It is hard to find anyone who has a good word to say for the water industry in England and Wales. Thirty-five years after privatisation, the industry is mired in pollution, debt and public hostility. The blame game is in full swing, with politicians blamed for pressurising regulators to hold down prices, the Environment Agency (EA) blamed for not properly policing the sector, OFWAT for allowing the wave of financial engineering, and the companies blamed for paying out large dividends and salaries and bonuses and generally mismanaging the assets.
All of this is being played out in the public gaze by the unfolding disaster that is Thames Water. Its failings cover almost everything that could fail. Thames has failed to properly manage its sewerage works by its own admission. Its owners have variously mortgaged the assets to gear up its balance sheet to the benefit of shareholders and not invested enough in the Thames catchment. Its multiple breaches of environmental law have led to large fines (from which it is now pleading to be exempt). It has failed to monitor and properly report sewage spills. It failed to appeal against the price caps set by Ofwat at the 2019 price review (PR19) even though it knew the consequences would be bad for its network, and now it is engaged in an attempt – aided by the government and OFWAT – to head off being placed into Special Administration, despite violating the licence conditions placed upon it.
Pointing the finger is easy to do. There is a cornucopia of targets. But much more difficult is to know what to do now – how to put the industry on a sound footing so the rivers and aquifers can be properly protected and enhanced, for our benefit and for future generations, and to restore public trust.
Performance politics
Once elected in 2024, the new Labour government moved very quickly to assert that it was indeed “doing something”. In a classic bit of performance politics, it told the public it was getting a grip, taking powers to lock up water company directors who caused pollution. In fact, all that its new Bill rushed through Parliament did was to take powers to lock up directors who did not cooperate in investigations. It probably had these powers anyway, but hoped the public would see its zeal in clamping down on the industry without noticing that it was doing anything but this.
Having done its bit of performance politics, the much harder challenge lay ahead. What is the government actually going to do about the water industry? Its first challenge was the final determinations of PR24, which set very large price increases, largely front-loaded. Customers now faced a very big bill for what they perceived as the faults of the companies, being made to pay to put right what many customers thought they had already paid for.
To the government’s credit, it did not intervene in the price increases. It was further encouraged not to do so, because it might have tipped three or four of the companies into administration, no longer able to finance their functions. It could of course have watered down the environmental requirements, and hence spread the required capital programme over a much longer period. Lower bills, less of a pollution clean-up. But it resisted this temptation too.
To the government’s further credit, it decided to take its time and properly consider what needs to be done. Instead of rushing into announcing wider reforms, it called in Sir Jon Cunliffe to lead an Independent Water Commission which would make recommendations, and then the government could decide whether to accept some or all of what the Commission came up with, with a view to legislating in 2026. That is where we have now got to.
What Cunliffe came up with
In passing the challenge to Cunliffe, the government at the same time constrained the options his Commission could consider. Nationalisation was ruled out, not least for fear of the overwhelming sentiment in favour amongst Labour MPs. For a related reason, the government wanted to head off Special Administration, for fear that, once in administration, the backbenchers would push through nationalisation.
With nationalisation pushed aside, and Special Administration resisted, the central dilemma for the Cunliffe Commission was, on the one hand, how to make sure that the private sector would keep on raising debt and equity to finance the investment programme, since current customers could not be asked to pay for what was, in effect, a large programme of capital maintenance, whilst on the other hand restoring public trust. The former requires higher returns (and hence higher customer prices) and stable and predictable regulation; the latter requires lower bills and more interventions.
The Commission’s Interim Report[1] is long on setting out the failures of regulators and companies (though soft some of on the government failures), but short on solutions. The Final Report[2] gives its headline solutions: the government setting clear objectives, tidying up the environmental legislation (essentially repeating the direction of travel set out in the Corry Report[3]), abolishing OFWAT, going to a supervisory model of regulation, and setting up new catchment forums.
This is the technocratic approach, and each of these recommendations has its pros and cons. But whilst there is great scope for improvement (there would be anyway after 35 years), none seriously addresses public trust, and none clarifies how investors are going to get more money with lower risks.
