The study, from the National Foundation for Educational Research, is a blow to traditionalists and Chris Woodhead, the Chief Inspector of Schools. They have argued for a return to such traditional methods as phonics.
Since the average reading ability of nine-year-olds has remained much the same since 1948, the report's authors say, a return to the methods of the 1950s would not help.
Between 1987 and 1991, there was a small decline in reading standards, says the report. But by 1995, standards had returned to the 1987 level.
The report compares the performance in reading of 2,000 nine-year-olds in England and Wales with pupils of the same age in 27 other countries.
England and Wales came 21st, towards the bottom of a middle group of countries whose scores were all close together.They were well behind the top group, which included the United States, Finland, Sweden, Italy and France, but were only just below Germany, Canada and Hungary.
In common with other international studies, the report found scores in England and Wales were pulled down by a "long tail" of under-achieving pupils. The report says part of the explanation may be a larger number of children with learning difficulties in mainstream schools than many other countries and the fact that we do not make the slowest pupils repeat a year.
However, the researchers argue that these two factors alone do not account for the high proportion of under-achievers. One of the most worrying findings is that pupils tested in 1995 and 1996 made slower progress between the ages of eight and nine than they did in 1987.
That may be because schools are putting more energy into getting seven- year-olds through national tests and neglecting other pupils, the report suggests.
In addition, pupils start to study a broader range of subjects between the ages of seven and 11, so that schools may spend less time reading.
It concludes: "Since all surveys of methods of teaching reading show the great majority of teachers use a mixture of methods, it is extremely unlikely that pedagogical differences could influence all results.
"The stability of the long tail over time and across curriculum areas would seem to betoken a stubborn underlying tendency, namely that the British educational system pays too little attention to low performers and could and should pay them more."
Tony Pugh of the Open University's School of Education said: "We didn't do terribly well and we had a bigger spread of achievement than most countries.
"One explanation may be that the national curriculum, with its emphasis on the normal, is not helping low-achievers."Reuse content