The question of whether or not initial teacher-training targets for Wales need adjusting is not a new one.

Back in 2001, the Welsh primary teacher workforce represented 6.5 per cent of the England and Wales total - but Wales had 8.7 per cent of the primary intake target for ITT.

At present, the Department for Education and Skills in London is mainly responsible for the underlying data analysis used in generating the ITT targets for Wales.

The DfES uses a number of models, and it assumes the data used to generate targets is uniform across England and Wales. However, this is not always the case.

For example, in the primary model, the split between undergraduate and postgraduate courses is assumed to be 57:43. However, in Wales, the split is almost exactly the other way round. This means there is a greater time lag between entry to the course and the arrival of new teachers in schools, because undergraduates are normally on a three-year degree course whereas training for the postgraduate certificate in education lasts for a year.

Secondly, pupil numbers are expected to decline significantly in England and Wales. However, the decline is not uniform across the age groups or across Wales.

For example, it is assumed that south-east Wales will have 4,800 more nought to four-year-olds by 2013, but 27,300 (13 per cent) fewer five to 15-year-olds. North Wales will have 2,100 fewer nought to four children and 8,800 (9.5 per cent) fewer in secondary schools.

Other factors affecting the teacher-supply model include possible changes to pension ages, the transfer of funding arrangements for induction to schools, and the short-term rise in the number of teachers reaching retirement age.

As many of the retirements will be among senior staff, direct competition for their posts is unlikely to come from newly-qualified teachers. Indeed, some of the posts may be filled by teachers returning to Wales from England.

Not everyone who completes ITT enters service immediately. A build-up of trained teachers willing to enter service, but not currently working as teachers, will intensify competition for any posts. In Wales, this may well be the current position in the primary sector.

Premature retirement or re-entry of older teachers can also contribute to turnover within the teaching profession. And turnover in the primary sector in Wales has traditionally been lower than for any region in England.

Finally, since the introduction of tuition fees, students have been opting for undergraduate courses closer to home. This trend is likely to intensify when top-up fees are introduced in England in 2006.

The extent to which the England and Wales ITT model remains useful for Wales also depends in part upon how far policy in the two countries diverges. Key policy questions that need to be considered before any outcome can be reached include:

* will students in Wales and England have different expectations regarding teacher-training in the new fees climate?

* what level of over-supply is acceptable?

* is there any greater responsibility to PGCE students when it comes to finding a teaching post or should they compete with those on undergraduate courses?

* what are the needs of Welsh-language schools and how can they be protected if training targets are reduced?

* are there any implications from the recent rise in the number of trainees who live in Wales?

After bringing together all the data, the issue remains regarding the likely implications for ITT in Wales and how the targets should be decided.

At this stage, only the position for the primary sector has been discussed in any detail. However, it is clear that the decline in pupil numbers will have consequences for the secondary sector in the future that cannot be ignored.

Should Wales follow the example of Scotland and Northern Ireland and develop its own teacher-supply model, or continue as part of the model developed in London? No model can ever be completely accurate, and it is better to err on the side of over-supply than to risk a shortage of teachers.

However, there is ample evidence that, unless there are policy changes, some difficult decisions will need to be made about the targets for both primary and secondary ITT places in Wales during the next decade.

To avoid taking these decisions may result in both unnecessary spending and a significant cadre of trained but unemployed teachers.

John Howson is a member of the Furlong group reviewing teacher training in Wales. This is an edited version of a paper on teacher supply in Wales published by the Assembly last week