Do things get better when the state sells?
Short answer: It depends mostly on the market. Companies sold into competitive markets become more efficient and more profitable, according to most studies. When a natural monopoly, such as water or the rail network, merely changes owner, results are weaker and depend on how it is regulated. Research knows least about quality in healthcare, schools and care. Who gains from the sale itself varies: buyers of Royal Mail made 750 million pounds on the first day, while those who bought Telia in 2000 lost most of their money within two years.
More on the state as owner: State-owned companies shows what the state owns and has sold, The banks mortgage rates, Schools independent school profits, Who owns the homes when rental flats become housing cooperatives, Elderly care where the private providers are, Infrastructure how rail deregulation went and Corruption and procurement how public procurement works.
What the studies say, outcome by outcome.
Privatisation means two different things: the state selling a company, and the public sector buying a service from a private firm instead of providing it itself. Studies diverge most between the two, and between markets with and without competition.
-
Productivity and profitability: usually better
A review of a large number of studies from around the world finds that privatised state companies generally become more efficient, more profitable and financially stronger. The effect is clearest for firms in competitive markets; for regulated monopolies the picture is mixed.
Megginson and Netter (2001), From State to Market, Journal of Economic Literature 39(2).
-
Cost when municipalities contract out: no clear saving
A meta-analysis of studies on water and waste finds no systematic link between private provision and lower costs. Competition for contracts is often weak, and monitoring costs money.
Bel, Fageda and Warner (2010), Journal of Policy Analysis and Management 29(3).
-
Quality: depends on whether it can be written into the contract
A private firm has stronger reasons to cut costs than a public agency. That is good when quality can be measured and required, but it can come at the expense of quality that cannot be written into a contract. The theory is applied to prisons, and the same argument is made about care and schools.
Hart, Shleifer and Vishny (1997), The Proper Scope of Government, Quarterly Journal of Economics 112(4).
-
Jobs: small effects on average
In around 30,000 firms in Hungary, Romania, Russia and Ukraine, privatisation to domestic owners changed employment by roughly zero. With foreign owners it rose. There are few studies of wages and working conditions after privatisation in Sweden.
Brown, Earle and Telegdy (2010), Economic Journal 120(545).
-
Distribution: often more unequal, at least in the short run
A review of privatisations in developing and transition economies finds that most made the distribution of wealth and income more unequal, at least in the short run, while access to electricity and water sometimes improved for poor households. For Britain, Florio finds that private ownership as such changed long-run trends in prices and productivity little, and that the gains went mostly to shareholders.
Birdsall and Nellis (2003), World Development 31(10); Florio (2004), The Great Divestiture, MIT Press.
-
Swedish welfare services: the evidence is thin
SNS's review of research on competition in healthcare, schools and care found the evidence so thin that no general conclusions could be drawn about gains or losses. At the time, almost one in five welfare employees had a private employer.
Hartman (ed.) (2011), Konkurrensens konsekvenser, SNS Förlag.
The rule of thumb that recurs in the literature: where customers can choose another supplier, a private company is forced to share its efficiency through lower prices or better service. Where they cannot, on a power grid, a water network or a railway line, the outcome depends on how well the state regulates and procures. The cases below broadly follow that rule.
Railways, water and the post.
| Financial year | Journeys, millions | Operational government support, £bn |
|---|---|---|
| 1990/91 | 1.8 | |
| 1993/94 | 2.3 | |
| 1994/95 | 735 | |
| 2006/07 | 10.5 | |
| 2013/14 | 4.9 | |
| 2018/19 | 1,753 | |
| 2019/20 | 8.1 | |
| 2020/21 | 388 | 20.9 |
| 2024/25 | 1,729 | 11.9 |
| 2025/26 | 1,830 |
Journeys in Great Britain from ORR, table 1220; figures from 2020/21 are provisional. Operational support in 2024/25 prices, CPI adjusted, ORR table 7270. Investment such as HS2 is excluded. An empty cell means the year is not included.
Railways
British Rail was broken up and sold from 1994 to 1997: the track to Railtrack, the trains to private operators on time-limited franchises. After the Hatfield crash in October 2000, when a broken rail killed four people, Railtrack went into administration in 2001, and the not-for-dividend Network Rail took over the track in October 2002. Since 2014 Network Rail has been classified to the public sector. Journeys have more than doubled since the mid 1990s, and regulated fares in 2026 were 8 per cent lower than in 1995 in real terms; all tickets together 6 per cent higher. Operational government support is at the same time several times larger than before privatisation. A review commissioned by the government found in 2011 that unit costs needed to fall by around 40 per cent to match the average in France, the Netherlands, Sweden and Switzerland.