What the Cunliffe Commission does achieve is to set the scene for the government to pick-and-mix from the recommendations, and buy time for the main act – legislation – to follow. This main act will increasingly be in the shadow of the next price review, PR29, with all the political challenges this will bring in the run-up to the next election.
Clarifying objectives
The Cunliffe Commission wants the government to clarify its objectives. But then every review of public policies starts with the objectives, and wants government to be clear. In theory, it is pretty simple. The government wants cheap water that is also sustainable and secure, just as it wants cheap energy that is also sustainable and secure, and just as it wants cheap rail and public transport that is also sustainable and reliable. Of course it does, like every government.
All of the above are largely signals of the preferred direction of travel, rather than well-specified objectives. It gets difficult when it comes to funding and finance, and to specifying trade-offs when there are losers as well as winners (as there almost always are).
Funding and finance run into the classic political difficulty we the electors impose upon our governments: we want lots of sweets, but we don’t want to pay for them, and we don’t want higher taxes. In the case of the utilities, that means that governments encourage the private sector to lend, and hope that pay-when-delivered (future voters not current voters paying) will kick this problem down the road.
Investors can see where this leads. More and more debt, and then some future government and regulators reneging on making future voters, as customers, pay. Investors rightly worry that we might not vote for those who will force us to pay, and that future customers might not actually be able to afford to pay anyway. Collapsing public trust just makes the prospect even worse.
What do rational investors do in such circumstances? They see higher risks, they see that this risk is political and regulatory, and they demand higher returns and more and more guarantees. They take the money out when they can.
What the Cunliffe Commission had to do was de-risk water investment and increase returns. For this, the Commission had to provide cover for prices to go up even more. To do this, the PR24 numbers have to be increased further and, for the government and the regulators, this is preferably done below the radar. OFWAT has already started on this path – the DPC (direct procurement for customers) model, piggybacking more large-scale CAPEX on the back of the utilities’ regulatory asset bases (RABs), with cost pass-throughs. But the real challenge here is to increase prices in another hike in PR29, and that looks very difficult.
Whilst it is true that water prices are low in real terms, the problems with another big hit in 2029 are multiple. The obvious thing to do is to not have a PR29, and the Cunliffe Commission baulks at recommending moving on from periodic reviews to a more granular approach to regulating prices. Indeed, the Commission wants both periodic reviews and supervision.
Abandoning or at least delaying PR29 is an open goal that the Commission has missed, even with OFWAT as the goalkeeper off the pitch as a result of the Commission’s abolition recommendation. The government needs to act fast on this one. The short-term answer is to have a rollover, with perhaps a one-year fix to get past the election. These sorts of ad hoc arrangements have been used by the Civil Aviation Authority in respect of Heathrow Airport, so there is precedent. To be clear, PR29 is not a good idea for government, the companies or the investors, and might well trigger further calls for nationalisation.
The central recommendation for government to clarify its objectives beyond the overarching bland statements of the direction of travel is a nice try but never going to be delivered. Being clear and explicit about the trade-offs is not the route to enhancing a Secretary of State’s future prospects, and governments almost always prefer to make it up as they go along, one decision at a time, and only in each case once the public debate has indicated the voters’ preferences. We the voters don’t actually want to be confronted with the harsh truths that we should pay for the necessary capital maintenance and investments, and as the ultimate polluters we should pay for the pollution that we cause every time we flush the toilet.
Once upon a time government tried to be explicit. Privatisation was accompanied by five-year, fixed-priced price caps (actually for water it was initially supposed to be a ten-year deal). Regulators were supposed to keep their hands off the industry until the five years were up. This was the Stephen Littlechild model of RPI-X. It was a brave idea, but it failed at the first hurdle, when Littlechild had in effect to re-do the 1994 electricity distribution price cap. Windfall taxes followed in 1997, and credible regulatory and government commitment never recovered.
