The Conservatives' controversial introduction of a competitive market in state education is paying off, according to new research.
The study, by economists at Lancaster University, claims that schools are better run than they were 10 years ago and that competition between local schools improves their performance.
Reducing class sizes could make a significant difference, say the economists, potentially leading to cuts in truancy rates as well as better exam results.
These findings have emerged from the first study to produce "efficiency" ratings for every non-selective school in England.
They will strengthen the Government's determination to press ahead with market reforms of its own, despite pressure from Labour traditionalists to abandon the Conservative approach.
The research is based on an analysis of GCSE results and attendance rates for 2,657 English secondary schools. It explores the effects of pupils' background and teaching quality on school performance and takes account of school type, proximity to other schools, gender balance and spending on teachers, books and equipment.
Using a complex formula widely employed by economists to measure "technical efficiency" in industry, the researchers have taken school performance data since 1993 - when league tables were introduced - to give each school its own efficiency rating.
Their key finding is that schools' overall efficiency - in terms of GCSE results and pupil attendance - has increased by 1.3 per between 1993 and 1998. More controversially, they also say that increasing competition has helped drive the improvement, with schools which compete with a large number of others tending to be more efficient.
The study suggests that the trend is for the gap between the best and worst performers to narrow. This is because schools with worst performances in 1993 have, in general, improved the most.
"Competition improves performance at the bottom end of the ladder even more than at the top," say researchers Steve Bradley, Jim Millington and Geraint Johnes .
Their controversial findings conflict with another study carried out by Rosalind Levacic of the Open University. This research, using a smaller sample of 227 schools, has adopted a different methodology, taking improvement of exam results over time as the measure of school performance. It also takes account of headteachers' own assessment of the degree of competition between schools in their area.
The OU study, preliminary results of which were reported in The TES last year, concludes that there is no evidence to link results with the degree of competition.
But, both studies agree that efficiency has improved since 1993.