Should Teachers Be Paid for Performance?

Paying teachers for performance should be a great idea, right?  In the corporate world, paying based on merit seems a great way to incentivize someone to do a better job.  For some reason, however, most teachers are against so-called “merit pay.”  I have to say, I was very surprised by this reaction.  My thoughts were similar to many critics, “Well, if you are doing a great job, you should have no worries and should welcome the recognition of your work.  So, it seems like the worst teachers would be the biggest critics, not being able to perform or having to work harder.”  I started to realize that a lot of teachers were quite against merit pay.  Not only that, but unions, who traditionally fight hard for increases in teacher pay, were also against it.  In this post, I hope to identify the arguments for and against merit pay to better understand whether it is a way to improve student achievement and teacher retention, among other things.

The Arguments for Merit Pay

Teachers are often severely underpaid and underappreciated, and the position is not attractive to people with large student debt.  There are some very compelling reasons why merit pay could strengthen public education:

1. After Michelle Rhee implemented bonuses for teachers being rated “highly effective” for two years or more, one such highly effective teacher who was considering leaving the District to teach elsewhere decided to stay. 
2. Paying good teachers more should make the profession more attractive to more highly educated individuals.
3. The private sector has, for years, used merit pay to improve efficiency and allow people to advance for performance rather than simply for working a long time or taking shallow credentialing.
4. Merit pay can be adjusted to reflect the goals we want to elicit out of the system: you can design the weight that standardized tests have in determining performance; what kinds of training or activities are deemed best practices, etc.
5. The way teachers are currently rewarded (premiums for having an advanced degree) has not been shown to correlate with gains in student achievement.
6. Being paid for performance has the potential to encourage teachers to be more creative in how they teach, trying to improve outcomes to earn these incentives.

The Arguments against Merit Pay

While rewarding and retaining quality teachers seem like pretty strong arguments, I have to give pause to some of the incredibly strong arguments against merit pay:

1. What if merit pay doesn’t work? According to a Harvard study of a New York City merit pay program, merit pay did not affect teacher behaviors, student achievement (in the areas measured: English and Math tests) actually declined, and made only negligible improvements in other measures like student attendance, behavioral problems, graduation exam scores, and graduation rates.  Fryer, the economist who conducted the study and was initially supportive of merit pay, noted that the literature was mixed on the effects in other countries, but that NYC’s $75 million investment was not worth it.  Research on performance pay continues to show that it does not work, not even for CEOs in the private sector.
2. How do you measure merit? In a letter to the editor of St. Louis Today, it was noted that measuring merit is usually largely determined by test scores.  As I noted in my post on standardized testing, this is not a measure of real learning or of good teaching, but of shallow learning and uninspired teaching.  The more test scores are used to determine pay, the more we are incentivizing mediocre instruction.  
3. Is merit consistent across all teachers? The St. Louis Today letter also noted that just because a dermatologist performs better than an oncologist in terms of number of deaths, you would still not want the dermatologist to treat every illness.  Likewise, a kindergarten and Junior Math teacher need two very different skill sets in terms of subject matter expertise and classroom management skills. 
4. Are the student’s entering competency and personal challenges accounted for? Just as it is more likely for someone treating cancer to have more patients die than a podiatrist, it is more likely for a teacher in a poor district to have lower test scores because students enter with lower test scores.  It is unreasonable to expect a teacher in a rich, suburban district to end up at the same place as someone from a poor, urban or rural district.
5. Are teachers (good or bad) offered the assistance needed to improve? The Washington teachers’ union criticized merit pay (the IMPACT program) in DC because it is used more as a stick (to fire bad teachers or simply offer a cash bonus to teachers who do well) rather than to provide tools to facilitate a teacher’s improvement.
6. Do teachers want merit pay? According to a survey of American Federation of Teachers members, there is limited support for merit pay based on improvements in portfolios of students’ work, but virtually no support for merit pay based on test scores.  Teachers were far more inclined to think that experience or taking on additional responsibilities are better reasons to extend a bonus to a teacher.
7. Is it redundant? Studies show that more experience tends to make teachers more effective in the first five years of teaching, so wouldn’t the natural increase in teacher’s salary for tenure that is in place already reward improved performance? Further, most teachers I know aren’t really in it for the money, so there is the question of whether it would elicit more performance or dedication from someone who took such a grueling job that pays so little in the first place.
8. Is bad teaching is the root of all problems in the American education system? As I noted in my post on some rather bombastic solutions for how poor kids can help themselves succeed, there are many problems that go beyond the ability of a good teacher.
9. Does merit pay serve as a self-fulfilling prophecy?  Psychology shows us that people respond more to positive feedback than to extrinsic rewards, and this article in the Washington Post raises good concerns about whether merit pay undermines the intrinsic value teachers have for their craft and if the pressures of constant focus on testing improving might make a good teacher perform worse.

Different Ways of Structuring Merit Pay

To be honest, I was surprised in writing this post to have my opinion changed so fully.  Merit pay seems like a brilliant way to infuse the system with new creativity, to get rid of those bad apples, and to reward actual work.  That said, the most salient point for me was that it simply hasn’t worked all that much, even in the private sector.

That said, I think teachers who do have a proven track record do deserve recognition.  I would urge that some of the following points be considered in structuring merit pay programs:

1.     Merit pay should be designed on improvement during the time the teacher is with the student. It is unfair to say “your students still read below grade level” if they entered two or three grade-levels below average in the first place. 

2.     As a corollary, teachers should be rewarded for taking on the most challenging classrooms, as that is where gains really need to be made.  If you just base merit pay on testing, teachers are incentivized to work in the classrooms where kids already start at or above level, which effectively entrenches or worsens the problem of teaching quality in the most struggle student populations.

3.     Merit pay should compare apples to apples.  Comparing an art teacher to a math teacher or a 1^st and 11^th grade teacher is not fair, as the positions require different skills to be successful.

4.     Merit pay should not be informed greatly by standardized tests.  We already have seen the deleterious effects of tests in my last post, and to reward teachers for producing bubble-filling automatons that have not learned any significant skills is an awful idea.

5.     Merit pay should be linked to programs to help teachers improve and work together.  An interesting model was one that rewarded a whole school’s teachers for improved performance, thus creating a sense of teamwork to improve all kids’ education.  Nobody thinks bad teachers should be in the classroom, but if the system offers only back-handed financial rewards and creates competition, it is not creating a healthy environment for students.  Further, students spending 6-8 years in a school will need all their teachers, not some of them, to be good, so creating whole schools with great teaching is key.

6.     Merit pay can include experience for the first five years to incentivize teaching retention, but this factor should lose weight after 5 years, as most studies agree that there are no benefits to experience after that point.  

7.     Merit pay should not be viewed as a panacea, as teachers—particularly in the poorest environments—are not going to be able to overcome the gaps in basic needs like proper nourishment and parental support that many kids bring to the classroom.

Speak out!

What are your thoughts on merit pay? Do you have an experience with merit pay (in teaching or elsewhere) you can share? In your opinion, what are the limits of merit pay?

What ideas do you have for how to structure merit pay in an effective manner for teachers so that quality teaching is rewarded and students are given access to the best quality teachers possible?