Supervisory regulation: what does good look like?
Reflections on Christopher Davis's (Oxera) session at Moody's UK Water 2026, and what it means for credit ratings. By Douglas Rowlings, Managing Partner.
Why the question matters now
"Supervisory regulation" is the centrepiece of the Cunliffe Review's prescription for water, and nobody can yet say precisely what it is. Christopher Davis of Oxera Consulting LLP opened his session at Moody's UK Water 2026 with that admission: ask the regulation directors of the 16 companies to define it and you would get a different answer from each. Neil Griffiths-Lambeth, introducing him, was blunter still: "I've got no idea."
Davis's reading of the mood was sharp. Neil Griffiths-Lambeth had asked that morning whether the sector had moved from gloom to optimism; Davis's answer was that it had moved to "uncertainty", and that word ran through everything said on the day.
Two deficits sit underneath it. The first is public trust: CCW research cited on the day found only 44% of households think their current bill is fair, and only 50% believe their company cares about the service it delivers, before the increases needed to fund investment arrive. Davis borrowed Mark Carney's line that trust arrives on foot and leaves in a Ferrari, "and the Ferrari may have screeched out of the car park".
The second is investor confidence. Companies forecast hundreds of billions of pounds of enhancement spend by 2050, possibly £300bn or more, before asset health. Investors say the proposition has changed: returns have been squeezed, environmental and asset risks have grown, and dividend expectations have shifted as returns are reinvested in RCV growth. With gilts at their highest level in 20 years, the cost of getting this wrong is very high.
What good would achieve
Davis's test for a successful regime is a list of outcomes, not a rulebook. His slide put five things on it:
Customers accept that bill increases are justified and represent value for money.
Networks are maintained and enhanced in a way that is sustainable for future generations, with progress tracked against long-term objectives.
Trust and confidence are restored across stakeholders and the public.
A stable and predictable environment for investment: fair returns in exchange for committing capital.
Companies are held to account for their performance and delivery.
He added two qualifiers to the last point that are easy to miss. Accountability should come with a learning culture, so that lessons are learnt rather than simply penalties paid, and when things go right the sector should champion them. "This is not just about punishment."
Anyone who reads rating methodologies will recognise the fourth outcome. It is close to the language Moody's uses to describe a supportive regulatory environment, and it is the one the rest of this article keeps coming back to.
A change in philosophy, not tools
The Independent Water Commission looked at the foundational tools of economic regulation and left them largely intact: five-year price reviews, ongoing performance monitoring, assessment of financial resilience and licence enforcement. What changes is how the regulator approaches those tasks and how it exercises judgement: from standardised to company-specific knowledge, from mechanistic rules to constrained discretion, from uniform outcomes to variation within guard rails, from periodic review to ongoing dialogue, and from retrospective enforcement to early intervention.
Organisationally that means supervisory teams facing off against companies. In-period, the regulator's job becomes spotting risks early and intervening before they become issues, with precedent from financial services for the monitoring of financial resilience. There is far less precedent for using supervisory judgement inside a price review, which Davis saw as both the risk and the opportunity: the sector can still shape it.
Two of his remarks deserve underlining by anyone who builds a financial model around the regulator's. "The models are one source of evidence, but they are not the true answer." And on the purpose of it all, the football analogy from the Q&A: today's system reaches back ten years, punishes, and leaves "an asterisk over that period of history". Good supervision would have stopped the behaviour in the first place.
Three things it depends on
Whether supervision proves more effective than what went before depends, in Davis's view, on three foundations. His warning was that "the same people, under the same external pressures, are likely to come to the same outcomes".
The calibre of people. Judgement-based regulation needs skill, experience and seniority in the supervisors making the calls. His slide described today's regime as "economics-led (no Chief Engineer)" and asked for a broader mix of skills and experience.
The culture and policy framework. Supervisors need to be told what they are optimising for and what their discretion is meant to achieve. The decisive external pressure is whether government treats the price review as a bill-reduction exercise or as continuous improvement in the sector.
The information available. Here the onus is on companies. Supervision can help with information asymmetry, but it cannot help with information gaps: if nobody knows the condition of the assets, or what spending does to that condition, no supervisor can fix it. Expect a move away from standardised, aggregated data that fits econometric models towards granular, company-specific evidence.
Trust needs safeguards at both ends. If the public comes to believe the supervisors have been captured by the companies, credibility goes quickly; if supervisors are seen to stray beyond their discretion, the same. "Constrained discretion" therefore has to mean something concrete, in guidelines and safeguards.
It will also cost more, not less. Asked whether this cut against the wish for simpler, quicker and cheaper regulation, Davis conceded that effective supervision could cost two to three times as much as a hands-off incentive regime, because supervisors must be intimately involved in what the business is doing, even if enforcement action falls as a result. "You cannot have a supervisory regime that is cheaper than a hands-off incentive regime."
PR29 is fast approaching
Davis's uncomfortable observation was that a full supervisory system will not be up and running by PR29. Recruiting the people takes time, and so does learning the assets. Meanwhile companies must do more of everything, in enhancement and in asset health, with infrastructure costing more than it did five years ago, after a period of high inflation and energy costs, and with gilts at a 20-year high.
Yet basic elements of the next review remain undefined. There is no national water strategy to say how trade-offs should be made. Regional system planners are back on the table, having previously been taken off it. Nobody can yet say who will conduct PR29, under what duties, or within which appeals framework, and the debate over public control continues alongside. His warning: "we must make sure that we're not trying to answer all of those tricky challenges in the second half of 2029, as we have done in some previous price reviews".
