Losing Experienced Teachers Is Bad for Schools, Right?

(Photo: www.audio-luci-store.it)

Maybe not. A new working paper (abstract; PDF) by Maria Fitzpatrick and Michael Lovenheim finds that offering early retirement to experienced schoolteachers doesn’t have a negative effect on students’ test scores, and in some cases leads to an improvement. The abstract:

Early retirement incentives (ERIs) are increasingly prevalent in education as districts seek to close budget gaps by replacing expensive experienced teachers with lower-cost newer teachers. Combined with the aging of the teacher workforce, these ERIs are likely to change the composition of teachers dramatically in the coming years.  We use exogenous variation from an ERI program in Illinois in the mid-1990s to provide the first evidence in the literature of the effects of large-scale teacher retirements on student achievement.  We find the program did not reduce test scores; likely, it increased them, with positive effects most pronounced in lower-SES schools.

Here are some of the factors that Fitzpatrick and Lovenheim had to wrestle with:

Ex ante, it is unclear what the effects of large-scale teacher retirements, such as those resulting from an ERI, will be. On the one hand, retiring teachers are highly experienced, and they typically are replaced with much less-experienced teachers or with new teachers. The evidence of the strong relationship between experience and effectiveness in the classroom (Wiswall 2013; Rivkin, Hanushek and Kain 2005; Rockoff 2004) suggests teacher retirements could reduce student achievement. Even among teachers who have the same amount of experience, teacher quality varies substantially (Goldhaber and Hansen 2010). If teachers with better job opportunities outside of Illinois Public Schools (IPS) are the most likely to retire, and if wages outside teaching are positively correlated with teacher quality (Chingos and West 2012), then the offer of an ERI would negatively affect student test scores.

On the other hand, teachers who are near retirement may put forth less effort than younger teachers or may be less well-trained in modern, potentially more effective, pedagogical practices. This may be particularly true for those teachers who desire to retire early. Alternatively, if productivity is negatively correlated with disutility from teaching, the teachers who choose to take up the ERI may be those that are least productive. Family and personal circumstances also influence the labor-leisure decision in ways that lead to ambiguous predictions of the effect of ERIs on achievement. Finally, principals and administrators may respond to large losses of experienced teachers, e.g. by decreasing class sizes, changing the assignment of teachers to students or purchasing additional non-teacher resources.

I cannot imagine many teachers’ unions embracing this finding. But it may be good news for school districts.

One story to think about here is that 1) until the 1960s, schoolteachers represented the best and brightest U.S. women, who didn’t have many job options available; 2) as many bright and well-educated women went elsewhere, the overall quality of teaching fell; but 3) there may be a renaissance in the appeal of teaching among bright women (and men), perhaps nudged along by an economic upheaval in which more traditional high-status jobs (in finance, law, medicine, etc.) have become, for a variety of reasons, less appealing.

Kevin Johnson

As a teacher in between 'new' and 'experienced'--Im entering my 9th year teaching at age 31--and what I worry most about this study's use of test scores as the only measure of what is best for students. Most of those tests are multiple choice tests and as I have gone through the system, more and more focus is being placed on test-taking strategies, cram sessions before tests, and even teaching more towards the lower-performing students. Most of the experienced teachers I have seen retire refuse to adopt these practices, but will do whatever it takes to help students learn. Newer teachers are more worried about their jobs as they are under provisional or non-tenure contracts and see test scores as their only measure of performance.

I have been pressured to remove writing assignments from my curriculum since the only measurement of my social studies class is a 70 question, end-of-the-year multiple choice test. Administrators have asked why I place details on current events when they are not in the curriculum, why I include more information than the students need to know, and why do I focus so much on computer technology if it does not improve test scores.

I know most of this seems anecdotal and based on one teacher's observances, but I would like to see more from this about AP test scores, IB accreditation, etc, as I think those are more rigorous and a better (though still flawed) measure of student learning and success as many state tests are easier, and not as adept at testing learning and quality education.



Ideally, you should always be "teaching the test", as it is the measure of student knowledge. If the test doesn't adequately assess that, the problem is with the test. That appears to be the problem here, but too often (nearly always actually) the response from teachers is that they don't want to teach the test, rather than (as you seem to believe) that students' knowledge is not adequately measured using current methods. Keep up the fight, and stay in the business; we need more of you.

