Can private schools teach the world?
Despite international pledges, there are still millions of pupils around the world without a basic education.
Could the private sector be a more effective way of reaching these millions of pupils who are missing out?
Should donors be supporting low-cost, low-fee private schools, rather than trying to build state education systems?
Or would such schools further deepen the barriers to education for the poorest and most excluded?
A meeting at the House of Commons last week heard strongly opposing views on such private sector involvement. Here are some of the arguments for and against, from Sir Michael Barber and Professor Keith Lewin.
SIR MICHAEL BARBER: 'LOW-COST PRIVATE SCHOOL IS ANSWER'
"Getting every child in the world into primary school and learning is proving to be a tough challenge.
In spite of repeated global commitments, we are not currently on track to meet the Millennium Development Goal of universal primary education by 2015. At present over 60 million primary children are not in school at all and another 250 million or more are in school but barely learning anything worthwhile.
Apart from being a betrayal of human rights this is storing up huge problems for those children and their families, and for all of us. In some countries such as Nigeria the number of out-of-school children is actually rising.
In spite of the best efforts of some non-governmental organisations (NGOs), some governments and the international donor community, it is clear therefore that the current approach is not working.
Around the developing world poor parents themselves understand this all too clearly. They know that the only route out of poverty for their children is a good education. When the state fails what do they do?
The answer for many is to choose a low cost private school, charging perhaps 5-10 dollars a month. There has been a vast expansion of the low cost private sector in the last 15 years.
Across the Punjab in Pakistan, out of 20 million children around 9 million are in these schools; and in Lahore, Karachi and Delhi in India, around 70% of children attend them.
By Western or Pacific Asian standards these are poor schools, but they are usually better than the government alternative, often significantly so.
So many poor parents have voted with their feet that it is no longer possible to solve the problem of universal primary education without taking the low-cost private sector into account. The cat is out of the bag.
Some leap to the conclusion that if the public sector is not up to the job then it should get out of the way and leave it to the private sector, but a purely private approach is no more likely to work than a purely public one.
The road to hell in education is paved with false dichotomies. What we need is to unleash the moral force, energy, commitment and investment that the two sectors combined could bring to bear.
What would this look like in practice? First, governments need to ask the right question: "How do we get every child in our country a good education as fast as possible?" They need a reform strategy that will benefit every child, not a list of initiatives or boutique projects.
As many countries have done in other sectors such as health and transport, this means government should be funder (and yes they will need to spend more) regulator and quality assurer of education - but not the sole provider.
Often NGO and for-profit providers of schooling are doing a better job; so learn from them, make them part of the system and see them as part of the solution. On enrolment drives, urge parents to get their children into a school, regardless of whether it is public or private.
While millions of very poor parents choose low-cost private education, some of the poorest can't afford even low fees. This group often can't afford 'free' government schools either because there are hidden fees, for uniforms or sometimes bribes.
One solution here is to learn from the Punjab Education Foundation in Pakistan which, with Department for International Development support, is providing targeted vouchers to poor families whose children aren't in school.
It is an experiment involving 140,000 children. The early signs are encouraging. Any school that accepts the vouchers has to demonstrate that children are making progress - and they are.
Alongside this of course every effort should be made to improve government schools too. This means getting the basics of management right - for example, it is an outrage that across India teacher absenteeism is often 25%.
It means fixing facilities by getting the money devolved to school level. It means tracking the data on enrolment and attendance at every level and acting when there is a problem. It means appointing managers at district level on merit, not for political reasons. And it means tackling corruption head on wherever necessary.
The major donors need to encourage governments to set ambitious goals and get behind strategies such as these.
They should reject the notion that money alone is the answer or that all it takes is a little more patience. How long do we think we've got? Equally they should reject the notion that it can't be done. We know what to do; we just need the courage and will to get on with it."
Sir Michael Barber is also chair of the Pearson Affordable Learning Fund.
PROFESSOR KEITH LEWIN: 'FEES WILL MAKE THE POOR POORER'
"The proposition that development aid should be used to support private, fee-paying, for-profit schools to educate poor children in low-income countries sounds odd to many people.
