve
these objectives.
AGENDA FOR
ACTION
There are two phases working in tandem with
each other:
— extending the collaborative approach
by organising multilateral discussions between the involved parties
to develop broad agreement on the need for appropriate measures;
— extending the technical/economic
research to develop industry-wide acceptable methods for providing
clearer guidance for the decision takers on how much to spend,
on what and when.
The collaborative approach (ie the water industry
working with other relevant stakeholders):
— explore and develop the definitions
of customer serviceability to get an agreed understanding of what
the customer wants and is prepared to pay for;
— explore and develop agreement as
to which pragmatic measures for assessing asset maintenance requirements
could be developed;
— explore and develop agreement that
new and more sophisticated grading schedules to identify status
of individual assets could be developed;
— agree a restricted range of assets
which lend themselves to a speedy solution in terms of developing
more sophisticated analysis for determining economic expenditure
requirements;
— ensure all stakeholders are involved
and/or kept informed (as appropriate);
— engage the Minister in the debate
as to the relationship between asset management and service provision
and the longer-term issue of stewardship.
The research approach (ie the water industry
in whole or in part, and if appropriate in collaboration with
stakeholders):
— customer needs and preparedness to
pay;
— pragmatic asset maintenance measures;
— risk analysis and econometric techniques.
Water UK intends setting up two work groups
for these two strands and also to continue to facilitate the transfer
of best practice (as it is identified) through forum based activity.
BACKGROUND
Sixty-nine water industry representatives[19]
met in London on 26 July 2000 to discuss together issues arising
from the funding of capital maintenance in the water industry.
The meeting was "closed". This report
summarises the presentations made, the issues raised during discussion
and the common themes and action proposals which emerged at the
end of the day.
The meeting, which drew together representatives
of government, the regulators, the water undertakings and suppliers
to the water industry, was organised by Water UK Information and
Learning and supported by British Water.
The theme of the think-tank was an exploration
of the process of deciding how much capital maintenance was "needed"
in the water industry given the importance of this calculation
in the overall periodic review of water industry prices as carried
out by the Water Regulator (Ofwat). The water industry is concerned
that not enough allowance has been provided in the recent review
(AMP3) to cover all the maintenance that could be needed and is
anxious to ensure that the process of deciding what element is
included for maintenance in AMP4 (the next review in four years'
time) is developed in advance of that review.
CREATING AN
AGENDA FOR
ACTION
Barrie Delacour, Corporate Strategy Director of
Southern Water set the scene
In the current five year period the determination
by Ofwat allows for a spend of £6.3 billion on capital maintenance—approximately
2/7 below ground and 5/7 above ground. Expenditure on these assets
shapes the level of service the water companies provide. However,
while considerable attention has been given to such issues as
quality of water and environmental issues, much less has been
given to the assessment of what capital maintenance is needed
to provide for the necessary levels of customer satisfaction in
the longer term. Companies, in making their submissions to Ofwat
prior to the review, felt that £8.3 billion was needed; Ofwat
felt that £6.3 billion would be adequate. The Ofwat determination
document also shows that the starting point for their judgement
was historic expenditure levels rather than the company plans.
This difference of opinion and the underlying apparent difference
of approach to calculating asset maintenance requirements was
the reason for the think-tank. It provides a springboard for going
forward. It is an opportunity to work towards an acceptable procedure
next time round. Such an approach can only work if all the parties
are involved and engaged early. A proper base can then be developed
by collaborative working.
What Were the Main Differences Between Ofwat's
Procedure and the Water Companies?
Professor Chris Binnie, FREng, Past President
of the Chartered Institution of Water and Environmental Management
Professor Binnie had been commissioned in 1999
by Water UK to review the implications for capital maintenance
expenditure arising out of the draft (subsequently final) determinations
by Ofwat for the period 2000-05. His report itemises the measures
used by Ofwat and by the industry and underlines the strengths
and weaknesses of the various measures (for the full report see
think-tank delegates' pack or log on to the Water UK website at
www.water.org.uk).
His main conclusions were:
— Ofwat used backward looking measures.
