Will Rahm Emanuel's Merit-pay System Work Where Others Haven't?

Last week, Chicago mayor Rahm Emanuel announced that he’s rolling out a merit pay program specifically for school principals, using $5 million in donated funds. The plan is particularly bold considering its announcement comes on the heels of quite a bit of evidence, from research to scandals, showing the faults of merit pay.

In March, we wrote about Harvard economist Roland Fryer‘s study on  New York City’s failed merit pay experiment, the Schoolwide Performance Bonus Program, which was shutdown last month. A subsequent RAND report echoes much of Fryer’s findings:

…the theory underlying school-based pay-for-performance programs may be flawed. Motivation alone might not be sufficient. Even if the bonus here had inspired teachers to improve, they might have lacked the capacity or resources — such as school leadership, expertise, instructional materials, or time — to bring about improvement.

Vanderbilt released a study last year critical of Nashville’s failed flirtations with merit pay, where, unlike New York, bonus money was given directly to individual teachers rather than institutions as a whole. The experiment in Nashville’s public school system has offered mixed and generally negative results.

Texas’s merit pay program, a combination of school and individual incentives, has had lackluster results. It’s recently been hollowed out and will likely face even more cuts in the next few weeks.

And then of course there’s the scandal that erupted in Atlanta this summer, where the emphasis on bonus pay for higher test scores resulted in widespread cheating and corruption.

So what makes Emanuel think his program will work any differently in Chicago? For starters, its carrots are directed at the top of the management chain, toward principals rather than teachers or other staff. Bonuses are expected to range from $5,000 and $10,000 per principal.

From the Chicago Sun Times:

…Emanuel said his principal merit pay program will be unique in that it will include training principals to a set of expectations outlined in a new “principal performance contract’’ that is still being drafted.

“One of the things that we’ve learned from the failure of others is you just can’t kind of throw performance pay out there [as] a Hail-Mary pass,’’ the mayor said at a news conference at Melody School in West Garfield Park.

His program will be a comprehensive, integrated approach — “the first of its kind anywhere in the country” — that doesn’t just rely on “‘You’ll get a bonus.’ It relies on objective standards, measured, re-training and training, and the performance is tied to exceeding those objective goals,’’ Emanuel said. A new Chicago Leadership Collaborative will oversee the project.

So it seems the major stated difference in Emanuel’s plan – besides principals themselves – is explaining  a contract that hasn’t been written. Beyond that, watchwords like “comprehensive,” “integrated,” and “objective standards,”  seem sadly familiar to what we’ve heard before.

But who knows, maybe Emanuel will pull it off and the program will be a success. Among other things, that will depend on Chicago’s school system proving to be less corrupt than Atlanta’s. So, we’re not exactly holding our breath.


Quick correction - the last line says "So, we’re not holding exactly holding our breathe." Probably best to get rid of that last e. And as a Chicago resident, I too am not holding my breath... everything is corrupt!

Matthew Philips

fixed thank you

Mike B

Merit pay systems break down when there the ability to earn the bonus is actually outside of one's ability. While teachers can clearly do poorly, to have one's class perform in an upper quartile is basically determined by the luck of the draw on day one rather than any extra hours put in by a fully qualified and competent teacher. Kids are either smart and motivated or they aren't, they have a supporting home environment or they don't. Too many dysfunctional children in a class and not only will it just be harder to teach them, but giving them the extra attention they need may reduce the attentions needed by other students. There are only so many hours in a day and teachers have 30 kids in their class. As one person they simply lack the tools to provide intensive care to a large amount of students. Therefore they just do the best they can and if that means a bonus then all the better.



Therefore teachers performance should be measured by the increase in performance (ideally of each individual student).

I agree that teaching is not an easy job. (one of the points why I would not consider it)


First of all, devoting more money to education is never a bad thing.

Secondly, why should you want to pay the principals more rather than the students achieving the higher standard. Can't there be a way of directly supporting a student (or group of student) achieving an above-average increase (so it is less biased for 'smart' and 'less-smart' students) in (test)performance.

Third, several bonus systems in the past have shown to be vulnerable for corruption. Why not try to increase the overall pay for teachers so that teaching becomes a more attractive job. In the Netherlands we had(/have) a general lack of (good) teachers mainly due to the low income they receive. I think the best solution then offered was to get part-timers, unemployed and retired people to fill this gap, experienced people who had an incentive (financial or social) to convey their experience and knowledge to the youth. Of course this was immediately shut down by schools due to the difficulty of having those people teach according to the curriculum.

