Environment, Food and Rural Affairs CommitteeFurther supplementary written evidence submitted by the Water Industry Commission for Scotland (WICS)

Thank you for your letter to Sophia Goring, requesting further information from the Water Industry Commission for Scotland on behalf of the Efra Select Committee. Before turning to your specific questions, I would like to provide some contextual information in relation to the issues raised.

My understanding from the Water White Paper is that the Government does not propose to mandate water and sewerage companies (or the water only companies) to separate their retail activities.

On balance, I consider that this approach represents the most appropriate course of action. It is in customers’ interests to ensure that the supplier who is best capable of meeting their needs is able to supply them. In other words, and as I said in evidence to the Committee, it is important that there is a genuinely level playing field so that all existing and potential entrants can enter and operate in the market on the same terms. This means that the retail operation of say, Thames Water, should have to offer a customer in London a level of service and value for money that is every bit as good as that offered by a new entrant if it is to retain the business of that customer.

The Government should have the same interest if it is to achieve its aim of a sustainable and efficient water industry. Such a level playing field could be achieved through:

separating the ownership of the retail and wholesale activities;

legal separation of the retail and wholesale activities (mandated by Government or opted for by the regulated water company); or

functional separation—under this option legal separation is not required; instead, a relatively restrictive Governance Code ensures a level playing field for new retailers and protects the interests of customers.

This Governance Code would reassure new entrants that they could be successful if they offer better services than those of the incumbent’s retail operation.

In practice, even if there is legal separation some form of Governance Code is likely to be required to ensure that the wholesaler does not discriminate in favour of the retailer it owns (although the Code would be less restrictive than in cases where there is no legal separation). Such discrimination could potentially still be possible in matters such as the priority with which the wholesaler responds to a problem raised by an end customer.1

The Water Services etc. (Scotland) Act 2005 required Scottish Water to establish a separate retail entity, but gave it a wide degree of discretion as to how to achieve this.2 Scottish Water opted for legal separation. In doing so, it was no doubt mindful of our view that there should be a level playing field for both Scottish Water and any new entrants.

If the UK Government were to require legal separation in England, there is the possibility that, in certain circumstances, some highly leveraged water companies with very restrictive covenants on their debt could potentially incur financial penalties. (Although in our view this possibility would be relatively remote if the new framework is broadly similar to that which has been tried and tested in Scotland.)

It could be argued that it is the owners of these companies (the equity holders)—and not customers3—who gave gained from such arrangements in the form of lower debt costs. As such, it could follow that the Government should not worry about imposing an obligation to separate. However, the Government should be keen to avoid any action which could be seen by infrastructure investors as increasing the political risk of investing in the UK.

I agree that the benefits of lower debt costs in securitised companies flow principally to the equity holders. However, I would still consider that the Government should be wary about mandating legal separation. In my view it is likely that investors would regard such an approach as increasing political risk in the water industry and other infrastructure projects that the Government may want investors to finance.

It would be preferable if Government made clear its commitment to a level playing field and to customers being able to choose the supplier that best meets their needs. It should also signal both the importance of a level playing field and the fact that legal separation is likely to require a less restrictive Governance Code. Wessex Water separated its billing operation into a new joint venture company with Bristol Water some ten years ago. The company reports that, in all that time, debt investors and rating agencies have never raised any questions about the implications of this separate legal entity.

The Experience in Scotland

As noted earlier, Scottish Water opted to create a separate subsidiary company to fulfil its retail responsibilities. However, it was not straightforward, even then, to ensure a level playing field. There were issues associated with ensuring that costs were properly allocated, that no informational advantage was available to Business Stream, and that Business Stream’s responsibility would be to its customers not to its parent company, Scottish Water.

Getting this separation between the former retail and wholesale arms of Scottish Water’s non-household business was central to successful implementation of the competitive retail market.

In October 2006 we required Business Stream to produce a Business Plan explaining how it intended to operate within the new framework. In response we granted Business Stream provisional water and sewerage licences. This allowed us to use our statutory licensing powers but restricted Business Stream’s activities.

We also specified three tests that Business Stream had to pass before it could be granted permanent licences:

1.Business Stream’s activities and assets were demonstrably separate from those of Scottish Water.

2.Business Stream’s governance was demonstrably separate.

