My water bill is £78 a month.
I know exactly why that number lodged in my head. It is because my electricity bill is almost nothing. We have solar panels, a battery, a heat pump and two electric cars, and between them they have taken a bill that would otherwise be one of the largest things we pay and reduced it to a rounding error. Every one of those was a rational response to a price signal — a spread between peak and off-peak, an export rate, a unit cost worth arbitraging against.
Then the water bill arrives, and it is £936 a year, and there is nothing to be done about it at all. We have spent a decade of cost-of-living pressure being told to shop around, switch, insulate, generate. On water there is no switching, no tariff to move to, and no meaningful way to use less. You simply pay.
Here is the fact that started this. An eight-minute shower uses about 80 litres. Heating that water takes 2.79 kilowatt-hours, which on a heat pump at 15p a unit costs 13.9 pence. On a gas boiler at the current cap rate it is 24 pence.
The water itself costs 59.7 pence.
In the South West, the water in your shower costs more than four times what it costs to heat it. Even an immersion heater — the most wasteful way of making hot water yet devised — is cheaper than the water going through it.
The letter
A few weeks ago South West Water wrote to tell me they are replacing my meter with a smart one.
This is not a local exercise. South West Water is fitting 440,000 more smart meters by 2030 and upgrading every existing household meter to advanced metering within ten years. Nationally, Ofwat's price settlement funds 10.4 million smart water meters by 2030 at a cost of £1.7 billion, taking household coverage from 11.9 per cent to roughly half.
That 11.9 per cent is worth sitting with. Seventy-one per cent of Britain's electricity and gas meters are already smart. Electricity has half-hourly settlement, time-of-use tariffs, export payments and a competitive market in tariff design. Water has one flat rate, charged identically at three in the morning on a wet Tuesday in February and at six in the evening on August Bank Holiday.
So the infrastructure that would make anything else possible is being installed across an entire region in the next four years, and the honest answer to what it is for is: more accurate bills.
I think it could be for something considerably more interesting. But to explain why, I have to start with why my bill is so high in the first place — because it turns out almost everything I assumed about that was wrong.
It is not expensive water
My £936 is 46 per cent above the national average household bill of £639. My consumption has nothing to do with it. Working backwards from the direct debit, we use about 105 cubic metres a year, which for three people is 95 litres per person per day. I built a model of South West household demand calibrated independently against South West Water's own published assessed charges, and it expects 98 litres for a household of that size. We are, to within three per cent, exactly average.
The gap is entirely geography. And the interesting part is which half of the bill it sits in.
Compare South West Water with Wessex Water — the neighbouring company, the same peninsula, the same coastline, the same bathing water obligations, and also an expensive company. At 110 cubic metres a year a South West household pays £977 against £763 in Wessex. Of that £214 gap, £184 — 86 per cent — is sewerage.
More striking still: South West Water's water rate is £3.0158 per cubic metre. Wessex charges £3.0573. South West water is slightly cheaper per cubic metre than its neighbour's. The entire premium is on the other side of the bill, where South West Water charges £4.6858 per cubic metre against Wessex's £3.1124 — the highest sewerage rate in England.
The South West does not have expensive water. It has expensive sewage.
Why
Two reasons, and they compound.
The first is rain. I pulled ten years of daily rainfall for the largest population centre in every water company region. Plymouth gets 1,258mm a year across 42.5 days of heavy rain. London gets 721mm across 16.5 days. Combined sewers spill when it rains, so rainfall is not weather — it is a cost driver. Across all ten companies, mean annual rainfall correlates with wastewater investment per head at r = 0.75. The two wettest regions, the South West and the North West, carry the two heaviest wastewater burdens per person in England.
The second is arithmetic. South West Water is spending £761 million between 2025 and 2030 on storm overflows and pollution — a sevenfold increase, covering 283 overflows. It is protecting 860 miles of coastline, over a third of England's total. And there are only about 818,000 households behind it.
That works out at £930 per household, or £1,274 per metered household. The same programme spread across Thames Water's customer base would be £48 a head. In the South West it is £423 — nearly nine times the burden per person.