I also don't get why the money would go to the principal rather than the school('s funds). Especially since the argument is made not to give it to the teachers because that would give them the incentive to corrupt, wouldn't the same go for principals? (who would otherwise have lacked in checking their teachers for not-being corrupt)

I think the effort and money put in to the projects such as this merit pay for principals idea could better be used to try and make teaching a more attractive job opportunity rather than giving the principals incentives to skew results, tests, or pressure teachers to obtain (or 'make-up') results so he can get his bonus.

If, however, they do make a plan somewhat based what they currently state, they should make sure the principal in no way has any control over the standard and the data used for determining whether he gets the bonus. Secondly, there should be clear consequences for any form of corruption.

To be honest I have no high hopes for such a project, who is going to check who? The mayor want to show his version works so he has no incentive, the schools/teachers can (in the case the principal 'cheated') chose to either accept the 'you performed well rating' or publicly announce their principal cheated and they actually didn't, so they have no real incentives either.

I do understand that there also are people actually trying to make education/their school better and not everyone is corrupt, I just think this isn't the right system to benefit those people.



"First of all, devoting more money to education is never a bad thing."

Is it never a bad thing? Has education performance increased or decreased over the past 50 years as we've thrown increasingly greater amounts of money per student into the system?

Don't parochial schools, with their lower costs per student, continue to out-perform public schools?

Maybe it's more accurate to say that increasing focus on education is never a bad thing. I'm just not sure more money is the answer.


I work in this field.

Anyone who worked in corporate Amercia though the 80s heard all the same refrains about performance measurement -- flawed formulas, outside factors, unfair bosses, cheating, etc.

Tough. It was a hand grenade science, but everyone got out of their chairs, became results-oriented, and if the goals and metrics had to be adjusted over time, that happened. You only have to look at the best companies in the world to see which high performing teams embraced carrot and stick management.

Teacher's unions oppose merit pay because it creates potential fissures in their nice, secure income base (dues payers). Every teacher is supposed to be a robot cog marching along their actuarial time on this mortal coil getting their annual incremental raises. There are no great teachers and there are no bad teachers -- actually, let me correct myself, there are only great teachers. And it is just fine and dandy if the 60 year old K-2 gym teacher spending her day setting dodgeball games up for 8 year olds makes 135k for 180 days of work and walks out the door at 2:35 every day, while the 35 year old with advanced degrees teaching high school special ed physics and staying after class until 6:00 pm every night (my wife) is making 42k and looking for a job in the private sector.

So how can you possibly introduce any sort of measurement into an education envirom -- oh, wait THE ENTIRE INDUSTRY IS BASED ON GRADING. From the time kids step onto the bus for first grade until the time they step onto a podium to get their hat they are measured and graded and quizzed. You can measure the customers but you can't measure the employees? Seriously?

Anyone that has ever worked in a school immediately knows who the crappy, mailing-it-in teachers are, and who the great ones are. And by not addressing the issue of lousy teachers you are really insulting the whole profession -- how important can the job be if lousy teachers are openly condoned?

Don't get me started.


Mike B

Lousy teachers are condoned because even if you fire them you are unlikely to get someone who is better and will in all likelihood, get someone who is just as bad or worse. If you want to deal with the problem of teacher quality then make it a respectable, well paid profession instead of some kind of glorified day care provider.

Prior to the 1970's most women only had two career options, teacher and nurse. Schools could set low wages and still attract top quality talent because the women applying to the positions either had a husband to support them or no family they needed to support and B) because they simply had no other option. Since that time talented women in the workplace can get just about whatever job they want and command salaries equal to their male counterparts (provided they don't take time off for family). However public sector schools don't have the budget or wherewithall to try to compete for this talent therefore they have been forced to retain an ever decreasing caliber of employee through things like job security and pension benefits.

So if you want to fix the problem with poor teachers, either find a way to increase their productivity so that any old $10/hr cog can read a script and say the magic learning words, OR, transform teaching into a well paid profession that seeks out top talent with top wages. The reason that teachers unions are resistant is because they know it's a trap. Towns will never make good on any promise to increase pay therefore any of these proposals are all stick and no carrot. I personally entered into a new pay for performance system and it turned out to be exactly that. We were told that instead of everybody getting the same raise, the top performers would get more and that $ would come from the lowest performers getting nothing. Instead the top performers get what the used to get anyway, the standard performers don't get anything and the low performers get docked. So more work = more money was actually more work = same money. This is what the teachers are rightly wary about.



"First of all, devoting more money to education is never a bad thing."

What an excellent sentiment, and one needs look no further than the California public school system as proof.

Eric M. Jones.

"Even if the bonus here had inspired teachers to improve, they might have lacked the capacity or resources — such as school leadership, expertise, instructional materials, or time — to bring about improvement."