3.Business Stream was financially viable and independent of Scottish Water.

Only when Business Stream had passed these three tests in January 2008 did we grant it permanent water and sewerage licences and allow it to operate freely within the market. Until it met these tests, new licensed providers were able to sign up customers for transfer at market opening and Business Stream was unable to respond.

Since market opening we have had to look carefully at some of Business Stream’s activities to assure ourselves that it was not misusing its incumbency to its advantage.

Legal and Functional Separation

The process that we followed contrasts with “functional” separation. Under this approach, the regulator has to seek undertakings that require certain behaviour. These undertakings would cover the same issues that were of concern to us in monitoring Scottish Water’s decision to separate its retail activities legally. With functional separation the economic regulator has to be much more “hands-on” in ensuring that new entrants can see that the playing field is level.

In comparing the costs of these two options, it is perhaps useful to explain the nature of the different costs incurred.

Set-up costs are the costs the economic regulator, the regulated company and the market settlement and registration system operator incur. These costs occur once (although perhaps over two-three years) in establishing the initial retail framework.

On-going costs are the costs that the economic regulator, the regulated company and the market settlement and registration system operator will incur each year that the framework is in place. These costs are incurred in ensuring that the playing field continues to be seen as level and that settlement and registration systems function effectively.

Functional separation may have lower set-up costs for the regulated company. But set-up costs for the economic regulator and the registration and settlement system operator are likely to be higher. Overall set-up costs may well be lower. However, because there is not the same transparency as exists under legal separation, more concerns are likely to be raised by other market participants about whether or not the playing field is genuinely level. There are also likely to be more regulatory concerns surrounding the allocation of costs. These issues are likely to increase the costs for the regulator and the registration and settlement system operator on an on-going basis. The regulated company is also likely to incur higher compliance costs on an on-going basis.

It may be appropriate to characterise legal separation as ensuring that the company takes responsibility and is fully accountable for how it interacts with other market participants. In contrast, functional separation requires the regulator to take the lead.

In my view, this explains why the Government should make it clear that it wants to see a level playing field and that the Governance Code arrangements are likely to be more restrictive if a company opts not to separate its retail activities legally. This is not the same as the Government mandating legal separation, as it allows investors to make their own assessment of the costs and benefits of different governance possibilities.

The Costs of Legal Separation—CCWater Evidence

On a separate, but related issue, I have noted the oral evidence that the Consumer Council for Water provided on the costs of legal separation.

While I cannot comment on the views of Defra or Ofwat, I would be surprised if the costs of establishing a non-household retail framework would be allowed to have an adverse impact on household customers. Indeed, I believe it should not.

As I indicated above, the Consumer Council for Water is correct in highlighting that the initial costs of a framework based on optional legal separation would be higher than the costs for functional separation. However, as I set out above, it is important to consider the set-up and on-going costs for the regulated company, the economic regulator and the operator of the settlement and registration systems. In Scotland additional savings were realised from the legal separation of Scottish Water’s retail and wholesale activities—in my view, similar savings could well be achieved in England.

The questions in your letter ask me to quantify the costs of frameworks based on legal and, alternatively, on functional separation. You also ask me to quantify the benefits of both options.

The information provided in this letter is based on our best estimates of the potential costs and savings in England and Wales, based on the experience gained in Scotland. We have included Wales in our assessment because we were unable to separate out the costs and benefits attributable to one company. Where we are less confident about the estimates (because they are not based on experience) we have made this clear. We have, however, included them in order to be helpful to the Committee.

What is WICS’ estimation of the costs to (a) Ofwat and (b) water companies (including any financing costs) were there a requirement for legal separation of water companies’ retail and wholesale functions to be completed within three years of a Water Bill receiving Royal Assent?

Our response

It is not possible to generalise about any additional financing costs that may be incurred by water companies as a result of a Government requirement to separate legally. The companies that are highly leveraged would be the most vulnerable to additional claims from investors for extra financing costs given the increased risk of a breach of that company’s covenants. Equally, there would, almost certainly, be some companies that would incur no additional financing costs as a result of a requirement to separate. Clearly, the longer the period, the less debt would be affected (assuming refinanced debt would not carry restrictive covenants).