The preference to make it up as we go along is not necessarily to be avoided, provided the terms of each investment are set in concrete, or at least the risks that the investors bear as their part of the deals are clear. Think broadly. Is it really a good idea to set an impossibly demanding 2030 target for net zero electricity? Investors wonder what happens when it is missed, or when an alternative government takes a slower pathway, provoked by the sheer scale of the costs a fast track requires. Would it really make a credible policy to commit to precise outcomes in five or ten years in water? Take the current Secretary of State’s promise to “halve sewage pollution from water companies by the end of the decade”. It is good performance politics, but it is sufficiently vague to be “flexible”, and anyway if the past is repeated with one Secretary of State at DEFRA per annum, there are at least three more to come in this current parliament. There may even be a new government in 2030.
The inevitable result of failing to define the trade-offs, of what might be called British pragmatism and empiricism, is some form of rate of return regulation. The broad framework for economic regulation is set, with a view to flexibly agreeing the company spending within the framework of a guideline rate of return upon which investors can rely. That is what happens across much of Europe and in the US. It is true that rate of return regulation does not have the same efficiency incentives that price cap regulation promised, but then, after 35 years, the England and Wales private water companies are not noticeably more efficient than their continental comparators. Over this period a higher return has been provided to motivate these incentives that have not really delivered the promised super-efficiency.
The pragmatic answer is where we have been heading for a while across many of the utilities. Rate of return regulation typically has an efficiency condition attached, notably in the US.
The danger of supervision models
The Cunliffe Commission ducks these issues. It wants to preserve the periodic reviews and it wants to graft on top its supervision model. It gets both badly wrong.
Let’s start with the periodic reviews. The Commission is long on what it thinks is wrong with the periodic reviews. For it, the problem is the economic models that OFWAT uses. It thinks this creates a one-size-fits-all mentality, and fails to properly take account of the specific characteristics of each company and catchment.
There are valid criticisms to be made, both of the detailed methodology and data used in the OFWAT models, and in the way they are applied in periodic reviews to set the price caps. But what else is OFWAT supposed to do? And more importantly, there are two ways in which the companies can challenge the application to their patches. First, they can appeal to the Competition and Markets Authority (CMA) against the price caps set by OFWAT. In PR19 most failed to do so, and for Thames Water this was a spectacularly bad decision. But there are also adjustment mechanisms, cost pass-throughs and interim determinations. If a company really thinks the application of the modelling to itself is wrong, resulting in its inability to finance its functions, then the law provides for redress – if the company can prove its case. Even if the CMA does not agree, it can go on to judicial review.
The economic modelling tries to use data and comparisons and this information is valuable, and should be part of any price determination. What the Commission could instead have recommended is that this approach is advanced to incorporate the new and immense resources of digital mapping and data, so that it is refined for each company and there can be a detailed exchange on modelling in each catchment. This is one of the key baselines for the Catchment Regulation Model that I have proposed to replace the existing approach. More on this below.
In the event, the Commission is not brave enough to carry through its criticisms. It ends up wanting to keep the periodic review and keep the economic modelling. What it proposes is to layer on top of the periodic review yet another layer of regulation. It wants a supervisory model borrowed from banking regulation on top of the periodic reviews. Make no mistake. The Commission wants more regulation.
In this supervisory model, OFWAT’s successor is going to have company-specific teams covering each of the companies. These water teams are going to have the expertise to understand their charges, and engage in the day-to-day understanding of the companies’ decisions. They are going to be able to intervene quickly and presumably be “agile”. They are going to be able to engage in turnaround processes (presumably modelled on the Thames Water case).
Why would anyone think this is a good idea? These extra regulators are going to break down the clarity of the separation of responsibilities between private and public, and they are going to, in effect, shadow the boards of the companies. They are going to have to be highly skilled and will inevitably be expensive. The main recruits would presumably come from the companies they are supposed to regulate. It will probably be even more a case of revolving doors. Many regulators already join regulated companies once their terms of office come to an end, or consultancies that advise regulated companies. This will be worse.
