Secondary schools, with their experience of national league tables, might prove less compliant. The National Association of Head Teachers has already advised its members against agreeing their targets with their education authorities - even though the latter are required to set out goals in their new education development plans.

There is also widespread concern about target-setting as a means of securing improvement. Do they raise standards or provide a basis for meaningful comparison between schools? How can we tell if schools are really improving without knowing what sort of pupils they start with and what value they are adding?

National targets certainly have their critics. Crude goals, they say, distort the work of a school and lead it to concentrate on some pupils rather than others and on some skills rather than others. In particular it is thought that an emphasis on grades A*-C has led them to neglect the weakest pupils. In other words, league tables and public targets have harmed rather than helped the 48,000 excluded from success.

“Weighing the pig doesn’t fatten it. The danger is that it can make things worse by focusing people on a narrow range of features we can test in three-hour exams - and that is not a good basis for a flexible workforce, ” says Dr Dylan Wiliam, head of the school of education at King’s College, London.

His research work, with colleague Professor Paul Black, has focused on what is actually going on in the classroom - in other words, discovering what teaching and assessment techniques lead to improved pupil performance. This, he believes, is the key to improving standards.

So does Professor Harvey Goldstein, of London’s Institute of Education, who raises the difficulties of maintaining GCSE exam standards over time in order to make valid comparisons. It is impossible to make judgments about whether schools are doing better without putting their raw results in context - which means value-added analysis.

“If you are setting targets, even if they are nominally achieved, you won’t know if it is because schools are doing better or the exams have got easier, ” he said.

“What you can do is collect evidence about what it is that does raise achievement. Is it certain kinds of teaching? That is the research we should be doing, not setting national targets.

“That is a sideshow, and a dangerous one, because if you concentrate on that and how a target is reached - if it’s reached - you could be ignoring a lot more important things simply for the sense of achieving the target.

“There is a contradiction. On the one hand, the Government is saying we must have value-added judgments (in performance tables) because this is the only way we can begin to make comparisons. On the other, it is ignoring those judgments when it sets targets because targets are not value added.”

John Sutton, general secretary of the Secondary Heads Association, believes there is no evidence that national targets raise standards - although school-based ones might.

“If a school sets itself smart targets - achievable, realistic - that is probably a good management strategy which may well help,” he said.

“But giving a school a target that is of no relevance is hardly going to be a helpful step to take. A lot comes down to how targets are set and by whom. ”

Schools will run out of arrows to shoot at all the targets, according to Dr John Dunford, soon to take over at SHA: “If results have gone down from one year to the next, or not gone up as much as the national average, then you really can’t make any sensible comment on the performance of that school unless you compare those results with the intake information from five years earlier. ”

But chief education officers seem to sympathise with the Government’s need to direct progress on secondary-school improvement via target-setting. Christine Whatford, vice-chairwoman of the Society of Education Officers, would prefer a bottom-up to top-down approach, but is more concerned that the targets are set at a realistic, achievable level.

The previous government’s National Education and Training Target of 85 per cent of 19-year-olds achieving five A* to C GCSE passes, or the vocational equivalent, seems like pie in the sky even with two years to go to the millennium.

“If I was the government, I would say that top-down targeting is the way to be sure there is going to be an overall raising of standards,” said Ms Whatford.

“It’s not the best model in terms of motivating schools. The best way is in partnership with schools. But the Government needs to be sure all the little things going on locally add up globally.”

Graham Lane, chairman of the Local Government Association’s education committee, believes a realistic goal for 16-year-olds is 50 per cent achieving five good grades within four years - compared to last year’s average in England of 45.1 per cent.

“Target-setting is useful if you set targets that are achievable but challenging,” he said.

“To improve more than two or three percentage points is doing well and achieving more than the national average. I think the Government would be unwise to go above 50 per cent. If you set something that is achievable, you might reach the targets sooner.”

There is, of course, the vague promise from Mr Byers of “funding and guidance” for individual schools to help them reach their targets.

Any extra cash will be welcomed by secondary schools which, despite years of declining real funding per pupil, have none the less presided over an upward trend in GCSE performance.

Value-added warning, page 26