I have also noted the comments of some witnesses to the Committee that the industry’s overall cost of capital could be increased by the proposals in the White Paper. The proposals to offer non-household customers a choice may entail an increase in the costs of capital that are allowed to the retail activities of a water company. However, in Scotland this has been immaterial relative to the reductions in costs that have been achieved by Scottish Water’s retail subsidiary, Business Stream.

In our view, it should be possible to ensure that, other than this small adjustment to the retail component of the current vertically-integrated companies, the impact of the Government’s retail proposals would not increase the cost of capital. Indeed, the proposals may actually lead to a reduction, as this single reform should reduce the risk of the wholesale business materially. For example, the fact that retailers pre-pay in Scotland, and bear the risk of non-payment, has removed the bad debt risk to Scottish Water’s wholesale business.

The Committee may want to consider that some companies may be unsuccessful in any retail market and would incur additional financing costs if they are unable to exit the market. In our view, a company should be free to exit the retail market.

It is also possible that the financing costs of the wholesale business could be reduced once investors have had time to understand the new framework. This could happen, for instance, if there was a working capital benefit to the wholesale business (ie retailers pre-paid wholesalers, as happens in Scotland) and all bad debt risk was borne by retailers (again, as happens in Scotland).

In our view, companies will respond pro-actively to the regulatory framework within which they have to operate. If a company can ensure that it faces a substantially less restrictive Governance Code by legally separating its retail activities, it is likely to pursue such an option. It would be in the interests of its investors to do so because the costs of complying with the Code would be much reduced.

We have a clear understanding of the costs that were incurred in Scotland from introducing a retail competition framework that also involved legal separation of Scottish Water’s non-household retail activities. These are set out in Appendix 1. These costs are presented in both cash and net present cost terms.4

On the basis of our experience in Scotland, we have estimated the costs that could result in England and Wales if a broadly similar approach were adopted.5 These estimates are set out in Appendix 2. For the reasons set out above, we have not included any estimate of the additional financing costs that may be incurred by some companies with a highly leveraged financing structure.

In our view the deadline will only impact on the costs of separating activities if it is unreasonably short. In Scotland, full legal separation was achieved well within three years after Royal Assent.

Table 1 summarises our estimate of the present value of the costs of introducing a retail framework with legal separation in England and Wales.

Table 1

ESTIMATED PRESENT VALUE OF THE COSTS OF INTRODUCING A RETAIL FRAMEWORK WITH LEGAL SEPARATION IN ENGLAND AND WALES

Cost item	PV over 10 years	PV over 30 years
Set up costs (one-off)	£182m	£182m
On-going costs	£166m	£360m
Total costs	£348m	£542m

What is WICS’ estimation of the costs to (a) Ofwat and (b) water companies (including any additional financing costs) were there a requirement for functional separation to be completed within three years of a Water Bill receiving Royal Assent?

Our response

We cannot be as confident about the costs that would be incurred if the requirement were for functional separation only. This is because the framework we implemented involved legal separation of Scottish Water’s retail activities.

We would not expect any additional financing costs to be incurred as a result of a requirement for functional separation. This assumes that a company could choose to exit its retail activities should that be the right commercial decision.

In terms of the set up costs, our view is that the costs regulator would incur higher costs in establishing the framework (specifically in relation to ensuring a level playing field, assessing the appropriate level of wholesale revenue, and agreeing an appropriate set of wholesale tariffs with the company). It is also likely that establishing the central registration and settlement systems would involve higher costs. However, the set up costs for the incumbent company may be materially lower.

As to on-going costs, we would expect regulatory (market supervision) costs to be much higher (given the greater potential for disputes between market participants. Similarly, the more restrictive Governance Code that would be required to ensure a level playing field would cost both the retailer and the wholesaler more.

In order to make an estimate for the Committee of the costs of functional separation, we have assumed:

a 10–20% increase in the regulator’s costs of establishing the market;

a 10–20% increase in the incremental costs of establishing the settlement and registration systems;

a 25%-40% reduction in the water company’s initial set-up costs;

a 10%-20% increase in the on-going running costs of the settlement and registration systems;

a 40%-60% increase in the incremental costs to Ofwat of managing the framework; and

additional compliance costs of between £3 million and £4 million a year.

Appendix 2 sets out the rationale for the assumptions we made.

Table 2 summarises our estimate of the present value of the costs of introducing a retail framework with functional separation in England and Wales.