Here is the part I did not expect. Total investment per head in the South West over this period is £1,673, against an England and Wales average of £1,704. It is two per cent below average. Current capital spending does not explain the highest bills in England at all. What explains them is the split: the water business is spread across 2.3 million people, and the wastewater business — the part the coast makes expensive — sits on 1.8 million.
Who the coast is actually for
South West Water serves 1.8 million residents. It also, by its own reckoning, serves 10 million visitors a year.
Charge the storm overflow programme to residents alone and it is £423 a head. Charge it across everyone who actually uses the coastline and it is £64. Residents are carrying a 6.6 times subsidy on the cleanliness of beaches that are, for six weeks of the year, mostly full of other people.
You can see the same thing in the water itself. Across the year, visitors are 7.1 per cent of the person-nights in the region. But in the 45 days of the school summer holidays they are 21.7 per cent of the people present and, because holidaymakers swim daily and face no marginal cost, roughly 29 per cent of the water and sewerage load.
The cross-subsidy, in pounds
None of this would matter if the tariff reflected it. It does not.
Capacity — reservoirs, treatment works, trunk mains, and above all that £761 million storm overflow programme — is sized by peak load. A flat tariff recovers it evenly across every cubic metre sold in the year. So a property that concentrates its use into the weeks that size the system pays less towards capacity than it causes, and everyone else makes up the difference.
I modelled seven household types for the region and allocated capacity costs by when each one actually uses water. The result:
Holiday lets and second homes are 10 per cent of connections, cause 22 per cent of school-holiday load, and pay 11 per cent of revenue.
A single person overpays by £112 a year. A low-income flat, £171. A family of five, £240.
A static caravan underpays by £512. A holiday cottage by £833. A second home with a swimming pool by £1,536.
Across Cornwall's roughly 21,000 second homes and holiday lets, that is about £15 million a year transferred from permanent residents to properties that are empty for most of it.
It is worth being blunt about who those residents are. Cornwall has among the lowest household incomes in England and, thanks to the coastline, the highest water bills. This is a cost-of-living transfer running precisely the wrong way: from working households in Camborne and Redruth to holiday properties in Rock and Padstow, invisibly, through a tariff nobody has ever had to defend because until now there was no alternative to defend it against.
Which raises a question I think is worth putting to a minister: should a second home in Cornwall receive an £800 subsidy from local water bill payers? Should one with a swimming pool receive £1,500?
Nobody chose this. It is what happens when you charge an annual average price for a service whose costs are set by six weeks in summer, and until now there was no way to charge anything else. That is exactly what the meter in my front garden changes.
What a seasonal tariff would actually do
The obvious design — cheap in winter, expensive in summer — barely works. South West Water is trialling one now. I modelled it and it moves a resident's bill by about two per cent, because everybody uses roughly half their water between April and September. You cannot give someone a meaningful discount on winter water when only half their consumption is in winter.
Two changes make it work.
The first is to seasonalise sewerage as well as water. Sewerage is charged volumetrically alongside water anyway, at 95 per cent of metered volume, and in the South West the case for it is stronger than for water: the regulatory target is defined in bathing seasons, and it is the summer visitor peak that loads coastal treatment works. That takes the reachable share of the bill from 41 per cent to about 84.
The second is to narrow the premium window to the school summer holidays — 18 July to 31 August. Forty-five days, 12.3 per cent of the year. A permanent resident uses 12.8 per cent of their annual water in that window: essentially their pro-rata share. A static caravan uses 42 per cent.
That gap is the whole mechanism, and it needs no registration, no means testing and no attempt to identify second homes. A tariff that is cheap for ten and a half months and expensive for six weeks self-selects. A permanent resident banks the discount. A property that only fills up in August never does. And it might, just might reduce the frequency of hose pipe bans too!
Run revenue-neutral with an eight-to-one differential, the results are:
Single person: −13.4 per cent
Low-income flat: −14.8 per cent
Family of five: −12.2 per cent
Static caravan: +54.7 per cent
Second home with a pool: +48.9 per cent
For ten and a half months of the year every household in the region would pay £3.19 a cubic metre instead of £7.47 — well under half. My own bill would fall from £78 a month to £66.69. And because almost everyone pays by monthly direct debit, I would never see the peak rate at all; the number on my bank statement would simply be smaller.