But there are those in the debate about universal access to primary education who have managed to convince themselves that this makes sense.
And they are not talking about "free schools" or academies, in the UK sense, or other types of grant maintained arrangements, but truly private for-profit, fee-paying schools, targeted on the poorest.
As I understand their rhetoric, it seems to argue that such private schools are sweeping up the poorest un-enrolled children all over Asia and Sub-Saharan Africa.
But the rising numbers of children in school in poor countries, up by more than 50 million children since 2000, have not been mostly enrolled in such fee-paying private schools.
This progress has been overwhelmingly the result of public investment in public schools.
The greatest enrolment gains have been after school fees have been abolished in government schools. Those who have not noticed this must have been unusually somnambulant.
Where there has been growth in low-price private school numbers, it is often because of specific circumstances, such as where a high level of migration into cities has not been matched by an expansion in public services.
When there is access to state schools for migrants and more state schools are built, such as in Beijing, the demand for low price fee-paying schools falls away.
Those proposing more support for private for-profit schools also seem to believe that such schools consistently out-perform equivalent state schools.
But it's impossible to generalise about whether for-profit or public schools are better, because the research produces contradictory results. There is a large overlap between the performance of different types of school.
Some studies show for-profit schools achieving better academic results than very poor state schools. But this may not be much of an achievement if it means being the equivalent of three years behind the national curriculum rather than four.
A school only has to be a bit better and affordable for it to begin to attract children from a worse school. Private for profit schools cluster in wealthier parts of poor or rural areas, since they follow the money.
Low accountability, poor governance, and little transparency afflict both private and state schools and are often associated with unacceptably low levels of achievement.
Even the title "private schooling" is a misnomer. There is only schooling which is privately financed. There is no "private physics" or "private history".
If some state schools are better than some private for-profit schools, which they invariably are, then the problem is not the ownership but the operation.
There are also questions about the quality of teaching - as those promoting private schools also seem to believe that teachers working for a fraction of a state school teachers' salary can still be as effective and not be exploited.
Such for-profit schools can only be low price if they pay teachers little more than a dollar a day and exploit local labour markets for unemployed, untrained secondary school leavers who, more often than not in South Asia, are female.
Turnover of such teachers is typically high, career development is not available, there is no job security, and pedagogy unlikely to be innovative.
If teachers are at the centre of efforts to improve quality, as most recent international reports suggest they must be, then such fee-paying schools will have trouble competing for the best teachers.
Another consequence of fees is that children and households desperate for educational advancement may borrow in the micro-credit markets that typically charge annual interest rates of at least 40%.
There are also claims that the introduction of private for profit schools means a wider choice - but the opposite is the case for the poorest families.
The growth of private, fee-paying schools can lead to very stratified choices linked to affordability. Competition between private schools can see them differentiating themselves by charging higher fees - with higher quality being associated with higher fees.
The widest school choice for poor households comes when there are no tuition fees at all.
Moreover, every pound households spend on school fees and other school charges is a pound not spent on child nutrition, health care, medicine, clothes, shelter or sanitation and clean water.
There are better alternatives. For instance, there are ideas recommended by the CREATE programme - the Consortium for Research on Educational Access, Transitions and Equity - which is funded by the UK's Department for International Development.
This sets out a programme that includes schools without fees, state and community-led enrolment drives, a more relevant curriculum, fully professionalised teachers, decent buildings and learning materials, better governance. This is where aid should be focused and not on the growth of for-profit services.
Private, for-profit schools are a reality - and they perform both well and badly. They do contribute to the education of those who can afford them, but they have limits to their growth, depending on their affordability and the quality of the free state schools with which they compete.
But they do not reach the poorest reliably, and have no future in countries aspiring to become middle income, except for households well above the poverty line who can pay for advantage.
The sustainable 10-year ambition should be universal access to schools with choice unrestricted by affordability.
This is what successful developing countries have achieved. Charging poor children for their education will never make sense, it will only reduce their choices and make their households poorer."
These are abbreviated versions of speeches given at a meeting in the House of Commons hosted by the All Party Parliamentary Group on Global Education for All.