There is a consensus in Sweden that their policy of school choice, introduced in 1992, has driven up standards. This has been the result of the expansion of independent schools from covering 1% to 10% of pupils. It is not just that the new schools have achieved good results but they have pulled up the results of the state schools competing with them.
However there is not unanimity in Sweden. Newsnight are pleased with themselves as they have got the top education bureaucrat in Sweden, a man called Per Thulberg, to tell them that the whole thing is a failure. But reading his paper, What influences education achievement in Swedish schools, upon which Newsnight based their report, one finds plenty of left wing assertions but not much in the way of solid evidence. There are general references to "research" showing this and that but no solid data. For instance the report says streaming in schools is a mistake because it has " a stigmatising effect."
Anyway the Burning out Money blog has done a very good rebuttal although I thought he was a bit unfair saying Michael Gove was "back-footed" by the BC attack. I thought old Govey did pretty well. If he had tried to quote the facts then Emily Maitlis would have thought he was very boring and shut him up.
BOM points out that the Swedish free schools get a Grade Point Average that is 20 points higher than the state schools.
What of the evidence of standards being driven up elsewhere?
A study by Anders Bohlmark and Mikael Lindahl of Stokholm University found that an increase in the percentage of free schools in an area increased pupil performance across all schools. Most of this increase was due to competition in the school sector, forcing all schools to improve their quality. Åsa Ahlin of Uppsala University found that a ten per cent increase in the number of children attending free schools led to a five per cent increase in Mathematics performance across the area.
See also this Bergstrom and Sandstrom study.
UPDATE.
Maria Rankka of the Swedish think tank Timbro emails to say:
"Per Thulberg is wrong. There have been academic evaluations of the school voucher reform and the fact is that schools in geographical areas where competition exists are performing better than schools in areas where there still is no competition."
The Swedish Association of Independent Schools have more details on their website. What helps comparisons is that not everywhere in Sweden has the free schools system:
The proportion of students in independent schools has grown considerably since the beginning of the 90s, although the sector is still very small. In school year 1990-91, about 0.9 per cent of all
Swedish pupils in compulsory education (ages 6-15, approximately) were enrolled in independent schools, whereas in 2007-08 the figure had grown to about 9 per cent. The same trend may be observed in secondary education (ages 16-18, approximately), where the share has grown from 1.5 per cent to 17 per cent during the same period.In about 210 of the 290 local councils in Sweden, independent schools “compete” with public schools run by locally elected school boards; as yet there are no independent schools in the other local councils. The urbanized areas of south and middle Sweden, in particular in the Greater Stockholm area, have the highest concentration of independent schools.
That makes it possible to judge the impact the free schools have on standards in the state schools.
Critics said that it is highly motivated parents who opt to take their children out of local council schools, thus leaving those schools with disadvantaged children, or that independent schools impose financial burdens on the school boards by disrupting their planning.
Where comparisons have been possible, independent schools have performed better in terms of knowledge and skills than local council schools, and have achieved this at a lower cost. This has inspired some local community schools to improve their organisation and teaching in order to improve results.