Standards based pay vs performance pay

Kim Beazley has been characterised by some in the media as being in favour of performance pay. Which is odd, because last Friday, he said that he opposed it:

Julie Bishop says the Government supports so-called “merit pay”. What do they mean in practice?

Well, it’s hard to say precisely, because this is a Government of spin. No plan and no vision, just a series of short sighted slogans.

But if the Howard Government means anything by so-called “merit pay”, they mean paying teachers more when students get higher marks and principals allocating performance bonuses within schools. The only problem? It just doesn’t work. You only have to look at the overseas experience.

In the US, the system is described by the OECD as a failure. Why? Poor implementation; lack of clear criteria; and a failure to give teachers useful and effective feedback.

And what is the result? Low morale among teaches and a less cooperative and constructive learning climate.

Merit pay is another Howard Government path to Americanisation that we shouldn’t take.

What I want to see is a new set of high national standards for the teaching profession and I want to see teachers paid more when they achieve those standards.

That’s why today I’m announcing the Government I lead will develop an explicit set of national teaching standards. These will be based on the existing national framework which was approved by all the Education ministers – including the then Federal Minister, Brendan Nelson.

Will ‘standards-based pay’ produce results? Maybe, but I’m sceptical. As Sara Mead and I pointed out last year, there’s no evidence (at least in the US) that teachers with a Masters degree produce higher test score gains. In preliminary work with data from Queensland, I’ve found similar results. Higher qualifications don’t seem to be correlated with student scores.

In short, teacher certification is a hoop, while kids’ test scores are an output. It would make more sense to pay for outputs, not hoop-jumping. Surprisingly, many Australian journalists (who are typically paid for writing good stories, not having extra degrees), don’t seem to understand the distinction.

* An aside: It’s not clear which OECD report Beazley has in mind. A Google search for “teacher merit pay” brings up this one, which seems pretty even-handed.

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

13 Responses to Standards based pay vs performance pay

  1. slim says:

    Kid’s test scores are an output, but by no means the only or most important one, other than a convenient one for politicians, bureaucrats and economists because they are the easiest thing to measure. Just what they are measuring, and how valid these measurements are in evaluating the effectiveness of a student, teacher, or school is far from clear.

    Standards-based pay doesn’t refer to someone with a Masters degree. It is a system of bench-marking professional standards supported by ongoing professional development and evaluation processes.

    Teaching is a profession underpinned by cooperation and collaboration. Schools in difficulty will usually be deficient in these qualities. It is a reasonable concern that competitive ‘performance’ bonuses will discourage individual teachers from sharing their trade secrets.

    And how do you even begin to make an equitable performance bonus scale based on kids’ test scores? Education is not a mechanised production line with easily measured and distinct raw ingredients transformed into a uniform can of PAL at the end of the line. Surely it is the aim of all educators to help their students achieve the best they can, and for some students, test scores may not figure so highly.

    Socioeconomic factors, unemployment, drug abuse, emotional neglect/abuse, family stability, non-english speaking backgrounds, how much a family reads and discusses as opposed to passively soaking up television, among others, all have a profound influence on a child’s ability to learn and achieve a ‘desirable’ test score, even among children with the same innate abilities.

    The most wonderfully effective teacher in a disadvantaged school has no chance of achieving the test score bonus payment. So then you have to start creating indices and scales and complex formula in order to approach some kind of effective evaluation of how good a teacher is based on relative and comparative test scores. Another government department in the making, I’m sure. Money better spent by investing in education, not endlessly measuring it.

  2. depressed says:

    This is an incredibly depressing statement by Beazley.

    Yes, there are difficulties with implementing performance-based pay in teaching but what he is saying is that Labor won’t even try to make it work. No trials, no modest experimentation, no attempt at policy innovation. All he is offering is the the most conservative union line on the issue and isn’t even attempting to innovate on the margins. Stale, stale, stale.

  3. Bring Back EP at LP says:

    slim’s pickings is what I have been saying on this topic all along.

  4. Pingback: Invest in education not measurement « Soap Box

  5. cba says:

    Nice opinion piece about performance pay for US teachers in the WSJ (31 Oct 06) by Terry Moe

    a (hopefully fair use) quote:

    “… The idea is not to impose a “one best system,” but simply to take reasonable steps, based on objective data, to see that good teachers are rewarded and mediocre ones aren’t.”

    “If incentive pay for teachers is practical, and if it makes good sense, then why is it so rarely used? The answer has to do with politics and power, not with what is best for children. By far the most powerful forces in the politics of education are the teachers unions, and they are opposed to incentive pay. The unions represent the occupational interests of all their members, not just those who are good teachers, and they have a deep resistance to any form of differentiated treatment that threatens member solidarity. Their demand is consistently for across-the-board-raises. Everyone benefits, no one surpasses anyone else. Their vision — if you can call it that — is one of stultifying sameness.”

    “Fortunately, the unions are surrounded by more incentive-pay brush fires than they can put out. The federal program was one of these, adopted despite union opposition. And the unions have also failed to stop two major incentive-pay programs recently adopted in the states. One is in Texas, which allocated $260 million for a program that started this year in roughly 1,000 low-income schools in high-poverty districts, and will be offered next year to all districts. The other is in Florida, which initiated a $147 million program distributing rewards to the top 25% of teachers in participating districts.”

    “The unions cannot hold back progress forever. Incentive pay is an idea whose time has come. It is an idea that is so unambiguously superior to the status quo — paying good teachers and mediocre teachers the same — that the need for reform is obvious. We can fine-tune the details of how to do it as fairly and effectively as possible. But the direction we need to be moving in is clear.”

