Here we go again. Handing headteachers the responsibility for deciding teachers’ pay increments according to performance sounds plausible – indeed, who could be against it? We need good teachers, don’t we? So what better way of attracting and encouraging stars than rewarding them accordingly?

Actually, if ministers consulted their history, they’d find that they are by no means the first to whom this bright idea has occurred. Victorian Gradgrinds thought it was plausible enough to try it more than 100 years ago. But, alas, as they also discovered, it’s in the large category of things (like George Osborne’s shares for rights) that only makes sense to those who don’t know the first thing about management.

In their essential book on evidence-based management (or more accurately the lack of it) Hard Facts, Dangerous Half Truths and Total Nonsense, Stanford Professors Jeff Pfeffer and Bob Sutton demonstrate that ideas that seem like common sense in management generally aren’t. They devote a number of pages to showing why merit pay for teachers is a classic example.

It’s not just that done according to the letter almost all appraisal-based performance pay systems quickly crumble. Either they create such acute problems of perceived fairness (the aptly named teacher’s pet syndrome) that they demotivate as many people as they encourage. Or managers compensate for the waves of bad feeling created by giving everyone satisfactory appraisals. If merit pay for teachers worked, it would already be in place. In fact, it is always more trouble than it is worth and quietly buried a few years later.

But leaving aside the practical difficulties that affect all such schemes, consider the conditions that would need to be fulfilled for performance incentives to be effective in the specific case of teaching. Are teachers’ effort and motivation (since intelligence is unfortunately insensitive to incentives) the single or most important causal factor in student learning and achievement? Are teachers motivated by money? Are tests and exam results a reliable guide to a student’s educational achievement? Is teaching a solo activity requiring little cooperation with others in the school?

A nanosecond’s reflection indicates that the answer in each case is ‘no’. Much the most important factor in student achievement – more than anything that happens at school – is the home environment. The logic of incentivising teachers rather than parents or indeed students themselves is therefore unclear. On the whole people motivated by money are more likely to go into the City or highly-paid private-sector occupations than teaching. Whether we would want teachers to be motivated by money is anyway moot: giving people incentives to influence something that is not directly under their control (students’ exam results) is an invitation to do so by other means, including cheating. Sure enough, research cited by Pfeffer and Sutton shows that in these conditions cheating is a predctable outcome, and one ‘quite sensitive’ to the size of the incentives on offer. Exam results are a reliable guide only to the ability to pass exams. Right on cue, the CBI recently complained that current emphasis on exam results was not supplying employers with the rounded individuals capable of joined-up thinking that they need. And of course school culture, quality of resources and facilities, and peer support and learning from colleagues all play an important part in teachers’ performance. Since the strings being pulled aren’t attached to any levers, it’s not surprising that trying to relate teachers’ pay to performance is ineffective. ‘Merit-pay plans seldom last longer than five years and […] merit pay consistently fails to improve student performance,’ conclude the authors.

Ironically, it’s a safe bet that those behind today’s scheme who automatically assume that teachers are motivated by money would forcefully deny that they are so influenced themselves. This persistent bias explains why even people who ought to have a clue about management consistently overestimate both the effects of money and the extent that it can be used to compensate for management shortcomings in other areas. In fact, money is what is HR folk call a ‘hygiene factor’: while not enough of it is a powerful demotivator, the reverse is not the case. Beyond a certain point it has little positive motivational effect. The bottom line: the bad news is that the effects of bad management can’t be undone by applying a dusting of cash. The good news is that the best thing to do with pay is to uncomplicate it: pay people enough that it isn’t an issue, so that they (and you) can stop obsessing about it and concentrate on the job instead.

Strikingly, the incentives in the government’s Work Programme, where private-sector providers are rewarded for for getting the unemployed back into work, are based on similarly dubious assumptions. Since it’s not in the job agencies’ gift to create new jobs from thin air, it’s not obvious why incentivising job agencies would be more successful than incentivising, say, the Chancellor to create the economic growth of which jobs depend. Expect, then, the scheme not to work very well, or if it does seem to be working, for there to be evidence of fiddling the books. It doesn’t, there is, and the case rests. Incentivising people to do things that are outside their control falls squarely in the third of Pfeffer and Sutton’s title categories. As the old saw has it, never try to teach a pig to sing. It wastes your time and annoys the pig.