Supervision of water companies: Which problems can it solve?

In summary

• The Government’s response to the Cunliffe Review placed supervision at the centre of water companies’ future regulation

• While supervision is often treated as a single regulatory philosophy or a universal solution to the sector’s challenges, it is better thought of as just one of the regulatory tools available

• Supervisory-style regulation will substantially change how companies are regulated, with it becoming the principal approach in some areas such as financial risk management

• The future framework for water regulation should not seek a single supervisory model but incorporate additional supervisory tools alongside other approaches, including the continued use of benchmarking and financial incentives

The publication of the Independent Water Commission's review put supervisory-style regulation at the centre of the reforms. "As part of a fundamental reset in the way the regulator engages with companies, the Commission is recommending that it adopts a 'supervisory approach'..." the review read.

This was reflected in the Government’s White Paper in response. This framed supervision as key to addressing a wide range of issues facing the water sector:

"A new supervisory approach will shift regulation towards being more proactive and targeted to the specific needs of each water company to improve performance – while also increasing grip over water company delivery. These reforms will establish a model for the new regulator to intervene early to help underperforming water companies recover faster, attract the investment they need, and improve long-term financial resilience, asset health, and environmental performance.”

This raises questions about the role that supervision should play, including the extent to which it replaces other regulatory tools such as benchmarking, financial incentives and the use of periodic price determinations.

To explore this, it is worth starting by asking what we mean by “supervision”.

One of the challenges with this area is that there is no single definition of supervision.

In financial services, a large part of regulation is referred to as being supervision. But this closely resembles what is referred to as “compliance monitoring” by some safety regulators, and how the ORR “regulates” Network Rail’s performance. A large part of Ofwat's oversight during periodic control periods also could be considered as 'supervisory', as does its oversight of company’s financial structures.

For our purposes, we can focus on a small number of characteristics that help to contrast supervision with “incentive-based” regulation.

This naturally exaggerates the differences and ignores the reality that regulators combine these approaches, typically backed by overarching licence requirements. However, it does help tease out when you might rely more on each of these approaches.

One way to approach this question is through the lens of how to reveal information about the regulated firm. Regulators try to overcome the fact that companies know more about their businesses, including the costs, revenue potential and ways to improve efficiency, than they do.

Currently, regulators do this by comparing between companies (benchmarking), providing incentives that encourage the companies to reveal the truth (financial incentives), and by collecting data from companies (monitoring).

These tools are better at tackling certain forms of information asymmetry than others. Consider the following two examples:

• A company may have good information about the risks of waste water treatment works failing. This is held privately but can – with detailed checking by informed experts – be largely discovered by a regulator.

• A management team may consider that they can generate efficiency savings by reorganising how they deliver maintenance. This is privately-held information but, unless it has been written down somewhere, cannot be uncovered by a regulator through checking. Even if a document exists, the management team may be uncertain about the level of efficiency that can be achieved.

So, when thinking about what regulatory tool to use, it is important to consider whether you are trying to discover private information that is ‘measurable and certain’, or information that is either difficult to quantify accurately or hard for anyone to know in advance (unmeasurable and/or uncertain).

Information collection is costly for all parties, so it is also important to consider the value attached to the information, once it is discovered. Again, consider these examples:

• The regulator discovers that a firm is using new technology to reduce the cost of replacing water mains, revealing that all other water companies can reduce their costs. It gives the regulator relevant information about other firms

• The regulator discovers the best way to design a large one-of-a-kind reservoir, which can help set cost allowances for that reservoir, but may have limited wider relevance and therefore value

These two ways of thinking about information can be used to inform the choice between supervision and incentive-based approaches.

This is summarised in the diagram below, which views information problems in terms of whether the underlying information can be discovered and how valuable it is to the regulator.

This highlights some useful distinctions:

• Financial incentives are well-suited for the discovery of information that is hard to find in other ways and where it can be applied to reduce costs in future, including at other firms. Intuitively, this is where the value of the information is larger than the cost to consumers of the incentive mechanism

• Where measurement is possible (and not too costly), detailed oversight and benchmarking avoid the cost and lags involved with financial incentives. Some of this is typically labelled as benchmarking or takes the form of detailed regulatory reporting

• There is a particular challenge when regulators tackle one-of-a-kind spend where there is high uncertainty. Unless you can describe and measure a clear outcome metric, the regulator may have little choice but to determine what should be delivered, and possibly in some detail. The growth in investment in resilience-enhancing projects is already pulling Ofwat in this direction

The parallel to financial services supervision is the strongest in the bottom left – overseeing and enforcing minimum standards of financial resilience on firms.

This way of thinking about supervision has some implications for the future regulatory framework in the water sector:

• Supervision is one of many tools available to regulated firms. The best approach will combine supervision with a range of others. It isn’t a single fix for every problem

• Financial incentives are good ways to reveal hard-to-discover information. This includes where innovation or risk-taking is needed

• Benchmarking and comparison between firms is useful for certain types of information. It performs less well when company-specific or project-specific factors are large

• To overcome information asymmetry the right regulatory tool varies, depending on what you need to discover and the value of that information

Legislation should retain the flexibility to combine different regulatory tools in ways that improve outcomes. Reflecting this, any obligations on regulators to supervise water companies need to be sufficiently flexible in practice.

It will be easier for regulators to explain their approach if it is clear whether they are prioritising approaches that deliver control (risk management), secure near-term cost savings or support longer-term innovation and future cost reduction.

Firms will be more effective in achieving the right balance between detailed oversight and financial incentives if their asks are backed by a coherent and consistent articulation of how it will improve outcomes for consumers.