The following piece appeared as a guest blog in the UK's Guardian  this past week.
A good public education system means public spending – but not necessarily public provision.
In OECD countries, more than 20% of public education expenditure goes to private institutions  – communities, non-governmental organisations (NGOs), faith-based organisations, trade unions, private companies, small informal providers and individual practitioners – and about 12% is spent on privately-managed institutions.
But does private participation mean higher quality education? Does it bring better exam results? Can it encourage greater equality?
Evidence shows that in the independent sector, where schools depend on fees, it is often the case that once you control for family background, the actual benefits of private schooling disappear. But in systems where access is not limited by selection or wealth, privately-managed schools can contribute to better outcomes.
In the Netherlands, 70% of enrolments are in "private" schools that receive a fixed amount of government funding per student (with extra funding for disadvantaged students). On average, families tend to be from a lower social class than those of pupils attending "public" schools, and yet test scores achieved are higher. The level of choice offered appears to provide incentives for Dutch schools to keep improving.
Japanese high schools use private tuition support, which has been shown to lower drop-out rates among students taking less academic study pathways.
The Charter schools in the US have had a real impact on narrowing achievement gaps. The Harlem Children's Zone , which combines schooling with community support such as help with healthcare and meals, could reverse the black-white achievement gap in maths, according to trials; the Knowledge is Power  schools have been criticised for only improving test scores through selection, but evidence shows that the largest gains are among young people with special educational needs and limited English.
The picture internationally is that involving the private sector can improve school performance – through competition, accountability and autonomy – as well as expand access. However, without strong systems of accountability, private schools with public funding aren't likely to produce large gains.
The best results come where competition is enhanced through choice, disadvantaged areas are targeted and there is plenty of autonomy at school level.
Any new approach – such as the free-schools model in the UK – needs to be subjected to rigorous evaluation of its impact. Small-scale pilots are needed initially, with investment going only to projects that have been proved to work.
And moving forward, each country has a lot to learn from others. Keeping a watch on the international picture, benchmarking education policies, will be important for raising standards and addressing inequality.
Harry Patrinos is lead education economist at the World Bank, on secondment as a visiting research fellow at the UK's CfBT Education Trust .
Photograph: Startraks Photo/Rex Features