Finland's schools system provides the best value for money in the world, according to a new report, which recommends that other countries consider paying teachers less and increasing class sizes to match their rival's "efficiency".
The controversial study by academics from the University of Sussex and the University of Malaga in Spain compares education outcomes in 30 countries - as measured by the Programme for International Student Assessment (Pisa) - against the resources provided by their governments. It ranks the UK in 11th place in a league table of "efficiency scores", with South Korea in second place and Brazil bottom.
The analysis (www.edefficiencyindex.com) has been praised by Andreas Schleicher, education director at the Organisation for Economic Cooperation and Development (OECD), which runs Pisa. However, it completely discounts factors such as levels of teacher training, teaching resources and the quality of school buildings as potential ways to significantly improve education. Instead, it states that higher teacher salaries and smaller class sizes are the "only two contributing factors, out of 63 different components studied, that have a demonstrable impact on education outcomes".
The report claims a "proven statistical link between teacher salaries or class size and Pisa scores". But it argues that the "efficiency" achieved by Finland, which has larger class sizes and lower salaries than the UK, demonstrates that it would be possible to increase the pupil-teacher ratio andor make "a modest cut" to pay and still match Finland's Pisa result.
The UK could either reduce average teacher salaries by 10 per cent or increase the average pupil-teacher ratio by nearly a quarter (23.8 per cent) to achieve the Finnish efficiency "target", the report suggests. It makes these recommendations despite describing the UK's current situation as "more efficient than effective".
Chris Kirk, chief executive of Gems Education Solutions, the company that commissioned the "efficiency index", said it provided data that could inform government policy. "It clearly shows that some countries spend their available resources more efficiently than others," he added.
But John Bangs, chair of the OECD Trade Union Advisory Committee's working group on education, described the study's conclusions as "very odd indeed".
"One of the things that Pisa says is you have to pay teachers properly," he argued. "If you raise class sizes and pay teachers less on the basis of efficiency, then the quality of the system itself is going to crash - and who is efficient then?"
The positions of OECD countries in the efficiency index broadly match their most recent Pisa rankings. But the Czech Republic and Hungary, which finished third and fourth respectively in the index, came 24th and 39th in the Pisa 2012 results.
In an introduction to the new report, Lord Adonis, a former Labour schools minister, notes that the two countries have "exceptionally low teacher salaries, in common with most of the former Eastern European states".
"This makes them less interesting to policymakers looking for models of excellence and reform," the peer writes. He warns that it is "important not to take simple lessons" from the efficiency index, but describes it as "invaluable".
Mr Schleicher, who also endorses the index, suggests in his introduction that the "apparent high degree of efficiency" of South Korea and Japan (which came fifth) may be owing to high parental spending on out-of-school tuition, which is not included in the analysis.
The OECD official has previously made statements that appear to contradict the importance that the report places on class sizes. In December, Mr Schleicher said that top school systems "when deciding where to invest.prioritise the quality of teachers over the size of classes".
The report's co-author, University of Sussex economics professor Peter Dolton, said the data showed that class size was a "core, core driver" of outcomes that far outweighed factors such as teacher training and school buildings when accounting for the differences between countries' performance.
But he conceded: "I am perfectly happy to accept that the data is limited, particularly with respect to detailed, fine-grained information about how teachers are trained and where they are recruited from."