Since the work of Alfie Kohn and W E Deming dismissed pay linked to performance as a motivator, no academic that I am aware of has endorsed individual performance pay for public sector professions, particularly teachers. There is no evidence - again, of which I am aware - that performance-related pay for teachers is effective; there are only ideological assumptions.
As you say, sufficient pay is central to recruitment, retention and motivation. While setting pay above the level of sufficiency may not make a difference, the reverse is not the case. If pay is set too low, graduates simply take their expertise elsewhere. There is plenty of evidence from previous periods of low pay to support this contention.
Your analysis took account of the pay freeze and pension hikes but did not mention the effect of the trebling of student fees. Qualified teachers would start paying back their loans from day one, and this will have a sharp effect on recruitment.
The elephant in the room when talking about delegating more pay powers to heads is how you would prevent abuse of the system. More defined criteria were introduced for incentive allowances in the 1980s precisely because additional pay had been distributed according to whether the head liked you; whether you were quiescent; whether your face fitted; and whether you were a union activist.
There is no desire among any of the teacher unions to return to those times.
Martin Freedman, Head of pay, conditions and pensions, ATL.