  6. Andrew Leigh says:

    Slim, you said “The most wonderfully effective teacher in a disadvantaged school has no chance of achieving the test score bonus payment.”

    No-one is proposing a merit pay system in which we pay teachers bonuses for having smart kids in their class. That would be daft. Ridiculous. Crazy. What some US school districts have done is to pay for test score *gains*. On that metric, the most wonderfully effective teacher in a disadvantaged school is exactly the person who will get the bonus.

  7. Sacha says:

    slim, the idea with measuring things in education is so that you can say something about it. And so, perhaps you can say that kids in NSW appear to have a much better (or worse) understanding of fractions than they did ten years ago. This can be done. Kids in different countries have different strengths say in mathematics, for whatever reasons. By measuring, you can try to see what these variations are.

  8. Shaun says:

    I’m afraid to say I agree with Slim here. Merit based pay is a good idea in theory but using school tests seems to me to be a very narrow-minded way of going about this. For a start, value-add tests are still highly SES correlated. More importantly, what teachers do is much more than just helping kids do better in tests. Using test scores, therefore, only seems to capture one part of a teacher’s capability. Since we know that head teachers are pretty good judges of overall teacher performance, if we were to introduce a performance pay system for teachers, I think it would be better to be done through senior management review.

    However, performance pay itself seems to me just a tiny part of a much wider issue. If we want to raise teacher performance overall, there are two things we could do that might be a lot more useful. The first is to allow head teachers much greater flexibility to remove underperforming teachers. The second is to look at rewards for teachers as a whole. Performance pay is fine, but the real problem here is probably underpay in its entirety and then if you look at incentives for teachers, I’m sure you’ll find that measly performance packages only add very moderately to test scores. A much better idea (aside from raising teacher wages across the board) is to look at other ways of recognising exceptional performance, perhaps through more flexible working conditions, lighter teaching loads, greater avenues for career progression, greater recognition in the school arena in other ways. The whole debate about ‘performance pay’ seems to me to be of largely academic interest, but of very little practical significance.

  9. cba says:

    Shaun: I think your interpretation of performance based pay is too narrow (although I also think some advocates of performance based pay view the issue too narrowly). Isn’t “removing underperforming teachers” a type of performance based pay, just stick instead of carrot? I think the basic idea here is to promote teachers who have the biggest positive impact on their students. A standards based system of pay does not do this – each widget is treated as equal conditional on years of experience and qualifications. It is then little wonder that teachers are attracted to schools based on other factors: being around smart polite kids, locating in more attractive suburbs and having lower student staff ratios. Moreover, the standards based approach is heavily focused on teaching the curriculum, not obtaining an outcome (that being improvement in students). There are many ways to obtain the outcome, so why not allow a diversity of approach?

    “The whole debate about ‘performance pay’ seems to me to be of largely academic interest, but of very little practical significance.”

    Let me reverse it on you then: under the present standards based system, academics have a big role in setting the standards and curriculum and thus indirectly determine pay and performance. Are these questions of academic interest with little practical significance?

    “the real problem here is probably underpay in its entirety”

    I don’t agree with this either. Sure the starting salary is pretty low, but so is the starting salary first year out of most undergraduate degrees. But, by the time you are five years in you make a comfortable wage of about $50,000 p.a. Now that goes a lot further in some areas of the country than others, but $50,000 per annum is not a bad wage for a very secure job. Moreover, teachers who want more have opportunities to take on additional work during holidays (even after school) or can move into school adminstrative positions (albeit that they’re teaching job is already pretty darn taxing – but plenty I know do it). To me the real problem is the 20 year career teacher sitting on his hands earning $65,000 a year (plus generous pension if he got in at the right time), with enthusiastic junior colleagues working twice as hard producing better results for significantly less.

  10. wpd says:

    “No-one is proposing a merit pay system in which we pay teachers bonuses for having smart kids in their class. That would be daft.”

    Spot on.

    “What some US school districts have done is to pay for test score *gains*.”

    Are these gains measured at the ‘individual’; ‘class’; or ‘school’ level?

    Each approach has its own difficulties.

    Merit pay is great in theory but anyone with a modicum of sociological insight and has some appreciation of what ‘education’ is about would understand that it would be a failure if not a disaster.

    Doesn’t anyone remember the Advanced Skills Teacher experiment. What a mess. Minister Bishop ought to study a bit of History.

  11. Shaun says:

    Cba, just to clarify, I’m not arguing for a standard pay system. I’m very interested in making the education system more flexible esp. dealing with underperformance.

    What I am worried about is this narrow academic focus on merit based pay using student test scores. A lot of the academic articles I see on this subject are solely based on giving more money to teachers if they improve test scores. I think a lot of these schemes have come about because the effects are quantifiable, people can measure them and economists like using monetary incentives. Pragmatically though, I think the measured outcome (test score) is far too narrow and the purely monetary incentives are not sufficient. You do see some small effects, but on the downside you then have perverse incentives as well as negative effects on staff morale. Btw, a very similar scheme has been tried in the UK, with no effect whatsoever.

  12. Andrew Leigh says:

    Shaun, does the UK scheme reward student value-added? I thought it was akin to Beazley’s standards-based pay, though I might be wrong. I was also under the impression that there had not yet been any serious evaluations of the program.

  13. Shaun says:


    Link to evaluation attached. I’m afraid to say on re-acquainting myself with the scheme, I’ve got it completely wrong. The scheme does operate similar to the Beazley system, and ironically, it shows some improvements in value-add test scores (probably because of its largely once-off nature). Does that mean we should all re-evaluate our opinions? Unfortunately, the scheme itself has largely been discredited in the UK due to the incredibly high take-up rate.

    Click to access wp113.pdf

Comments are closed.