Supervision is a dynamic process. Each problem begets more problems, and each requires more information and more interventions. A supervisor is very exposed. It’s a bit like the probation officer whose supervisee commits a crime. Any supervisor of a water company will need to have in mind what happens if anything goes wrong on their watch. That leads to risk aversion, and more delving into the detail of decisions. Indeed, why not in effect steer the decisions themselves? Hence the process leads to a gradual erosion of the boards of the companies as the supervisors start to implicitly decide the key items. If this is really what the Cunliffe Commission has in mind, then perhaps nationalisation would be better.
Shifting the deck chairs around
The Cunliffe Commission’s model keeps periodic reviews, it keeps the economic modelling, and it adds on supervision, and thus requires more regulators to do the supervision. This expansion of regulation is hidden behind the Commission’s headline recommendation: abolishing OFWAT. It is another example of performance recommendations and hence performance policy. It helps to underwrite a sub-text: “it’s all OFWAT’s fault” (except for the lack of clarity of government objectives), and hence OFWAT will be abolished.
The regulatory deck chairs are rearranged accordingly. What happens when OFWAT is abolished? Its functions (now including supervision) go into a merged body with the EA and for good measure the highly effective Drinking Water Inspectorate gets thrown in too. Notably, the EA does not lose major functions, such as flooding. So we have a new super-regulator.
There are a number of good reasons to have a single regulator, though experience elsewhere is not entirely encouraging. OFCOM is torn in multiple directions. The merger of the National Rivers Authority (NRA) + Her Majesty’s Inspectorate of Pollution (HMIP) + the waste regulators to form the EA left the EA top-heavy on water and flooding. But in the OFWAT + EA model, the marriage of the environmental requirements with the economic regulation of the capital and maintenance programmes has merit.
Moving the deck chairs around rarely delivers the sort of quick results commissions and governments anticipate. In the short term, there are lots of administrative costs and distractions of creating new organisational structures, boards and so on. Where the chance of success is greatest is where the opportunity is taken to properly define the roles and objectives of the new body at the outset, rather than simply bundling the existing organisations into a whole and leaving them to sort it all out (which is what happened when the EA was set up).
The big opportunity is not really with “abolishing OFWAT” but rather rethinking the EA. When it was decided to create the EA to pursue integrated pollution control (the buzz term at the time), the model that was rejected was the Environmental Protection Agency (EPA) – small, sharp and focused on inspections and prosecutions. Instead of modelling itself on HMIP, which was merged with the NRA to create the EA, it was the NRA model that was chosen. The result was to incorporate a staff of over 6,000 – to grow to over 10,000 – many of whom were engaged in operational works including flooding.
Now is the opportunity to split out the flooding and make the EA into a proper EPA.
What then of OFWAT? The Commission’s recommendation encourages the thought: “OFWAT is dead, long live OFWAT!” Though it will be under the same board within the new body, ask yourself: what will actually be different? Will it be more friendly towards environmental considerations and hence increase water company spending and investment more, and therefore push up bills further? OFWAT is accused of holding down bills and taking a jaundiced view of environmental matters in the past. Will it now relax? If the new body takes a more environmentally benign position, then PR29 will make interesting politics.
After the hiatus of the process of moving the deck chairs around, perhaps the outcomes may not be much different. It will be a big distraction, and it is a distraction worth having if the result is a well-structured regulatory framework.
Here is what it could be: OFWAT is merged not into the EA but rather into a broader utilities regulatory body. This was indeed proposed by a previous Labour government in the late 1990s, though in the end only OFCOM emerged from the 2000 Utilities Act. The EA is shorn of its flood defence responsibility, shedding most of its current staff, and it is restructured as an EPA. With this structure in place, the Catchment Regulation Model could provide the stable and sustainable regulatory framework for the next 30 years, and beyond.
What would really make a difference – the catchment approach
The Cunliffe Commission nods in the direction of taking catchments seriously, but ducks the issue of making them the heart of water regulation. Instead, it proposes a sort of stakeholder grouping for each, but without sorting out who is in control. It is more an engagement to deal with what the Commission sees as a democratic deficit, and hence it is the stakeholders plus local government.