ORR, table 1220, 7270 and 7180; NAO; McNulty (2011)
Railways taken back
A law from November 2024 lets the state take over passenger services as contracts expire. South Western Railway transferred in May 2025, followed by c2c, Greater Anglia, West Midlands Trains, Govia Thameslink and Chiltern. Great Western and Avanti West Coast are next, and all government-contracted services are due to be taken over by the end of 2027. A new state company, Great British Railways, is to bring track and trains together; the bill for it was before Parliament in autumn 2026.
Gov.uk, Great British Railways and the public ownership programme
Water
The water companies in England and Wales were sold in 1989 free of debt. Today the industry owes around 60 to 70 billion pounds, depending on source and year. There is no official figure for how much has been paid out since 1989; in 2023 the Financial Times calculated 78 billion pounds over 32 years. Thames Water had net debt of 18.5 billion pounds in March 2026, was fined 122.7 million pounds by the regulator Ofwat in May 2025 and in October 2026 reached a conditional agreement with a group of lenders. In July 2025 the government announced that Ofwat will be abolished, following a review of the whole water sector.
Thames Water, annual results 2025/26; Ofwat, May 2025; Gov.uk, July 2025
Royal Mail
The state sold 60 per cent of the postal service in October 2013 at 330 pence a share, 1.98 billion pounds. On the first day the share closed at 455 pence, 38 per cent higher, and the buyers made 750 million pounds. The National Audit Office judged that the department had been cautious on price, and the priority investors had sold almost half their shares within a few weeks.
Pharmacies, Telia, electricity, trains and healthcare.
Pharmacies 2009
Apoteket AB had a monopoly with 955 pharmacies. The state sold 465 pharmacies to four buyers and 150 to small business owners, receiving about 6 billion kronor. In September 2025 there were 1,399 pharmacies. Almost all new ones opened within two kilometres of an existing one, and the share of people with more than ten minutes by car to the nearest pharmacy is largely unchanged. Shortly after the reform it became harder to get medicine within a day, and the number of rural pharmacy agents fell; in 2025, 95.5 per cent of prescription items were dispensed immediately, slightly more than in 2019. Prices of prescription drugs are set by the state through TLV.
Govt. comm. 2012/13:33 on RiR 2012:10; Statskontoret 2013:7; TLV 2025, pp. 40 to 61
Telia 2000
The state sold 30 per cent of Telia at 85 kronor a share in June 2000, and over a million Swedes became new shareholders. It was at the peak of the dot-com bubble. By June 2002 the share stood at 21.60 kronor, a loss of around three quarters for those who bought. Here the gain went to the seller, the state.
Placera; Govt. bill 1999/2000:84
Electricity market 1996
Electricity was deregulated on 1 January 1996, but not sold: generation and trading were opened to competition, while the grids remained regulated monopolies. Vattenfall is still state owned. No overall assessment of what the reform meant for prices has been found; today prices are driven mostly by the Nordic and European market, see The electricity price.
SOU 2004:129, chapter 6
Trains
SJ is state owned, but since 2010 other companies may run passenger trains in competition, and the regions tender their services. When MTR Express began running Stockholm to Gothenburg in 2015, SJ cut its prices by 12.8 per cent on average, all in first class. Track maintenance has been competitively tendered since the 2000s, which cut costs by around 11 per cent without lower quality.
Vigren (2017), Economics of Transportation 11; Odolinski and Smith (2016), Journal of Transport Economics and Policy 50(1)
Choice in primary care and elderly care
With the primary care choice reform of 2010, private health centres could open freely, and the number of providers rose by almost 20 per cent. Visits increased more for the population at large than for those with the greatest needs, but the latter were not crowded out. In elderly care, private homes had fewer staff per resident but better results on several measures of how care is delivered.
Swedish Competition Authority reports 2012:2 and 2014:2; Vårdanalys 2013:1; Stolt, Blomqvist and Winblad (2011), Social Science & Medicine 72(4)
Independent schools
Independent school profits, grades and school vouchers have their own page.
Who wins when the state sells.
If the state sells at the right price, it simply swaps a stream of future dividends for a lump sum, and neither the state nor the buyer gains. Proceeds from Swedish sales have generally gone to paying down government debt, which lowers interest costs instead of the dividend coming in. What the state has sold since the 1980s, and what the companies pay out today, is on State-owned companies.
