Can the UK nationalise water, rail, gas and electric, and would it make anything cheaper?

The question underneath all of this#

Prices keep going up, wages barely keep pace, the water companies are a mess, and every few months someone says "just nationalise it". I wanted to know if that is actually true, or if it is a slogan that skips over the hard part.

So this post tries to answer a handful of questions in order. Can it be done for water, gas and electricity, and what about railways and the other natural monopolies? Can we change how the market sets energy prices? What can realistically be done about the cost of living? What about tax? Where is the post-COVID debt now? And if the state does own things, how do you stop it being run badly?

I am not an economist and this is not financial or political advice. I have tried to use primary sources (ONS, OBR, Ofgem, HM Treasury, Defra, the IFS) and I have flagged where a number comes from someone with a stake in the argument. Everything below is as of 30 October 2026.

Is nationalisation possible?#

Legally, yes. Parliament can pass a law taking an industry into public hands. It has done it repeatedly. The government nationalised British Steel in July 2026, and took stakes in banks in 2007 and 2008. Rail passenger operators are moving into public ownership one franchise at a time.

The IFS makes a useful point: "public control" is a spectrum, not an on/off switch.

• Regulation: private companies, but the state sets prices and standards (Ofwat, Ofgem).
• Franchising: private firms run a service to a government specification (buses in Greater Manchester).
• Partial ownership: the state holds a stake (Sizewell C, Great British Energy).
• Full nationalisation: the state owns and runs it (Scottish Water, and now most rail passenger services).

The current government is deliberately in the middle of that spectrum. On 29 September it said the Water Bill will remove the cap on government shareholdings in the big water companies, which is currently 1% to 3%. It has said it will not fully nationalise the whole industry. For energy, the headline is Great British Grid, a publicly owned company that will compete for grid connection projects alongside the private network owners.

So the honest answer is "yes, it is possible, and the government is choosing a partial version of it". The interesting part is what each version would cost and what it would actually change.

Water: the strongest case, and the most expensive one#

Water is the textbook case for public control, for a simple reason. You cannot shop around for a water company. It is a regional monopoly, and the usual benefits of competition do not apply.

It is also where the private model looks most broken. Thames Water carries roughly £20bn of debt and has drawn £2.65bn of a £3bn emergency facility. This week its creditors approved a rise in that facility to £5.4bn to keep it going into autumn 2027. The first £3bn was lent at 9.75%, and campaigners warn that the interest alone could add around £40 to each household's bill. The creditor group says it will not be added to bills.

What would full nationalisation cost?#

Nobody agrees, and the disagreement is mostly about how to value the assets.

• Defra: about £100bn, using the industry's regulatory capital value as a stand-in for what the debt and equity are worth.
• Frontier Economics: £144bn by 2030, using the same method but projecting forward. Note that this was commissioned by Thames Water's creditor group, who have an interest in a high number.
• The OBR: about £78bn added to public sector net financial liabilities (£91bn of liabilities less £12bn of financial assets). That is a different measure, not the purchase price.
• A law professor writing for Common Wealth: "close to zero", on the argument that the compensation owed could be very low. Other estimates sit in between.

The IFS also points out that the up-front number matters less to the fiscal rules than it sounds, because the rules look forward. What really lands on the public finances is the ongoing bill, and the sector needs a huge amount of investment. One estimate puts it at over £290bn of enhancement spend over 25 years.

Why people still make the case#

The financing argument is the strongest one. An academic working paper by David Hall finds that finance costs (shareholder returns plus interest) took about 35% of customer bill revenue in England and Wales in 2023-24, versus about 8% at publicly owned Scottish Water. The paper's estimate is that public ownership could save £3bn to £5bn a year in finance costs. Hall's work is advocacy research, so treat the exact figure with care. The direction is still hard to argue with: the state borrows more cheaply than a highly geared private company.

Scottish Water is the proof of concept we already have. Household charges are about £490 against £603 in England and Wales. Its own regulator's 2024-25 report shows it doing better than the English and Welsh average on supply interruptions, pollution incidents and internal sewer flooding, and worse on leakage, per-person consumption and discharge permit compliance. That is a good result, not a miracle.

What I think is realistic#

A single step to nationalise every water company at £100bn or more looks unlikely and is probably not necessary. The more realistic route is the one being taken: abolish Ofwat in favour of a single regulator, lift the cap on state shareholdings, use special administration for the companies that fail, and move toward public control one company at a time. Thames Water is first in the queue. Special administration has never been used in the water sector, and the Environment, Food and Rural Affairs Committee has said the grounds for it are too vague to rely on while the company is technically solvent.