There is of course a local government model. This is the one used across Europe and one that once ruled the roost in the UK. But there is a world of difference between local government being involved in the regulatory processes alongside other stakeholders and it having a direct stake in the companies. Local government in the UK has been so emaciated by successive governments that it does not have the capacity or capabilities to take on water too.
What would really make a difference is to move entirely to a catchment-based regulatory framework. I have set this Catchment Regulation Model out elsewhere in detail,[4] but let’s recall the main rationale. The sources of pollution, flooding and the water quality before treatment are all in the catchments. Catchments are systems, and in a system everything depends upon everything else. There are few marginal decisions that make sense independently of the catchment system impacts.
Given the system nature of the pollution, the flooding and the water supplies, the rational step is to take an integrated approach. Farming is responsible for around the same amount of pollution as water company sewage spills. Not to bring farming into the framework is economically silly, and adds greatly to the costs and hence the bills that customers face. Catchment solutions to floods, sewage and river pollution and the costs of water treatment are almost always also cheaper.
The system is increasingly amenable to detailed digital modelling. In the catchment model, the regulator has ownership of the digital system map, and makes this public for all to see and indeed play with using their own simulations and preferred options. The catchment digital map is the basis for identifying how to achieve better outcomes.
The outcomes are delivered by multiple parties. When the system outcomes have been identified through the mapping and the public interface with it, anyone and everyone can bid to carry out the necessary measures and works. These include flood defence groups, landowners and farmers, local authorities, the Highways Agency, conservation NGOs and trusts, and of course water and sewerage companies.
The process starts with indications of interest, and these options are then winnowed down for a formal bidding process. The monies are recovered through the use of system charge to water and sewerage customers, and against a catchment system RAB.
Out go the periodic reviews, down go the costs of meeting any specified objective as a result of competitive bidding, the monopolies of the water companies are now open to challenge, and up goes the transparency and accountability of the regulatory processes. It harnesses the new digital technology along with all sorts of flood and water management options that can be tested. It makes sewage spills utterly transparent and the state of the assets can be publicly inspected.
Better still, the Catchment Regulation Model goes with the flow of ever-greater mapping abilities, as AI and greater resolution add to the details and AI allows multiple scenarios to be run at a fraction of the cost and much faster than now. It subsumes all the information used currently in OFWAT’s economic modelling. The digital map is therefore useful to planners and developers, to farmers to look at water storage and reserves, to road design and maintenance, and much else. It even has value to health, as it enables the detection of viruses in the water coming through the sewers and rivers.
Think what is also avoided by going for the Catchment Regulation Model. There is no need for OFWAT, other than some residual monopoly regulation which is best done through a single utility-wide network regulator. It is not abolishing OFWAT to reinvent it on an ever-larger scale inside the new integrated body that the Cunliffe Commission has in mind. The EA can focus on becoming a slimmed-down EPA (and benefit from the catchment digital maps). No need for the cumbersome supervisors the Commission has in mind. Competitive bidding removes much of the vulnerability to single monopolies.
Better still, the rates of return are determined through the bidding process, and not by the cumbersome CAPM/WACC (capital asset pricing model/weighted average cost of capital) modelling that the Cunliffe Commission favours. Put aside that the WACC – as an average – over-rewards debt and under-rewards equity and hence has been a positive spur to financial engineering and gearing up the balance sheets. It would not be needed at all. Out goes another Cunliffe Commission recommendation on a standard CMA-based WACC.
Rethinking catchments and water and sewerage management
At the heart of the current water and sewerage model is a focus on end-of-pipe solutions, and sewage and storm flows integrated into a single pipes system. The Catchment Regulation Model starts the other way around. Why build lots and lots of new tanks at sewage treatment works to hold more storm water and hence to limit sewage spills, when it is possible to start the other way around? Why not deal with the reasons storm water run-off hit the sewerage works so quickly? Why not think about the source rather than the end point?