What is needed for capital maintenance is a forward looking approach;
— some £7 billion of new assets
had been constructed over the last five years with similar over
the next five years. Many of these were shorter life or had been
designed down to obtain capital efficiency. Effectively Ofwat
had not allowed funding for capital maintenance of these new assets;
— Ofwat measured serviceability using
several criteria. Many of these had little relevance to capital
maintenance needs and new criteria of serviceability needed to
be set up;
— Ofwat serviceability criteria were
failure based. The object of capital maintenance should be to
avoid failure;
— Ofwat had ignored asset condition.
This did have a strong relationship with the need for capital
maintenance and should be one of the criteria for funding;
— there must be doubt whether companies
could meet the financial efficiency targets set by Ofwat without
short termism;
— the water industry is a long-term
industry which must adopt a long-term approach to sustainability.
In his view the recent determination has resulted in a serious
underfunding and will jeopardise the service and funding for future
generations;
— Ofwat letter MD161 suggested an approach
of economic appraisal of capital maintenance and operations, service
to customers and risk and who should bear it. He welcomed this
initiative.
Professor Binnie therefore concluded that it
was important for the industry to work with Ofwat to develop a
wider range of more accurate and pertinent measures in order to
ensure proper capital maintenance spend in the future.
How Can Asset Maintenance Requirements be Predicted?
Two, complementary routes were explored—the
management of risk and econometric models.
The management of risk. Dr Jeremy Lumbers, Chairman,
Tynemarch
Dr Lumbers pointed out that current methods
of assessing asset maintenance budgets in the water industry often
lacked a consequence of failure dimension. Using the remaining
life concept was unrealistic for most asset types because there
is in practice no single determinable future date at which renewal
will be required, and because of the mitigating effect that Opex
can have on failure rate. Capital Maintenance predictions are
highly sensitive to uncertainty in the estimated remaining life.
A risk analysis approach avoids the need to use remaining lives
and allows maintenance requirements to be explicitly linked to
the consequences of asset failure, expressed in terms of a quantified
loss of serviceability.
A number of other factors were important in
determining the most appropriate method of risk analysis including
the potential expenditure, the importance of getting it right,
the availability of data and the nature of the alternatives being
compared.
In carrying out risk analyses it was also essential
to be aware of the difference between corporate objectives and
operational objectives. The Board could be concerned with:
— avoiding loss of licence;
— maintaining customer satisfaction;
— increasing share value;
— maintaining competitive advantage;
while the operational side would be concerned
with all aspects of the function and performance of the business
in which failure would impact on the Board's objectives. This
also encompasses risks which the company needs to take account
of but which Ofwat, looking at it from a customer service point
of view, would be less concerned about (eg installing fencing
to prevent unauthorised entry and subsequent injury to trespassers).
Risks can then be scored in terms of the expected
impact on operational and/or business objectives. The risk reduction
potential for different solutions ranked and compared and a comparison
of serviceability versus expenditure can be derived. This approach
allows companies to take more informed decisions in terms of asset
maintenance and/or replacement.
Maintaining serviceability is the main criterion
adopted by Ofwat to determine the appropriate level of asset maintenance.
A risk-based approach to prioritise expenditure
will provide an auditable means of identifying priorities on a
consistent basis.
The Econometric Approach—Jeremy Atkinson,
NERA
The aim in capital maintenance programmes is
to maximise the net present value of benefits of service delivery
less the total costs of service provision. However, in the real
world the task is to minimise the net present value of total future
expenditure subject to the constraints of meeting the required
customer service standards as set out by the regulator.
There are three linked steps in the process:
1. determine how expenditure changes the
asset's observable future (its age, its condition etc);
2. determine how changes in the assets affect
their service performance;
3. determine how changes in the assets' service
performance affect customer service.
The requirement is to demonstrate relationships
between observable asset characteristics and customer service.
While some data is already available it is questionable
whether all water companies are in a position to develop adequate
econometric methods of analysis. Indeed, as each link in the chain
introduces a degree of probability as to outcome and performance,
the relationships, once found, do not remain static over time.
Each component is multidimensional and the aim is to minimise
expenditure consistent with a range of service performance criteria.
It is obvious that the process is not applicable to all assets,
nor can it be brought into play quickly. Full blown application
of the framework will take time. The key steps which now need
to be taken are:
— define "condition" in terms
of observable physical proxies;
— identify which observables best indicate
asset serviceability by an analysis of experienced asset serviceability;
— undertake an analysis of customer
serviceability to define it as a function of asset serviceability.