Push harder — bring holiday parks and campsites in from the non-household market where they currently sit entirely outside the household tariff, and roll the standing charge into the volumetric rate — and a South West resident lands at £652 a year against a national average of £639. The gap closes almost completely.
The regional average bill does not change, because the reform is revenue neutral. What changes is who pays it.
Four things that are wrong with this
I would rather say them than have them said back to me.
It might work too well. Allowed revenue is fixed. If holiday lets respond to the signal by fitting flow restrictors and recycling greywater, the peak-season revenue falls and rates must rise to collect the same money. A 50 per cent reduction in leisure peak use would halve the resident's saving, from £232 to £116. The tariff cannot both raise money from visitors and stop visitors using water. The resolution is that these work on different timescales — less peak demand means less of that £761 million needs building, which is a better outcome, not a worse one — but the transfer is transitional and the article that pretends otherwise is lying.
It is rougher on families. The school holidays are precisely when children are at home. A family of five uses 19.6 per cent of its water in the window against a single person's 17.6, so it gains least. Every resident household still gains, but the gradient runs the wrong way and it would need watching.
It is not only about tourists. A detached house with a large garden in Exeter is also underpaying — by about £91 a year. The tariff cannot tell a holiday let from a sprinkler, and it should not pretend to. This is a charge on summer peak use, not on second-home ownership, and it should be argued for on that basis.
It would make one existing absurdity worse. Thirty minutes with a hosepipe uses 540 litres and costs £4.03. Of that, £2.40 is a sewerage charge — on water that goes into the soil and never sees a sewer. That is not an error; it is the published 95 per cent rule. Sixty per cent of the cost of watering a garden in Devon buys a service that is not delivered. A seasonal tariff would amplify that, and the honest fix is to pair it with a proper allowance for outdoor use — which, conveniently, is another thing smart meters would finally make administrable.
The carbon footnote
While I was at it I worked out the emissions. That 30-minute hosepipe run is 103 grams of CO2e — about 2.2 miles in an electric car. Our entire household's water and sewerage for a year is 37 kilograms, roughly a quarter of one domestic return flight.
Which is the point. In electricity, price and carbon move together, so when I optimised the bill I decarbonised as a side effect. In water they are almost completely decoupled. That £4.03 is not four pounds of carbon. It is four pounds of reservoir, trunk main and treatment works, sized for the afternoon in August when everyone in Devon waters the garden at once.
Carbon does not care when you use water. A January hosepipe and an August hosepipe emit the same 103 grams. Capacity cares enormously. Which is precisely why the signal has to be seasonal rather than simply higher — there is nothing to fix in February.
What should happen
The meters are coming regardless. Ten million of them, £1.7 billion, by 2030. The question is whether the water industry does anything with them beyond sending more accurate versions of the same bill.
Electricity got smart meters and then got time-of-use tariffs, export payments and households that generate more than they consume. It took about a decade and it required somebody to decide that the meters were for something. Water is getting the meters and, so far, nothing else. Ofwat's charging trials are real but tiny — 1,653 households on a seasonal tariff at South West Water, 601 on a summer peak rate. There is no plan I can find for what happens after them.
Meanwhile 1.8 million people in Devon and Cornwall are paying the highest bills in England for 860 miles of coastline that 11.8 million people a year enjoy. In the middle of a cost-of-living squeeze, a family in Camborne is subsidising a swimming pool in Rock by an amount neither of them has ever been told about.
My smart meter goes in next month. It will be able to tell South West Water, to the hour, exactly when I use water. I would quite like somebody to have a reason to look.
The model behind this piece — daily demand simulation for seven household types, calibrated against South West Water's published assessed charges and its WRMP24 critical-period factors, with tariff designs solved for revenue neutrality — is on GitHub. Rates are from the 2026/27 household charges schemes of South West Water, Wessex Water and Northumbrian Water; bill comparisons from Water UK; investment allowances from Ofwat's PR24 final determinations; emissions factors from the UK Government's 2026 conversion factors. Assumptions, and the places where they are load-bearing, are documented alongside it.