It didn’t work then, but can it work now? Michael Duffy assesses a study of the first year of performance pay

Introducing the first system of performance-related pay for teachers, the vice-president of the Education Board made a famous undertaking. Addressing the House of Commons in 1862, Robert Lowe said he couldn’t promise that it would be economical, and he couldn’t promise it would work, but he insisted it would be one or the other: “If it is not cheap, it shall be efficient; if it’s not efficient, it shall be cheap.”

That system failed. After 30 years of teaching to test, payment by results was abandoned, lingering only as a folk memory of mistaken central intervention. Outside Britain, though, Lowe’s promise was not to be forgotten. Performance Pay for Teachers, the latest survey from Exeter University’s ongoing research into appraisal and performance, shows that the concept has endured for more than a century. School districts in the United States have often tried it; so, more recently, have New Zealand and Australia. One of the oddities of the story is that performance-related pay was abandoned in the Australian state of Victoria in 1999, just as it was being reintroduced here.

As Professor Ted Wragg and his colleagues show, there is extensive research on the subject. Very little of it, though, is wholly positive. Many schemes have failed to produce the expected benefits of harder work and better teaching. Some don’t last long enough; some have unexpected side-effects, such as the neglect of unrewarded tasks or the concealing of shortcomings.

In Britain there are problems with misrecorded data or (a typical Wragg touch) “lying, as it’s called”. There are also significant costs. But, in 1999, teacher recruitment and retention were at crisis point; significant extra pay was needed, and performance pay was the price the Treasury demanded. The threshold assessment system introduced that year and the performance management regime that followed it were centrally directed, but managed by a private contractor; there was, as the Government intended, no role for unions or local education authorities. It was very much a New Labour initiative. But how well did it work?

The research team examined the reactions of heads and teachers to the first full year of the new regime (in practice, rather more than a year, given the time it took for all the assessments and appeals to be finalised), and to the first 18 months of performance management and appraisal. The data was based on a questionnaire sent to a sample of schools across Britain.

With a response rate of more than 50 per cent, well over 1,000 heads were surveyed: in a sub-sample of these schools, teachers were surveyed by questionnaire and interview.

At one level, threshold assessment worked better than heads or teachers had expected. Heads were universally scathing about the private sector training they were given or, as most of them put it, subjected to; “the worst experience of my professional career” was by no means an untypical reaction. They were angry, too, about the way the message changed in the lead-in period, from the initial guidance that only about half of the teachers qualified to apply would succeed, to the later (and, as it turned out, accurate) forecast that the great majority of applicants would “meet the standards”.

They were annoyed at the uncertainty - in the case of progression along the upper scale, still unresolved - about where the money would come from to meet the new salary costs and about the amount of time they had to spend assessing applications. They resented the bureaucracy, but accepted that the system was fair and consistent. Surprisingly, perhaps, the great majority rated the contribution of the external assessor highly.

About 40 per cent of the heads supported threshold assessment in principle; no teacher did. For most of the teachers, performance pay was, and was meant to be, divisive: it ignored, they thought, the collegiality that they valued. They didn’t, however, object to the standards themselves, which were a realistic description of the job they were doing. So failure, for the 3 per cent deemed not to meet the standards, was often devastating.

The Government saw performance management as an essential part of the new pay process, and the final part of the Exeter research investigates the extent to which, in 2001 and 2002, this was happening. Again, heads and teachers were interviewed in depth. Most were positive: properly handled, performance review would, they said, improve performance and raise standards, and in a minority of schools, where objectives were specifically linked to school and professional development plans, teachers believed this would happen.

But some schools were going through the motions, even to the extent that previously agreed initiatives such as structured classroom observation were being squeezed out. The researchers emphasise the reliance on documentation and “evidence”; performance management and pay are, they conclude, data-led.

But do they work? It’s too soon to say. There has probably been an effect on retention, says the team. Many threshold applicants (most of them in the initial round already in their forties) said they applied “to boost my pension”. It’s less clear whether the extra pay has altered teachers’ motivation: there are many other more important factors. There is certainly evidence that schools and teachers have become more adept at handling monitoring and assessment data, but there is little evidence that they have changed their teaching styles and strategies because of it. We simply don’t know whether performance pay by itself is raising standards, whether, in Robert Lowe’s phrase, it is “efficient”. What we do know is that it hasn’t been “cheap”.

It isn’t going to go away, though, so this survey, sharply focused on the realities of schools and teaching and rich with glimpses of good (and occasionally spectacularly bad) practice, is much more than merely interesting reading. It’s comprehensive, balanced and persuasive: a significant contribution to an important, indeed urgent, debate.