What can change without waiting for the new regulator? His slide listed five implications, "underpinned by returns that are globally competitive and reflect sector risk":
Lower risk: a revised incentive package and a risk calibration that takes account of the potential "doom loop". Low risk is not no risk; equity must carry some risk for incentives to bite, "but it should be better balanced. We should not, one year into a price review, have £240 million of ODI penalties."
Recognise regional factors: a targeted role for supervision to capture company-specific circumstances, with variation within guidelines. Companies have asked for this over multiple price reviews and the bar has always been set very high.
Prioritise asset health: build on the asset health roadmap and the cost-change process of the last 18 months, with clarity over activity levels and what is funded.
Simplify: a streamlined performance regime and fewer regulatory mechanisms and reconciliations, so that boards understand and act on the incentives. "Simple regulation is not necessarily effective regulation", but the current system is overly complex and duplicative.
Hold accountable: funding linked to delivery through price control deliverables, and ongoing performance supervision. If companies are spending hundreds of billions of pounds of customer money, they should be clearly accountable for it.
None of this, he said, is rocket science and none of it requires legislation, "but it does require will and planning over the next year to 18 months".
What it means for credit ratings
Stability and predictability of the regulatory environment is the first factor on Moody's scorecard for regulated water utilities, and the sector's score has already stepped down twice. In May 2018 it went from Aaa to Aa, after Ofwat's consultation on capital structures and a perceived rising risk of political interference. In November 2024 it went from Aa to A, on a regime judged less supportive of credit quality, more enforcement-led, and more volatile for cash flow. Stefanie Voelz's slides at the same event posed the next step as "A → ?", citing a delayed transition plan, an undefined future framework and "political rhetoric running ahead of policy". Moody's has a negative outlook on the sector for 2026, puts the average rating at a weak Baa2, and sector gearing at around 70%.
The reason this matters to a finance director is mechanical. Each step down has tightened the ratio guidance for a given rating: a Baa1 now needs gearing of about 68% or below and an adjusted interest cover ratio of at least 1.6x, against 75% and 1.4x before 2018.
So how does supervisory regulation map onto that framework? It can cut both ways.
The case for upside. Everything on Davis's PR29 list points towards lower cash-flow volatility: a narrower corridor of returns, fewer and better-calibrated penalties, asset-health funding that reflects actual condition, preventative intervention rather than retrospective enforcement, and funding tied to delivery. Lower business risk, consistently applied, is exactly what would stop the stability score falling further and, over time, support a higher one. Moody's lists the hallmarks of an Aaa framework as an established track record and consistent application of rules, transparency, independence from government interference, publication of regulatory parameters and financial models, and a defined and independent adjudication process with timely conclusion. A supervisory regime can be built to those specifications, or not.
The case for caution. Discretion is the opposite of mechanistic predictability unless it is constrained in ways an outside analyst can see. "Variation within guard rails" means company-specific outcomes, so rating differentiation across the sector is likely to widen rather than narrow. Judgement-based decisions are harder to model, harder to forecast and harder to appeal. On that last point Davis was clear: restricting the scope for appeals at the very moment the regime changes fundamentally would be strange, and an appeals framework that can accommodate supervisory judgement will need careful thought and case law. Rating agencies price those things. Southern Water's own slide listed "a route to challenge via the CMA" among the things that had helped its turnaround.
And the transition itself is a credit event in slow motion. Until the new regulator's leadership, legal status, duties and strategic guidance are settled, the "A → ?" question stays open, and Moody's deck is explicit that near-term pressure is downward, with no upward pressure expected over the medium term. Davis's final slide made the same point in cost-of-capital terms: "the current lack of clarity on this framework risks increasing the sector's cost of capital".
What issuers can do now
For a water company CFO or treasurer, the practical conclusion is to treat the transition as a live rating factor rather than a 2029 problem.
Plan to today's guidance, not yesterday's. Measure headroom against the thresholds for an A-scored regime (for a Baa1, gearing at or below about 68% and AICR of at least 1.6x), and run a scenario for a further step down. Equity, dividend and hedging decisions taken now will be judged against those numbers.
Close the information gap before the supervisor asks. Granular, verifiable asset-condition data is the currency of supervisory regulation. A company that can show what its spend does to asset health is more likely to win the variation it argues for, and the agencies will read it as lower execution risk.
Stress-test discretion. Model company-specific outcomes rather than sector averages: ODI exposure, price control deliverable clawbacks, cost-change windows and the timing of in-period interventions.
Engage early and consistently. Supervision rewards credibility built over time and penalises surprises. The disclosure discipline that builds trust with a supervisor is the same discipline that builds it with a rating committee.
Track the milestones that decide the "A → ?" question: the appointment of the new regulator's leadership, its legal status and independence, Ofwat's new strategic policy statement, the shape of the appeals regime, and whether a national water strategy finally sets out the trade-offs.
Davis's own conclusion is the right place to end. Regulation needs to help mitigate rather than amplify risk, so that financing costs stay down while the sector delivers a vast enhancement and asset-health programme. Supervision could play a positive role in that, and in holding companies to account, but only if government supplies the clarity on objectives and trade-offs that supervisors will otherwise lack. Without it, in his words, "we risk the same failures of the past".
If the regulatory transition touches your own rating headroom, we would be glad to talk it through. Arrange a free, strictly confidential initial consultation via our contact page.