If the shares are sold cheaply, the buyers win. Privatisation shares have on average been sold at a discount, and Royal Mail is an example with figures: 750 million pounds on the first day. Telia in 2000 shows the opposite, where the state sold high and the small savers who bought lost.
Who owns the shares also changes over time. In Britain the number of individuals owning shares rose from around 3 million in 1979 to 11 million in 1990, largely through privatisations. Yet the share of UK quoted shares held directly by individuals has fallen, from 54 per cent in 1963 to 11.6 per cent in 2024, while overseas owners hold 58.8 per cent.
Megginson and Netter (2001) on the discount; NAO (2014) on Royal Mail; shareholder numbers for 1979 and 1990 are secondary figures from British surveys; ONS, Ownership of UK quoted shares 2024.
When state ownership may be better.
Research points to three situations where the case for state ownership is strongest. The first is natural monopolies, where only one network makes sense: power grids, water mains, rail track. There is no competition to push prices down, so the state must either own or regulate, and the British water companies show how hard it is to regulate a company that can borrow against its revenues. The second is quality that cannot be written into a contract. The third is supply that must work in a crisis, where an owner counting only profit does not take the full cost of an outage into account.
In Sweden the state owns the national power grid through Svenska kraftnät and the rail track through Trafikverket, while train services and electricity trading are open to competition. Municipalities own most water and sewage systems. Against state ownership: political goals can get mixed into business decisions, as the studies on state banks below show, and the state as owner does not find it easy to control costs either: the cost gap McNulty measured existed when the track was already run by the not-for-dividend Network Rail.
Hart, Shleifer and Vishny (1997); Megginson and Netter (2001); Bel, Fageda and Warner (2010); McNulty (2011).
Can a state bank push down interest rates?
The question has two parts: a state bank competing with private ones for customers, and state banks in general. Neither sets the policy rate, which the Riksbank decides on its own.
-
SBAB: mortgages in competition
SBAB started in 1985 and is wholly state owned. Its mandate is to increase competition in the mortgage market while earning a market return for its owner, so not to lend below cost. How large the mortgage margin is and how it has changed is on The banks. No study measuring how much SBAB has pushed down other banks' rates has been found.
Government Offices, annual report on state-owned companies; SBAB, annual report
-
Negative: slower growth and political lending
Across 92 countries, high state ownership of banks went with slower financial development and lower growth. In Italy, state banks charged lower rates to similar firms, most of all where the party controlling the bank was strong. A later study questions the link once countries' institutions are taken into account.
La Porta, Lopez-de-Silanes and Shleifer (2002), Journal of Finance 57(1); Sapienza (2004), Journal of Financial Economics 72(2); Andrianova, Demetriades and Shortland (2008), Journal of Development Economics.
-
Positive: lending when others pull back
State banks cut lending less than private ones in downturns, and in rich countries they increase it. Germany's KfW raised its commitments by around 75 per cent to about 135 billion euros in the pandemic year 2020.
Bertay, Demirgüç-Kunt and Huizinga (2015), Journal of Banking & Finance 50; KfW, Förderreport 2020.
-
The Riksbank is something else
The policy rate, which affects all other rates, is set by the Riksbank. Under the Instrument of Government no authority, not even the government, may decide how the Riksbank conducts monetary policy, and the goal is stable prices. A state commercial bank can lower its own rate, but not the country's interest rate level.
Instrument of Government ch. 9 § 15; Sveriges Riksbank Act (2022:1568), ch. 2.
Where the figures come from.
- Research reviews and studies: Megginson and Netter (2001), Bel, Fageda and Warner (2010), Hart, Shleifer and Vishny (1997), Brown, Earle and Telegdy (2010), Birdsall and Nellis (2003), Florio (2004), SNS Konkurrensens konsekvenser (2011), La Porta et al. (2002), Sapienza (2004), Bertay et al. (2015). Linked at each finding.
- Britain: ORR data portal (tables 1220, 7180 and 7270, retrieved 2026), NAO on Royal Mail (2014) and Network Rail, McNulty (2011), gov.uk on Great British Railways, Thames Water annual results 2025/26, ONS on ownership of quoted shares.
- Sweden: the National Audit Office and Statskontoret on the pharmacy reform, TLV follow-up 2025, SOU 2004:129 on the electricity market, the Competition Authority and Vårdanalys on primary care choice, Vigren (2017) and Odolinski and Smith (2016) on railways.
- State companies and SBAB: the government's annual report on state-owned companies, the same source as State-owned companies.
Figures transcribed by hand from the sources, autumn 2026. British pounds are not converted to kronor.