Table 2

ESTIMATED PRESENT VALUE OF THE COSTS OF INTRODUCING A RETAIL FRAMEWORK WITH FUNCTIONAL SEPARATION IN ENGLAND AND WALES

Cost item	PV over 10 years	PV over 30 years
Set up costs	£119m to £145m	£119m to £145m
On-going costs	£209m to £231m	£454m to £501m
Total costs	£328m to £376m	£573m to £646m

More detailed information on these costs is set out in Appendix 3.

What is WICS’ estimation of the savings to water customers over the first decade of a reformed regime (assuming that the White Paper proposals are implemented)? Can it provide comparative figures for estimated savings over the first decade of a reformed regime in which water companies were required to (a) legally and (b) functionally separate their retail arms?

Our response

Appendix 4 sets out the savings that have been achieved in Scotland and the present value of these savings on the assumption that the existing level of costs is maintained in real terms. It should be noted that, in this analysis, we have not sought to quantify the benefits of:

the improvements in service levels now being provided to customers;

the environmental and carbon benefits of the water efficiency achieved; or

the cost savings that Scottish Water has achieved through identifying redundant activities when it separated its retail and wholesale activities or from the company’s greater clarity and focus on service level outcomes.

In Appendix 5 we also calculate the net present value of the changes over a 30—year horizon and over a 10—year horizon (excluding any benefits accruing after the 10—year time horizon).

On the basis of our experience in Scotland, we have estimated the savings that could result if a broadly similar approach were adopted in England. These estimates are set out in Appendix 6 to Appendix 11. The potential for these cost reductions is based on two different approaches6 and uses published information7 that the water companies provided to Ofwat.

In the functional separation scenario we assume that the savings that could be achieved as a result of the new framework may be up to 10% lower than under legal separation.

Table 3 summarises our estimate of the present value of the savings from introducing a retail framework with legal or functional separation in England and Wales.

Table 3

ESTIMATED PRESENT VALUE OF SAVINGS FROM INTRODUCING A RETAIL FRAMEWORK WITH LEGAL OR FUNCTIONAL SEPARATION IN ENGLAND AND WALES

	PV over 10 years	PV over 30 years
Legal separation	£463m to £606m	£1,850m to £2,171m
Functional separation	£417m to £546m	£1,665m to £1,953m

Table 4 summarises our estimate of the net present value (NPV=total savings less total costs) of introducing a retail framework with legal or functional separation in England and Wales.

Table 4

ESTIMATED NET PRESENT VALUE OF INTRODUCING A RETAIL FRAMEWORK WITH LEGAL OR FUNCTIONAL SEPARATION IN ENGLAND AND WALES

	NPV over 10 years	NPV over 30 years
Legal separation	£115m to £259m	£1,309m to £1,629m
Functional separation	£40m to £218m	£1,019m to £1,381m

Conclusion

In our view it is important to distinguish between a requirement in legislation for companies to separate and a decision by a company voluntarily to separate its retail and wholesale activities.

We agree that the Government should not require companies to undertake legal separation of their retail activities. However, we do consider that the Government should emphasise its determination to ensure a level playing field for new entrants to the area of any appointed company. It should also be clear that the Government would expect the Governance Code to be more restrictive where a company chooses not to separate legally its retail activities but does specifically want companies to be allowed to pursue only functional separation. Such an approach would mean that additional financing costs (that are not more than offset by other cost reductions) are avoided.

We have a detailed understanding of the costs and savings that have been achieved in Scotland. As such, we consider this allows us to make relatively robust estimates of the potential costs and savings that would occur if all companies were to opt to separate legally and avoid additional financing costs. We have estimated the costs that would result if companies were to implement only functional separation.

Our analysis suggests that the 10—year cost of establishing a retail framework is broadly similar under both the legal separation and functional separation only options. The 30—year costs are much lower for the legal separation option.

Our analysis suggests that the 10—and 30—year savings from establishing a retail framework are higher for the legal separation option than for the functional separation only option.

Our analysis suggests that the 10—and 30—year NPV (benefits less costs) of establishing a retail framework are higher for the legal separation option than for the functional separation only option.

I hope that these answers to your questions are helpful. I would, of course, be happy to provide further evidence if this would be useful to the Committee.

May 2012