Gas and electricity: nationalising the shop does not fix the bill#

This is where the slogan and the numbers part ways.

Here is where the money goes in the Ofgem price cap for a typical dual-fuel household from 1 October (£1,723 a year):

• Wholesale energy: 47%, about £812
• Networks (the wires and pipes): 24%, about £417
• Supplier operating, debt and industry costs: 17%, about £291
• Policy costs: 6%, about £96
• Supplier profit allowance (EBIT): 2.7%, about £46
• VAT: about £41, all on gas, because electricity VAT is zero until 31 March 2027

If the state ran every energy supplier tomorrow, the most it could save from the profit line is about £46 a year, and the supplier would still have to buy energy at wholesale prices. Gas is traded globally. Most of what drove the last rise (wholesale up 11% in three months) was the conflict in the Middle East and LNG supply fears. The UK is a net importer of gas. A British-owned supplier does not change that.

Owning more of the generation and network helps in other ways, such as stability and long-term investment. But it does not change the price of a cargo of LNG.

Can we change how the market sets the price?#

This is the better question. Under the current system wholesale electricity is priced by the most expensive generator needed to meet demand, which is very often gas. That is called marginal pricing. The government says gas still sets the price around 60% of the time, down from about 90% at the start of the decade, and expects around 50% by 2030.

The result is that electricity costs about 26.3p per kWh and gas about 8.0p per kWh. Electricity is over three times the price per unit, which makes it harder to justify switching to heat pumps or electric heating, even though most of the grid is increasingly not gas.

The options on the table:

• Zonal pricing (different prices in different regions). The government rejected it in 2025 in favour of "reformed national pricing". Parliament's Energy Security and Net Zero Committee says that package may save a typical household only around £20 to £40 a year by 2040, and mostly by cutting network costs rather than wholesale costs.
• Fixed-price contracts for existing low-carbon generators. The government is offering voluntary long-term contracts to generators not already on them (around a third of British power), so their output is not repriced every time gas spikes. This is the most direct way to "break the link".
• Taxing windfalls. The Electricity Generator Levy has been raised from 45% to 55% to take more of the extra revenue when gas spikes.
• Pay-as-bid instead of pay-as-clear. The committee wants the government to publish its work on this by the end of 2026. The risk is that generators simply bid higher to cover what they expect, so it is not obviously a saving.
• Putting gas plants under a regulated asset base or strategic reserve, so gas is paid to be available rather than setting the price. Spain and Portugal capped the gas price used for power in 2022 with mixed results.
• Moving levies off electricity. Policy costs sit mainly on electricity (about £72 of the £96, with gas carrying about £24). Moving them toward gas or into general taxation would narrow the price gap and make electrification cheaper. Nesta has costed options like this. The Budget did something small in that direction: the electricity VAT cut is extended to 1 July 2027, and the planned gas bill levy for hydrogen is scrapped.

My view is that there is no clever market redesign that makes this cheap overnight. The big, durable lever is the one the government itself names: build enough clean generation, storage and grid that gas becomes a backup rather than the price-setter. That is slow, and it needs the grid investment Ofgem puts at around £70bn between 2025 and 2031.

Railways and the other natural monopolies#

Water and energy are not special cases. They are examples of a bigger category that economists call natural monopolies: networks that are so expensive to build that it makes no sense to build two of them. Nobody wants two sets of rails between Manchester and Leeds, two water mains down one street or two sets of pylons across the same fields. The IFS is clear that this is the strongest argument for public control, because the usual benefits of competition simply do not apply.

The useful trick, which Britain has used inconsistently, is to separate the network from the service on top of it. The wires, pipes, track and ducts are the monopoly. The trains, energy retail and broadband packages running over them can often be competitive. Where that separation is clean, you can regulate or own the network and let competition work above it. Where it is bundled together, as in water, you end up with one regional monopoly for everything.

Railways#

Rail is the clearest example of Britain trying every option over thirty years.

• The track is already public. Network Rail has been a public sector body since September 2014. It borrows directly from government and is funded through a five-yearly review run by the independent regulator, the Office of Rail and Road. The government supported its operations, maintenance and renewals with about £8bn in 2023-24, and passenger services with about £4bn.
• The trains were privatised in the 1990s on a promise. Franchising was supposed to bring subsidy down towards zero. A critical academic study found support actually rose from £2.93bn in 1994-95 to £5.28bn in 2013-14 (in 2013-14 prices). That study is part of a critical literature, and the passenger numbers also grew a lot in that period, so it is not the whole story. But the "no subsidy" promise did not hold.
• Operators are now moving back to the state as franchises expire. South Western Railway went first on 25 May 2025, then c2c, Greater Anglia, West Midlands Trains, Govia Thameslink and Chiltern (on 20 September 2026). Great Western Railway follows on 13 December 2026 and Avanti West Coast on 7 March 2027, with the transfer completing by the end of 2027.
• Great British Railways is the real change. The Railways Bill, introduced in November 2025, would bring 17 organisations, including Network Rail, into one public "directing mind" that runs track and train together, with powers over fares and ticketing and a new passenger watchdog.

A point that is easy to miss: this is cheap to do. The IFS notes that when a franchise ends, the old operator is not owed compensation. No company changes hands, the contract just is not renewed. The government is also not buying the trains themselves, which are leased from private rolling-stock companies, because Labour argued in 2024 that the compensation would be too high. So rail shows that "nationalisation" can be a very different thing depending on which part you take.

It is also too early to say it works. The five London-serving operators taken over since May 2025 all show higher cancellation rates and more delays than just before takeover, though only two have had a full year. Across all 24 operators over the year to August, publicly owned ones averaged fewer cancellations (3.53% vs 4.14%) and fewer late stops (18.76% vs 23.31%) than private ones, but that is a crude average and does not prove ownership was the cause. The Southeastern model, which put track and train under one leadership team, reports punctuality up from 84.2% to 85.7% and cancellations down from 2.5% to 2.1%. Stephen Glaister, a former chair of the Office of Rail and Road, told the BBC that a decade is needed to judge it. Many delays come from signals, track and weather no matter who runs the trains.

My read: the case for rail is not mainly about ownership, it is about integration. Track and train had separate owners with separate incentives, and each could blame the other. One accountable body fixes that, whether it is public or a well-regulated concession. The gains will come from new trains, staff and investment in the network, not from the logo on the carriage.

The wires and pipes in energy#

This is where I think public involvement has the strongest case in energy, much stronger than in retail. Networks are 24% of the price cap, about £417 a year, and they are going up because the grid needs rebuilding.

Ofgem's RIIO-3 price controls run from April 2026 to March 2031. They allow a return on equity of 5.70% for electricity transmission and 6.12% for gas. Ofgem expects network charges to rise by around £108 for a typical household by 2031, offset by about £80 of savings from lower constraint payments (money paid to generators when the grid cannot carry their power), so a net rise of about £30. The electricity transmission investment could reach around £70bn, with at least £44bn already committed.

That is a lot of long-term, low-risk infrastructure spending where the main costs are financing. This is the same logic as the water finance argument, and it is why Great British Grid, a public company competing for grid connection projects, is a more interesting idea than a public energy retailer. The caveat is the same too: borrowing for it still counts as public debt, and the state is not immune to cost overruns.

Broadband#

Openreach, which owns most of the UK's telephone poles and ducts, is the example of a third route: regulate a monopoly so that competitors can build around it, and step back when they succeed.

Ofcom's 2026 to 2031 review keeps Openreach regulated where there is no network competition, lets rivals use its ducts and poles at cost-based prices, caps the price of products up to 80Mbit/s, and says it will deregulate if competition emerges. Full-fibre coverage went from 24% of premises in May 2021 to 78% (23.7m premises) by July 2025, with investment of £3bn to £6bn a year. That was done without the state owning the network. It is a decent example of regulation done well, though it only works because duplicating fibre is possible in dense areas. In rural areas, where it is not, Ofcom is adding minimum repair and installation standards.

Buses, airports and the rest#

• Buses are not a natural monopoly in the strict sense, but franchising gives public control without public ownership. London has used it since 1995, Greater Manchester launched it in 2023, and Liverpool, the West Midlands and Wales have announced plans.
• Heathrow is regulated by the Civil Aviation Authority, which has looked at competition and franchising models because there is little realistic competition on the airport itself.
• Scotland and Northern Ireland already run water publicly, and Northern Ireland runs rail and buses publicly, which gives a ready-made comparison.

A rough test for any natural monopoly#

I would ask five questions before deciding between ownership, franchising and regulation:

1. Is there a real network that cannot sensibly be duplicated? If not, competition is probably the cheaper answer.
2. Can the service on top be competitive? If yes, keep it that way and regulate or own only the network.
3. Is financing the main problem? Highly geared private networks pay a lot for money. Public borrowing is cheaper, though it lands on the national debt.
4. Is the investment long term and lumpy? Grid, track and pipes need stable commitments for decades. That favours public funding or very stable regulation.
5. Is the regulator actually working? If the regulator cannot discipline a firm, as with Ofwat and Thames Water, the case for moving up the spectrum gets stronger.

That is not a full theory, but it already explains why rail track and the grid are strong candidates, why energy retail is not, and why water sits in the middle.

What can realistically be done about the cost of living?#

Let me start with where we actually are.

• CPI inflation was 3.5% in the 12 months to September, up from 3.1% in August. Transport is up 6.6% (fuel is the driver) and housing and household services 5.3%. Food is only up 1.3%.
• Average regular pay is growing 3.5%, which is only 0.6% in real terms. Wages are barely ahead of prices.
• The energy price cap is up 4%. Pay by direct debit and you are on £1,723 a year. Pay by standard credit and it is £1,861.

The Chancellor's 28 October Budget was explicitly a "breathing space" package: a one-off £100 household energy payment for council tax bands A to D in England, the electricity VAT cut extended, tariffs removed from more than 100 products, a National Living Wage rise to £13.21 in April 2027, and a request that Ofgem abolish gas standing charges. The IFS called the energy payment relatively small and badly targeted, because it is based on council tax bands that date from 1991.

What actually moves the needle over time, in my view, comes down to a short list:

1. Energy. Insulation, heat pumps and solar do more for a bill than any one-off payment. The Budget's Warm Homes loans claim up to £550 a year off bills for households that install solar, batteries and heat pumps. That is only useful if people can afford the up-front cost, which is the point of the loans.
2. Housing. Rent and mortgage costs are the biggest bill for most people. The only lasting fix is more homes, and that is a planning and construction problem, not something nationalisation answers by itself.
3. Wages. Real pay growth of 0.6% is too low. Productivity growth is what raises it. Nobody has a quick answer, but investment in grid, housing and transport feeds it.
4. Targeting. Spreading small payments across 80% of households is easy to run and poor value. Better data on who is hit hardest would allow bigger help for fewer people. That takes investment in data the government has not made.
5. Stopping the leaks. Standing charges, poor-value tariffs and finance costs in monopoly utilities all add cost with nothing to show for it.

None of this is dramatic. There is no one lever that fixes the cost of living, and anyone who says there is has not looked at the Ofgem breakdown.

What can be done about taxes?#

The honest starting point is that the tax burden is heading for a record high regardless of what anyone promises. The OBR has taxes at around 37.2% of GDP in 2026-27, passing the previous historical high in 2027-28 and peaking at 38.6% in 2031-32.

A lot of that is not rate rises. It is fiscal drag. The income tax personal allowance (£12,570) and higher rate threshold (£50,270) have been frozen since 2022 and stay frozen until April 2031. As wages rise, more of your pay is pulled into tax and more people cross into the 40% band. The OBR estimates the freeze raises somewhere around £55bn to £58bn a year by 2030-31 and drags millions more people into income tax and the higher rate. Nobody voted for a higher rate. The thresholds just stood still.

The Budget kept to its promise of no increase in the headline rates of income tax, National Insurance or VAT. It did raise taxes elsewhere:

• savings and property income tax rates go up by 2 percentage points from April 2027
• the bank corporation tax surcharge doubles from 3% to 6% for 15 months
• gambling machine duty goes up
• online marketplaces become responsible for collecting VAT owed by their sellers

The IFS counted 98 separate measures and was blunt about it: lots of small tinkering, another tax rise on landlords with no attempt to fix the underlying problems in how they are taxed, and a new Founders' Relief that "looks bonkers".

So what could realistically be done?

• Index the thresholds again. It is the most honest option and it is scheduled for 2031. It costs tens of billions a year, so it needs to be paid for.
• Fix council tax. It is based on 1991 property values, and it is why a payment tied to bands A to D is crude. A revaluation or reform is politically painful but it is the obvious structural fix.
• Simplify before adding. Fewer reliefs and special rates would raise revenue without raising rates, and make the system easier to understand.
• Tax the thing you want less of and cut the thing you want more of. Moving costs off electricity and onto less climate-friendly sources is an example of the same logic.
• Accept the trade-off. If you want a bigger state, including bigger public ownership, you pay for it. The only question is who, and when.

I do not think there is a painless tax answer. There is a reasonable argument about fairness between workers, savers, landlords, older people and companies, and a less reasonable one that assumes the money will appear somewhere else.

Post-COVID debt: where are we now?#

This is the bit people forget when they discuss new spending.

• The COVID bill. HM Treasury's 2026 cost tracker puts the lifetime cost of the government's COVID measures at about £385bn, with £380bn spent. That includes furlough, business support, health spending and expected losses on loans such as Bounce Back Loans.
• The jump. Borrowing hit £313bn in 2020-21. Public sector net debt went from about 84% of GDP to about 97% in a single year.
• Where it is now. The ONS says net debt at the end of September 2026 was £3,005bn, or 94.0% of GDP. That is 0.7 percentage points lower than a year ago, and provisionally the first time debt has gone above £3 trillion. The ratio is at a level last seen in the early 1960s.
• The cost of carrying it. The OBR's forecast for debt interest in 2025-26 was £111bn, 8.3% of all public spending. The 2026-27 forecast is around £124bn, or 3.9% of GDP. The Budget says the Middle East conflict has pushed borrowing costs up by an average of £14bn a year compared with the spring forecast.
• The outlook. The Budget forecast has debt edging up to around 97% of GDP in 2028-29 before slowly falling, and the government says it meets both its fiscal rules with a buffer.

Two things to keep in mind. COVID was not the only cause. The 2008 financial crisis took debt over £1 trillion in 2010, and the 2022 energy crisis and rising interest rates added to the bill. And the ratio is "stable", not "falling fast". A country with debt near 94% of GDP and interest at nearly 4% of GDP has less room to absorb shocks, which matters because it is exactly the room you would need to nationalise anything big.

If the state owns it, how do you make it work?#

This is the part the slogan skips. Public ownership is not good or bad by nature. The track record is mixed, and it depends how it is run.

The early rail results above show why you should not expect miracles: the first operators taken over got worse before any improvement, and the best early signal comes from integrating track and train, not from the change of owner. Scottish Water shows the other side, that a public body can do well over twenty years if it is set up properly.

Here is what I think the evidence says you need to get right:

1. A clear, narrow mandate. "Run the water network safely, at low cost, and invest to a plan" is a job someone can do. "Do everything that is good for Britain" is not. Mixed goals are how state firms end up serving ministers instead of customers.
2. Arm's length from ministers. The old nationalised industries were damaged by politicians using them to manage inflation, hold down prices and shave investment when the Treasury needed cash. Independent boards, long-term funding settlements and a rule against raiding the capital budget protect against that.
3. A strong, independent regulator. Public ownership does not remove the need for someone to say no. Scottish Water is still regulated by an independent body, which sets what it can charge and checks what it delivers.
4. Hard budgets and honest accounts. If a failing public company can always be topped up, there is no pressure to improve. Publish the accounts, set efficiency targets and let the National Audit Office look.
5. Benchmarking. You lose competition in a monopoly, so replace it with comparison. Scottish Water against the English companies is a natural yardstick, and its operating costs have fallen by over 30% since 2002 partly because they are measured against someone.
6. Pay and skills. Engineers, project managers and finance staff are expensive. A public body that cannot hire them will contract everything out at higher prices, or build badly.
7. Published performance data and a real customer voice. Leakage, pollution incidents, outages, complaints and delivery against plan should be public, on a regular schedule, with consequences when targets are missed.
8. Integration where it matters. Splitting track and trains hid accountability on the railways. The same logic applies to water, where one body should be accountable for supply, sewage, and the environment, not four.

None of that needs full ownership. Some of it, such as a stronger regulator, a single accountable body and published performance data, is available under the current model. That is why I think the government is right to be cautious about the form it takes and why the details in the 10-year plan, expected around December, and the 2027 spending review matter more than the headline.

Where I land#

• Water: public control is justified, because the financial model has failed and the monopoly has nobody to compete with. Full nationalisation at £100bn or more is probably not the first move. A stronger regulator, state shareholdings and special administration for the worst offenders is the path being taken, and it looks sensible.
• Gas and electricity: nationalising suppliers would change little. The wins are in getting gas out of the price-setting role, building clean generation and grid faster, and moving policy costs off electricity. A public body investing in the grid has a better case than a public energy retailer.
• Rail and other natural monopolies: separate the network from the service on top, and own or tightly regulate the network. Rail's big change is integrating track and train under one accountable body, and it is too early to call. The wires and pipes are a stronger case for public involvement than energy retail. Openreach shows regulation can work without ownership.
• Cost of living: there is no silver bullet. Cheaper energy, more homes and rising real pay are the things that count, and they are slow.
• Taxes and debt: the room for manoeuvre is small. Debt is 94% of GDP, interest is nearly 4% of GDP and the tax burden is heading for a record. Whatever the government does with ownership has to fit inside that.
• Making it work: ownership matters less than governance. A badly governed public monopoly will be as poor as a badly regulated private one. A well-governed one can be good, and Scotland has shown that in water.

I would rather argue for the boring version that works than the exciting version that sounds good in a speech. If you spot a number I have wrong or a source I should have used, get in touch.