CUSTOMER EXPECTATIONS
John Banyard, FREng, Director of Asset Management,
Severn Trent Water Limited
What the customer wants is clean (visually and healthily)
water available whenever it is wanted, allied to "invisible"
disposal of waste water and products without any damage to the
environment. In short the customer wants dependability.
Dependability is delivered through the serviceability
of the asset base. But how do you define serviceability? It is
a combination of condition and performance? Initially the industry
believed it was but Ofwat has equated serviceability only to performance.
To date, condition has been defined by means
of a grading system one to five drawn up by a nationally funded
project by WRc. The general rule has been that any asset in grade
five is highly likely to fail and needs to be replaced within
a five-year regulatory period. If not all grade five assets are
replaced in any five-year review period then the likelihood is
that there will be an increase in quantity of assets which are
likely to fail and hence a situation which is guaranteed to reduce
serviceability and adversely impact on dependability in the future.
It is to avoid that situation occurring that companies have to
carry out asset maintenance programmes.
If reactive management is to be avoided and
customer serviceability preserved, it is essential that all grade
five assets are recognised as needing replacement within an AMP
period. Sophisticated techniques such as risk analysis or economic
damage functions are not a substitute for replacing worn out assets.
It is recognised that there may well be some grade three and grade
four assets which, if they failed, would have a severe impact,
and it is in addressing these assets in a situation where there
is capital rationing that the real value of more sophisticated
analysis lies.
Clearly, the industry cannot replace all of
its grade three and four assets, and there is need for more comprehensive
criteria which take into account the specific roles that an asset
fulfils in order to prioritise need over and above that of simply
replacing worn out assets.
AND FINALLY,
OFWAT'S
FUTURE APPROACH
Bill Emery, Director of Costs and Performance
and Chief Engineer, Ofwat
Service and serviceability has improved through the
nineties (the last two five-year periodic review periods). This
is because there has been better information available within
companies, and companies have used it more efficiently to improve
systems and performance and reduce costs. There have been much
higher levels of expenditure on capital maintenance than in the
earlier (pre-privatisation) period. There has been a substantial
quality improvement programme carried through. All this is good
news.
Therefore, the capital maintenance picture is
not one of doom and gloom. What needs to be done in the current
period is to ensure a timely delivery of the agreed programme
so as to achieve compliance and meet the new standards on time.
Ofwat will maintain effective regulatory action if there is any
slippage.
Ofwat will also work with DWI and EA to explore
the determinants of serviceability in relation to performance
of assets. Ofwat has also published a note to all Managing Directors—MD161,
12 April 2000—which outlines Ofwat's approach to determining
the maintenance of serviceability to customers. It points out
that:
— services need to be maintained to
customers at least cost, recognising a trade-off between risk
and cost whilst ensuring compliance with statutory duties;
— appraisals of capital maintenance,
operating expenditure and risk can be compared using discounted
cashflow;
— information will be needed on:
— operating costs and their impact
on Capex;
— the basis for timing decisions
on asset replacement;
— impact of obsolescence and the
lower costs of new technology;
— the terminal values and the discount
rates used.
By 2004 Ofwat will be wanting to know if companies
have met their primary obligations in the AMP3 period, ie the
maintenance of the agreed level of service to their customers.
There will have to be consistency between accounting methods used
on the income side of the account and the expenditure side.
There will have to be close correlation between
planned expenditure as defined at the start of AMP3 and actual
expenditure during the period.
Finally, there is one last, but very important,
additional factor. By 2004, the industry will have been privatised
for 15 years. Secretaries of State will be giving thought to renewal
of licences. They will be particularly concerned with aspects
of stewardship. How have the companies maintained the asset base?
EMERGING IDEAS
Participants in the think-tank then took part
in three parallel discussions.
Developing criteria for the future
What are the key issues?
— a problem is building up. Current
maintenance levels are below those needed to maintain serviceability
in the longer term;
— need to differentiate between customers'
needs and maintenance needs. Customers' priorities may not be
engineers' priorities;
— some assets are critical for serviceability,
others are not;
— must redefine serviceability objectives.
Then the industry can define what it needs to spend to meet these
objectives;
— econometric approach is complex and
cannot be developed for universal use by AMP4;
— penalties of failure to comply may
be greater from the customer and the shareholder than those currently
available to the regulator. Boards have to decide in terms of
the three pressure points—regulator, customer, shareholder;
— a pragmatic rule of thumb is needed
now as the current grading system is seriously flawed;
— there are far too many asset management
capital schemes (many of them small) in any one company to carry
out risk analyses on all of them;
— Ofwat is not interested in individual
assets—only in their combined use to deliver the service
the customer requires;
— the industry has complained in the
past about Ofwat wanting too much detail. Now it is considering
providing more detail;
— fourteen years after privatisation
will be a good time to look at overall stewardship of assets by
companies prior to a new round of licensing;
— efficiency gains may not be as obtainable
in the future as in the past;
— running down assets, with the inevitable
subsequent additional and very high expenditure will lead to much
higher prices at some point in the future. Will politicians, let
alone customers accept this?;
— Ofwat is not looking to agree methods
now (2000) which will not be used until 2004;
— Ofwat is prepared for bilateral discussions
with individuals (or appropriate groups of) companies.
What new approaches should be developed?
— develop more sensitive methods of
assessing capital maintenance requirements;
— redefine condition grade categories
for assets—pragmatically in the first instance but taking
into account the impact of asset failure;
— redefine serviceability;
— look to using econometric (where
appropriate) and risk analysis (more often) methods with companies
in order to provide firmer, more appropriate ways of assessing
capital maintenance requirements;
— look forwards, not backwards. Some
new assets may have shorter life than those they replace.
Action ideas:
— work with Ofwat and the other regulators
to redefine serviceability and grading of assets;
— work within companies and collectively
to strengthen the processes used to assess capital maintenance
programmes. Adopt those processes that a prudent commercial company
would use anyway (regardless of Ofwat's stated needs);
— be open and transparent in this work
and keep all stakeholders involved or informed as appropriate.
Engaging the participants
What are the key issues?
— involving all the relevant stakeholders
including:
— Ofwat (important but not sole
stakeholder);
— consultants, contractors and
suppliers;
— what about the customers?;
— Ofwat is driven by what customers
want;
— customers may be happy to let
"us" get on with it provided there are no failures;
— customers don't look 50 years
ahead;
— do we need to do more/better
customer research? Can we do industry-wide research?;
— or are customers over-researched?
Are they sick of being asked?;
— there is a need for a more stable,
continuous investment planning process. Is the five-year cycle
too short or too long?;
— do we need an annual review?;
— should we have a 20-year plan?
What new approaches should be developed?
— the industry needs to understand
the needs of all the other stakeholders before it can engage and
work constructively with them—their points of view, their
aspirations, their agendas;
— look to the future, not the past;
— MD161. This is a first step in a
new approach. Ofwat clearly want feedback. At present it is looking
to bilateral discussions with separate companies. Can the industry
develop a wider view with Ofwat involvement?;
— Ofwat is concentrating on serviceability.
It is developing its understanding by involving DWI and EA. Can
the industry assist? What about "dependability"?;
— what is current best practice? Can
we share it?;
— if companies want to alter the AMP
programme then justification will have to be provided;
— seek to learn from the situation
in Scotland.
Action ideas:
— work towards open, transparent, predictable
and consistent process and methods;
— involve practitioners and stakeholders
together in the process;
— Alert Minister (Michael Meacher)
to the extended involvement by and contribution from stakeholders
and enlist his views on stewardship;
— companies work closely with Ofwat
to establish common ground for requirements;
— Ofwat to host an MD161 working group;
— draw agreed relevant information
together and inform the wider audience including MPs and their
researchers; the City; civil servants and political advisers.
Whose risk is it anyway?
What are the key issues?
— the risks lie with the companies
but Ofwat has a stake as it regulates the return companies achieve
for taking risks. Is Ofwat prepared to share the risk?;
— new "risks" now lie on
the table including possibility of charges of corporate manslaughter;
— the current system categorises assets
too simply;
— the need to link condition and future
serviceability;
— risk is forward looking, currently
the process operates mainly on backward looking data—largely
driven by the regulators' requirements. Deferral of expenditure
is sub-optimal in the long term;
— Ofwat tends to discount companies'
views on risk of failure;
— what is the cost of asset failure?
If customers are not affected how should this affect serviceability;
— how can all companies develop agreed
systems;
— sewer flooding is a specific failure
problem.
What new approaches should be developed?
— some hostility to econometric methods.
Any new systems must be simple;
— look to adopting risk assessment
methods more widely;
— essential to adopt assessment procedures
that use predictions of failure and the impact of that failure—even
if these have to be best guesses at the moment;
— must work with Ofwat to convince
them of validity of any new systems of assessment that are being
adopted, there is a credibility gap;
— better definitions of serviceability
and performance are needed;
— must develop new methods in time
for AMP4.
Action ideas:
— start work immediately, as timescale
is tight, on agreeing new definitions of serviceability;
— develop a dialogue with Ofwat and
the other regulators to redefine the relationship vis-a"-vis
asset management decisions. Then work to develop agreed measures
for taking those decisions;
— give special attention to developing
new methods of categorising assets, especially those likely to
produce severe consequences if they fail (particularly refers
to sewers).
THE ESSENCE
OF THE
DEBATE
The outcome of the last periodic review was
to identify a gap in perception. Companies had argued that a spend
of around £8 billion was needed to ensure that the asset
base delivered the level of service required. Ofwat felt that
a spend nearer £6.3 billion would achieve that. The debate
in the think-tank exposed differing views as to how to measure
customer serviceability (ie what the customer is happy with) and
what proxies can effectively determine the relationship between
spending money on maintenance and the achievement of customer
serviceability.
Without adopting more rigorous techniques, such
as risk assessment and econometric modelling, water companies
could lay themselves open to scrutiny by their shareholders and
their customers in terms of failing to extract the medium economic
and service advantage from their assets. Ofwat would obviously
have something to say about such a situation.
However, sophisticated methods require sophisticated
data which can be expensive to collect and analyse. To embark
on a major exercise to put in place such techniques across the
whole industry capital asset base would take a significant amount
of time and call for heavy expenditure. Indeed, part of the initial
analysis has to be the determination of which assets lend themselves
to giving an economically viable answer when subjected to these
techniques and some assets just don't warrant such attention.
Secondly, there are some areas which are already
ripe for attention—these could be tackled first. Thirdly,
there are some pragmatic approaches which could be introduced
quickly which, even if they do not provide the best longer-term
solution, could certainly improve the situation in terms of a
better focused pattern of spend on maintaining the capital assets
and ensuring high levels of customer serviceability.
Ofwat has suggested that it is in the companies'
own business interest to adopt clear well-proven methods of calculating
the expenditure required and provided that is done, they are happy
to discuss, on a company by company basis, the appropriateness
of the methods for achieving the requirement of customer serviceability.
The water operators feel that it would be more
valuable to have industry-wide agreed methods which it is confident
would not only satisfy business and customer requirements as they
see it, but would also be acceptable to Ofwat in the next periodic
review.
Given that Ofwat does determine the price customers
pay for water it clearly does have a role to play in also determining
the price customers are prepared to pay to ensure that their criteria
of service are achieved. The draft agenda for action which has
been developed as a result of the day's deliberations is designed
to help close these gaps.
AGENDA FOR
ACTION
There are two phases working in tandem with
each other:
— extending the collaborative approach
by organising multilateral discussions between the involved parties
to develop broad agreement on the need for appropriate measures;
and
— extending the technical/economic
research to develop industry-wide acceptable methods for providing
clearer guidance for the decision takers on how much to spend,
on what and when.
The collaborative approach (ie the water industry
working with other relevant stakeholders):
— explore and develop the definitions
of customer serviceability to get an agreed understanding of what
the customer wants and is prepared to pay for;
— explore and develop agreement as
to which pragmatic measures for assessing asset maintenance requirements
could be developed;
— explore and develop agreement that
new and more sophisticated grading schedules to identify status
of individual assets could be developed;
— agree a restricted range of assets
which lend themselves to a speedy solution in terms of developing
more sophisticated analysis for determining economic expenditure
requirements;
— ensure all stakeholders are involved
and/or kept informed (as appropriate); and
— engage the Minister in the debate
as to the relationship between asset management and service provision
and the longer-term issue of stewardship.
The research approach (ie the water industry
in whole or in part, and where possible, in collaboration with
stakeholders):
— customer needs and preparedness to
pay;
— pragmatic asset maintenance measures;
— new grading schedules; and
— risk analysis and econometric techniques.
Water UK intends setting up two work groups
for these two strands and also to continue to facilitate the transfer
of best practice (as it is identified) through forum based activity.
19 A list of those present is shown in Annex 1. Back
| 