How can one inspire improvement? Money won't do it. I see this as a major flaw in all such plans (like Bill Gates's). Money is and has always been a short-term motivator. I don't think you can "inspire" teachers to be great...or artists to be great...or ballplayers to be great.

That's not how it works.


@eric. I gave your post a thumbs up, but I have to say I disagree in some ways. All sorts of things inspire different people for different ways. For some it might be is recognition or status, for most it is money, though. Our whole economy is based on people being inspired by the opportunity to do better and live better lives for you and your children by your work efforts -- most times this means money.

Personally, I am sick to death of teachers unions running behind emotive defenses everytime the subject of managing their efforts is raised. Anyone who won't blindly throw tax dollars their way hates children or doesn't respect the great profession of teaching.

Many if not most teachers in my state are over-compensated when you factor in their family helathcare plans to the grave and pensions. Take whatever salary you want say it is too low, but in most US states it is for 180 work days, rarely eight hours mandated. I say mandated because many teachers take work home, papers to grade, etc, and put in many more hours, like my wife. Others put a movie on and grade papers at their desk.

So you take a 45 year old teacher with a bachelor degree who is making 60 or 65 large PLUS incredible healthcare PLUS an incredible pension PLUS a 180 day work year PLUS little or no work supervision PLUS tenure.....go replace that in the private sector.

PS: anyone interested in this topic HAS to see "Waiting For Superman".


Eric M. Jones

I can't think of ONE job where a person will become a star by being paid more money.

Can you?

Steve Nations

We need better teachers and principals. I don't think that's controversial. And I don't think there is a single industry in the world in which you couldn't make the claim that said industry needs better workers.

But we also need better students. We should spend some effort talking about the ways our society fails to incentivize students to do better. The drop-out rate in Chicago is outrageous. But most of the blame for that falls to the students and their parents -- not the schools.


This won't work as expounded on in the "Frekonomics" book. In the book there are two examples: Business managers, and Sports Teams. As the book talks about there seems to be no correlation between the person at the top and performance at the bottom. It turns out that, especially near term, performance isn't as easy to affect as we like to believe.

In this particular case, what can the superintendent due, in the near term, to affect near term results? Not much. The staff is what it is, the curriculum is what it is, the students are what they are and the superintendent can't do much about it. The requirements for a teaching certificate are such that all teachers have learned the same methods. The curriculum is set by the state. Public schools must take all students. Apart from all that, any change that is made is going to take years, possibly decades to propogate through the system and make a real impact. By the time that happens the current superintendent is going to be long gone.


Devon Blair

Have there been any attempts to look longer term with improvement? For example the number of students that go on to college and then graduate from college as the bases for compensation. Or are all the merit pay systems related to test scores? Granted that the teacher/ teachers would have to receive the bonuses years after the kids graduate.


@eric. Not sure how you define "star" in terms of this discussion. I suppose some of this is bell-curved -- the greatest and the worst need no incentives. It's the vast majority in the middle where you can move the needle.

Merit pay has to fall under a broad range of education reform -- tenure repeal, school choice through vouchers, etc

As far as it being "societal" or "cultural"....how do you explain that in many areas charter schools and parochial schools across the street from PS Hellhole are graduating 98 percent of their kids and the public schools are graduating less than 30? And, no, it isn't merely because of where the bad kids wind up, because in many voucher programs, it is the worst students whose parents sign up to try something else.

Too many inner city schools are broken, and the cost to society is devastating. Think of it this way: if you had to design a human system for minimal performance, wouldn't you start by eliminating all measurement, incentives, and threat of termination?


Steve S.

I think we already know how Matt Damon feels about this topic: http://www.youtube.com/watch?v=WFHJkvEwyhk


Funny how in all this discussion there is virtually no mention of the people who have the biggest impact on student achievement....PARENTS!

After all, students are usually under parental supervision for a lot longer each day (8-10 hours) than they are under school supervision (about 6-7 hours). If you want to create a merit-based system that actually works, incentivize the parents!

After all, whose fault is it if a child doesn't do their homework, doesn't eat a healthy breakfast, or doesn't show up for class prepared to learn? The principal's???

Travis G

Isn't the whole debate over merit pay kind of over-analyzing things? Can anyone name a private sector job/career where one isn't evaluated constantly based on their individual "merit"? Are there other jobs out there where you aren't evaluated (and compensated) based on the value of your work?


Merit pay systems are always going to be limited in what they can achieve. The core problem is one that economists are well aware of--it is notoriously difficult to measure productivity in the service sector in general. In the education sector, where so many exogenous factors affect student achievement, the problem is even worse. Compounding the problem is the fact that if you try to tie teacher or administrator pay to student performance, and you measure that student performance with the usual tool of standardized tests, you are setting yourself up for a classic case of Campbell's Law: "The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor."