Pareto Improving Performance Pay

Unless I’m misreading her, Federal Education Minister Julie Bishop appears to have taken the view that it’s possible to move to merit pay without raising average pay.

Politically, I can’t see how this is workable. All the merit pay schemes I’m aware of (eg. Denver, Cleveland, Florida, Israel, UK) have been what economists call a Pareto improvement – no teachers are paid less, and some are paid more.

Economically, raising average teacher pay probably makes sense too. Chris Ryan and I have been beaten up a bit by teacher unions for our paper on trends in teacher aptitude. But I suspect they wouldn’t be unhappy with Figure 13, which put together a 30-year series on how starting teacher pay compares with the pay of new graduates (in case you think this is all about the denominator, see also Appendix Figures 2-5). Teacher pay has steadily declined relative to new graduates. The same is true if you compare teachers to a different group, such as all professionals, or all workers. I have another paper showing that a 1 percent rise in the salary of a starting teacher boosts the average aptitude of students entering teacher education courses by 0.6 percentile ranks. If merit pay raised average teacher wages, that would be no bad thing.

This entry was posted in Economics of Education. Bookmark the permalink.

12 Responses to Pareto Improving Performance Pay

  1. Bring Back CL's blog says:

    that is certainly what she said.

    She seems to be confusing performance with qualifications,standards etc also.

  2. Leon says:

    I’m not sure if we’re referring to the same interview, but I thought the impression that she gave was that *instead of being paid more as time goes by*, merit-based pay would be introduced, i.e. funds previously dedicated to increasing teacher pay based on experience would be diverted towards pay increases based on results.

    If it were only applied to new graduates in teaching, it wouldn’t necessarily involve lowering anyone’s wages.

  3. Aussie Equitist says:

    I have two children in NSW public schools and I’ll give Mistress Android my feedback on teacher performance right here and now: public school teachers are grossly under-valued, under-paid and under-resourced. Ditto re the children they endeavour to teach!

    I would also implore the increasingly draconian Howard Government to cease and desist with its control-freak mentality – and stop demonising and vilifying public school teachers and the inequitable education system in which they operate.

  4. conrad says:


    I think that having one rule for new graduates and one rule for old teachers is going to be approximately as popular as a nasty disease. Just like similar laws were in France.

  5. Sacha says:

    Andrew, I agree with you that it’s probably not politically possible to reduce some teacher’s salaries.

  6. Kevin Cox says:

    Improving teachers salaries improves “quality”. Why bother with merit pay – just increase all salaries and give more to those who do the things that others do not want to do – like teach in difficult schools or in country areas or in subjects that have difficulty finding recruits.

    Merit pay has many difficulties associated with measurement of merit and potentially unexpected adverse side-effects and does not seem worth the bother.

    Continue to run trials or experiment with one of the State Systems such as Tas or SA.

  7. Pingback: Club Troppo » Missing Link

  8. James Farrell says:

    Whether or not anyone’s absolute pay goes down, it amounts to the same thing. Over time, a large chunk of teachers will be stigmatised as poor performers, and their morale and actual performance will sink accordingly.

    I agree with AE and Kevin’s well expressed statements.

  9. Ken Lovell says:

    I still can’t get anyone to tell me: (a) which kids get the teachers identified as duds, and (2) how does that decision get justified to the parents of those kids?

  10. Andrew Leigh says:

    Ken, at the moment, the answer is (a) kids in the most disadvantaged schools, and (b) it doesn’t. If you wanted to set up a pay system that encouraged the most talented teachers to teach the richest kids (without being too obvious about it), you’d design something that looked like the present system. Let’s not pretend that uniform teacher salary schedules are serving the interests of disadvantaged kids.

    A better system would encourage the best teachers to stay, and help underperformers to find a more suitable career.

  11. Ken Lovell says:

    Sorry, my questions were badly worded. I meant to ask what would happen under a system that categorised teachers in terms of their performance. It’s easy to identify the benefits that would flow from being able to allocate the best teachers to disadvantaged kids but the flip side is that the worst teachers would have to go … where? To the most advantaged kids? It’s no good suggesting they ‘find a more suitable career’, we can’t fill the vacancies that exist now. Even if that could be changed, any performance-based system will inevitably rate some teachers as ‘worse’ than others and nobody will want to be lumbered with the ‘under-performers’.

    That’s what I see as the fatal flaw in any performance-based scheme (or one of them). There is no way any group of parents would accept their kids being allocated second-rate teachers as an act of deliberate policy.

  12. Andrew Leigh says:

    Ken, we probably want to think about allowing more dismissals too (I mentioned in my AFR oped the fact that only 3/40000 Victorian teachers got dismissed last year). As to inflows, there are always area-specific shortages, but teacher labour markets as a whole tend to be in a state of oversupply. And of course, raising mean pay will increase supply too.

Comments are closed.