What if Britain had never privatised water?

In this Britain, water is a service you pay for at cost, not a monopoly you pay a dividend on.

Every spring the same envelope lands, and this year it landed harder than any year since 1989: the average household water bill in England and Wales rose 26% in one go, to about £603, the steepest single rise since the industry was sold. At the same time the country's biggest supplier, Thames Water, was busy not collapsing under roughly £17bn of debt, and the rivers were carrying a record amount of sewage. Three headlines, one system.

This file exists to finish the sentence the water companies never finish. Bills are high, they say, because pipes are old and investment is dear. Half true. The other half is that a slice of every bill is not buying pipes at all. It is servicing debt the companies chose to take on, and paying dividends to their owners, on an asset the public handed them for nothing. So we ran the counterfactual. Same pipes, same rain, same leaky Victorian network. One change: Britain keeps the water system it actually owned until 1989, run at cost by a public body, the way Scotland still does. This is not a manifesto for renationalisation. It is the invoice for the dividend.

Britain ran the experiment for us, and drew the control group along a line on the map. In 1989 the Thatcher government floated the ten regional water authorities of England and Wales on the stock market. Scotland did not. Scottish Water stayed in public ownership; so did Northern Ireland's. Same island, same rain, same regulator template, one difference: who owns the tap. Thirty-six years later, this is what each system charges a household, drawn to scale:

Read that chart next to your bill, because they are the same story. The two publicly owned systems on this island are the two cheapest. And the gap is not bought by starving the pipes: over the last two decades Scottish Water has invested more per household than the English companies, not less. Roughly £180 a head against about £120 in the current period. Cheaper bills and more investment, from the system that pays no dividend. That is the control experiment, and it points the opposite way to everything privatisation promised.

A private water company is a strange animal. It has no competitor, cannot lose a customer, and sells a product nobody can refuse. Water is the one bill you truly cannot shop around: you get the pipe your postcode gives you, at the price the regulator allows. Which means the return to shareholders does not come from winning a market. There is no market. It comes out of the bill, and only out of the bill.

The Competition and Markets Authority put a figure on it: on average about a fifth of a household water bill goes not on treating water or fixing pipes, but on servicing the companies' debt and rewarding their shareholders. Publicly owned Scottish Water pays no dividend at all, and spends only around 8% of its revenue on debt, borrowed cheaply through government rather than from private markets that demand a premium. Same job, same pipes, roughly a quarter of the finance bill. The researchers at Greenwich who model this call the difference the finance-cost gap, and it is the single cleanest reason a Scottish household pays less for the same glass of water.

This is the part that turns a high bill into a scandal. When the ten companies were sold in 1989 they were not just handed a monopoly. They were handed it debt-free: the government wrote off more than £5bn of their existing debt and threw in a "green dowry" of roughly £1.5bn in cash on top, a leaving present from the taxpayer to smooth the sale. The companies started life owing nothing and holding a cheque.

Today those same companies owe about £60bn between them. That debt did not build a second network; England's pipes leak much as they did. A great deal of it was borrowed to pay dividends, an accounting trick that lets owners take money out while the customer, not the shareholder, is left carrying the loan and paying the interest through the bill. Over the years since privatisation, shareholders have taken somewhere between £53bn on Ofwat's own narrow measure of declared dividends and around £85bn on the widest reasonable count. Meanwhile the shareholders themselves have put in next to no new equity of their own. The money that fixed the pipes was mostly the customers' and the lenders'. The money that left was the shareholders'.

And while it left, the rivers filled. In 2024 the water companies discharged raw sewage into England's rivers and seas for a record 3.6 million hours across more than 450,000 spills, now that every overflow is finally monitored. Thames Water alone accounted for nearly 300,000 of those hours. This is the estate the dividends were drawn against: a network leaking at both ends, owned by companies that borrowed to pay themselves and were fined, in Thames's case £122.7m by Ofwat in 2025, for the privilege. The dividend did not fall from the sky. It came out of the bill, and out of the river.

What the same household would pay on the public model that already runs north of the border, next to what it pays now.

Notice that the saving is not a rounding error, and notice who it favours. The households paying the most, in the regions where the private companies loaded the most debt and spilled the most sewage, are exactly the households the public model would help most. There is no version of this where the customer of a broke, leaking, dividend-paying monopoly comes out ahead of a Scottish household on the same island paying at cost.

And £85bn is only the money you can see leaving. It leaves out the interest quietly folded into every bill for the next thirty years, the fines that come out of the same pot the dividends did, and the clean-up cost of the sewage the whole arrangement was too busy paying itself to prevent. This is not a bill you settle once. It is a charge on a thing you cannot refuse to buy, forever, set by a company that answers to its owners and not to you.

This is where privatisation articles usually pick a team, so we won't. Four things are true at once, and a file that tells you only one of them is selling something.

