Should bad teachers be paid less?

Here's a tricky question for those of us in favour of moving to more localised pay in the public sector. Would you prioritise that over other reforms to pay and conditions in the public sector?

I was reminded of this by recent reports that "Bad teachers could be paid less under new Government proposals aimed at driving up standards in state education. The Education Select Committee insists a new results-based system is needed so poor teachers are not paid the same as colleagues doing a better job." [Or to take the more extreme version, 'forget paying less, we should sack them'.]

There are, of course, two simple answers to this question: we should do both, we should do neither. However, if you are convinced by the need for both types of reforms, and you think that the government needs to pick it's battles, the question of priorities is a far trickier one.

Evidence on local pay suggests that the quality of service is higher in areas where real wages are higher. This research suggests this effect comes about because higher pay improves the quality of staff (in the NHS, specifically, through less use of agency staff). It's reasonable to think that the same mechanism would work if you more closely linked individual pay to quality (because the evidence on local pay suggests quality staff are attracted by higher pay).

That said, individual pay may be more difficult to implement because it requires assessing the quality of individual teachers. In contrast, local pay is 'easy' to implement in the sense that, e.g., difficulties in filling vacancies provide a good signal of areas where pay may be too low. Local pay is also conceivably less divisive because perceptions of unfairness rely on comparisons of pay with workers based elsewhere. In contrast, with individual pay you may only need to look at the classroom next door to see someone getting paid more than you (albeit for 'doing a better job' if assessments are accurate).

Overall, I don't have a good feeling for which would have the larger impact on public service quality (and suspect that we don't know based on the available evidence). On balance, I think I lean towards local pay first: 1) The difficulty in filling vacancies provides a clear(ish) signal of areas where pay needs to rise; contrast that with the difficulties of assessing individual teachers. 2) The evidence suggests that there is a negative impact on public sector performance from low public sector pay in high cost regions. 3) Local differences in 'real wages' are hard to justify on equity grounds and addressing this less likely to be divisive for a given school, hospital etc.