Thinking in a catchment way starts with upstream re-wriggling of rivers and natural capital interventions, and with porous surfaces for developments and for roads and front gardens. It starts with water storage at home, at housing developments and at new factory and warehouse sites. Storage leads to recycling. Data centres can harvest water, store it, use it for cooling and recycle it, for example.
The additional advantage of at-source solutions is that they address and reduce flooding as well as sewage disposal. In all of this, farmers are central, controlling around 70% of the land, and currently responsible for water run-off from cropping and land-management activities, run-off of pesticides and nitrates from fertilisers, slurry spills, and poultry faeces. Uniquely amongst polluters, they expect to be paid not to pollute.
Now imagine if farmers were confronted with digital catchment maps that continually improve the mapping of pollution and run-off from each parcel of land, down to a very fine resolution. It is no longer out of sight, out of mind. If the Catchment Regulator sets objectives to reduce specific kinds of pollutants in rivers, the digital mapping exercise allows farmers to bid to improve performance against a baseline and in competition with water companies and others. They become part of the solution not part of the problem. Money spent on buffer zones along rivers, on making sure unprotected soils from maize fields in winter are not running off into the river, and holding flood water on the land can be cheaper than some other current end-of-pipe solutions for meeting these overall objectives.
There are problems, not least perverse incentives. Why not plough up a field and then get paid not to do so? But the scope is now limited because all this shows up in the digital mapping and in real time. Digital mapping shines a torch on what is actually going on in the catchments.
It is time for the full benefits and opportunities of digital mapping and AI to be reaped. DEFRA has been a notoriously analogue department, and the EA is now trying to catch up with the technology. Instead of both the EA and DEFRA trying to have a complete mapping ability, it would be better – and cheaper – to start with the catchments, digitalise these, and then these maps can be aggregated when required at the national level. It is hard to underestimate the opportunities for improvement here.
The new Water Act 2026 – how the government could set the water industry on a sustainable pathway through to 2050
It is not hard to work out what will happen on the current pathway from the Cunliffe Commission to the Water Act in 2026. The government has already indicated which bits of the Commission’s recommendations it likes. The main prize for the Secretary of State is the abolition of OFWAT.
What has not been committed to is the format of what replaces it. If the deck chairs are rearranged into a super-body combining the EA, OFWAT and the DWI, the consequences are likely to be disappointing. It doesn’t improve customer trust, it doesn’t improve investor returns and stability, and it doesn’t do much to move towards an at-source approach. Rather, it cements in lots and lots of new tanks at sewerage works, buying time until they fill up and overflow too, and at high costs.
It is still open for the government to abolish OFWAT and do something far more sensible, which is capable of putting the industry on a sustainable long-term path, one that could endure for 30 years or more and take us past 2050.
The necessary steps are: abolish OFWAT; adopt the new Catchment Regulation Model; split out flooding from the EA; turn the EA into an EPA; embrace digital mapping and AI in open-access catchment maps, allowing anyone to do their own scenarios; and abolish the periodic reviews and move towards competitive bidding. This takes time, but the transitions can be handled by rolling over the current PR24 past 2029.
The above means that: there is no need for supervisors (Special Administration Regimes are sufficient); and there is no need to set a WACC or get the CMA to do so.
What is not to like about this? Perhaps kicking the can down the road is easier short-term politics, but as with the immediate issues with Thames Water, can-kicking comes with a price and that is already biting. In the meantime, calling in a Special Administrator for Thames is a necessary start on the path to sorting out the water sector for the medium and longer terms – and finding cheaper ways of achieving better outcomes.
[1] Independent Water Commission (2025), “Interim Report”, 3rd June.
[2] Independent Water Commission (2025), “Final Report”, 21 July.
[3] Defra (2025), An independent review of Defra’s landscape, 2nd April.
[4] See Helm, D. (2017), “Regulatory Reform and The Systems Operator Model”, January; (2019), “The Systems Regulation Model”, February; and (2020), “Floods, water company regulation and catchments: time for a fundamental rethink”.