First, the private model genuinely did buy something. In the 1980s the network was falling apart and the Treasury would not fund the repair. Privatisation pulled in roughly £190bn of investment over the decades since, money a cash-strapped government of either party might never have voted for. Britain's beaches and drinking water are cleaner than they were in 1989. That was the deal's real argument, and it is not nothing.

Second, the public route is not free money, and reversing course now is dear. Public ownership does not abolish the cost of building pipes; it changes who carries it. Scotland funds its larger investment programme by borrowing through government, cheaply and with no dividend stacked on top, but that debt still sits on the public balance sheet and is repaid through bills or taxes all the same. Swapping dividends for state borrowing lowers the price of the money; it does not make the pipes free, and moving that borrowing off the Treasury's books was part of the whole point in 1989. Unwinding it today would be dearer still: buying the companies back at market value, or letting a Thames fall into special administration, lands the state with tens of billions in cost or inherited debt. The saving in this file belongs to a system that started public and never took the private detour, not to a wand you can wave over 36 years of financial engineering.

Third, take away the profit motive and you take away a discipline you then have to replace. The chase for profit is also a pressure to control costs, and several of Britain's pre-1989 nationalised industries drifted into overmanning and slack management once that pressure was gone. Scottish Water is the counter-example, lean and consistently well run, but that is a product of how it is governed and regulated, not a gift that arrives free with the word "public". A public monopoly answers to ministers rather than owners, and ministers are not always the more demanding boss. Remove the shareholder and you have to put something just as exacting in its place.

Fourth, and this is the important one, the villain is the model, not the pipes. Scotland's advantage is not sentiment. It is cheap government borrowing, no dividend, and tight governance, nothing more mystical than that. Which means the same result is reachable without full nationalisation: hard caps on gearing, a real ban on borrowing to pay dividends, dividends tied to actual investment and performance, a regulator with teeth. Ofwat waved through the debt-loading and the payouts for a generation and is only now, with Thames on the brink, discovering the brakes. The cheap water in this file is a choice about how you finance and run a monopoly, not a law of nature. Britain made the expensive choice, and defended it as prudence.

Britain did not fail to build a water system. It built one, owned it, and then sold it debt-free to people who borrowed against it, paid themselves, and let it spill. Scotland kept the same service in public hands and charges its households about £115 a year less while investing more in the pipes. That is the whole argument, and it is not a theory. It is a border with a control group on each side.

The bill for the English choice is not filed in a regulator's annexe. It is on the doormat every spring, and this year it rose more than a quarter in a single step: a fifth of it flowing to debt and dividends on an asset the public gave away, the rest stapled to a network that leaked a record amount of sewage into the rivers the very same year. The dividend was real. The water in the river was the receipt.

Scotland pays for its water once, at cost, and owns the tap. England pays for its water every month, forever, plus a dividend to whoever owns the tap this decade. That was the choice. Your bill is the receipt, and it never stops printing.

• Ofwat, average bills 2025–26 and Water UK, annual average bill changes 2025–26: England & Wales average household bill ≈ £603, up £123 (26%), the largest rise since privatisation; Southern Water ≈ £703 (+47%), Northumbrian ≈ £506 the lowest.
• WICS / Scottish Water, 2025–26 charges: average Scottish household charge up 9.9% (about £44) to roughly £488, among the lowest of any UK water provider.
• Hall & Gray, "Leaking Money", PSIRU / University of Greenwich (2025): finance costs of the privatised companies ≈ 35% of revenue (2023–24) vs ≈ 8% for public Scottish Water; shareholders have injected next to no new equity since 1989.
• Ofwat / CMA on returns and dividends: on average about a fifth of the household bill services debt and rewards shareholders.
• History of the 1989 privatisation: the ten companies were floated debt-free after a write-off of >£5bn of debt plus a "green dowry" of roughly £1.5bn in cash; they now carry about £60bn of net debt.
• We Own It / Ofwat dividend data and Left Foot Forward: dividends since privatisation between ≈ £53bn (Ofwat declared) and ≈ £85bn on the widest count.
• The Rivers Trust / Environment Agency, 2024 spill data: a record 3.6 million hours of sewage discharge across >450,000 spills in 2024, with 100% of storm overflows monitored; Thames Water ≈ 298,000 hours.
• Investment per household, Scotland vs England (drawing on Greenwich/PSIRU): publicly owned Scottish Water has invested more per household than the English companies while charging less.
• Full model: the current average household bill in each nation (Water UK/Ofwat for England & Wales; WICS/Scottish Water for Scotland), compared like for like with standing charges included. The whole England-to-Scotland gap is attributed to "the private system", a deliberately conservative framing that does not isolate the dividend-only effect. Aggregates use ≈ 25 million English and Welsh households.