Other items re Steven's remarks:

1. Public sector unions shouldn't be permitted. Employees whose employer can legally require customers to buy its product, at whatever price they demand whether the customer wants it or not, have no incentive to moderate demands in contract negotiations.

2. Schools (and all other government entities that do so) need to stop automatically paying employees more simply because they have a postgrad degree, and stop making that degree a requirement for promotion (implicit or explicit).


Enter your name...

Schools in particular should not pay more for teachers getting graduate degrees in education, because those degrees have been demonstrated to have no positive effect on student achievement.

Isidro Fernández

From my own point of view, experience is not make the same thing along many years, but is do many different things.

Here is Spain education is public, and managed by the government (local, for our CCAA, similar to states). So teachers are public servants, they pass a test once and then according to their results get a destiny or they must wait till a vacancy appears.

So we can find several cases, for example a teacher that started to work in 1985, 30 years ago in a small town, and never moved, and didn't attend to many courses to recycle its methods or practices, what is the real experience of this person?

On the other side, a young teacher, working less than 5 years, no vacancies in that time, so his main experience is working as a substitute teacher. when another one gets sick, in those 5 years this teacher has worked in maybe 10 different schools, of different towns, with different social background and different problems. What is the experience of this teacher?

On the other way, how can we evaluate the effectiveness of a teacher? Comparing the results of a class group this year, with teacher A, with the results of the same one last year, with teacher B?

Well to be accurate maybe we might compare several years, 3 and 3 or 5 and 5 years, then we are studying a period of 5 to 10 years, so we might add to the equation some other variables like the evolution of the students (sometimes results of students in different subjects increase or decrease along years), and we might add the different educative plans (in this country every new government changes the contents of the plans of previous governments to adapt those to their criteria), it is said that in the last 10 years none of the students in Spain began going to school following an educative plan and ended in the same line.

So my conclusion is that will be really difficult to determine what is an experienced teacher and how that reflects to the students, maybe the best we could do is give it a more marketing view: "are the students happy with that teacher? are the results good enough?"

(Excuse my english, that subject is not one of the most important in our schools).



This debate could go on for days and days; however, there are some other points made here that are misleading also. The line: "One story to think about here is that 1) until the 1960s, schoolteachers represented the best and brightest U.S. women, who didn’t have many job options available; 2) as many bright and well-educated women went elsewhere, the overall quality of teaching fell..." is inherently misleading.

I attended a school with a wonderful teacher preparation program, one of the best, and very selective. There were many very smart and bright women (and men) who were foregoing other, more lucrative fields because they believed in what they were doing. Unfortunately, after a few years of being beat down by state testing, administration who only care about test scores and are not fulfilling their leadership roles due to it, uncooperative parents, children with lessening attention spans and awful boxed curriculum, many of these smart and bright individuals leave for greener pastures. Also, being one of the only professions in which we require professionals to take advice from higher-ups who have no training in their subject matter or pedagogy, pass laws about their right to work, and treat them like monsters, is it any wonder that some veteran teachers feel defeated and may not be at the top of their game? Sure, you can eliminate older teachers and bring in young ones who can raise test scores for a short period, become burnt out and then quit and then bring in a whole new round of young idealists, but is this really sustainable for the profession?



"There is suggestive
evidence of heterogeneity: in lower SES and lower-performing schools, retirements from the ERI
program led to larger increases in test scores, particularly for reading. Although the differences
are not statistically significant, to the extent that there is a difference, we show that some of it is
driven by how schools were affected by the ERI program."

This might be relevant?

Linc Wolverton

There is another factor at play here. In a large school system with multiple elementary, middle and high schools, the assignment of teachers to facilities is not random, I believe (as a former long-term, school-board member). Poor and mediocre teachers in affluent-neighborhood schools frequently are pressured out. That is, parents go to the principals and school boards with their complaints and their political advantages and pressure the school district to remove such teachers from those schools.

What is supposed to happen according to many union contracts is that those poor teachers must undergo plans of improvement to correct their problems after they have been so identified (under a bureaucratic process that takes months to years). Mediocre teachers after their probationary periods simply are allowed to remain.

However, the parent and political problem is relieved by the administration transferring those teachers to other schools in the district until they find a place where the parental and political pressures are at a minimum -- in the poorest areas of the district typically. There, they may remain until retirement.

It is therefore not surprising that the loss of experienced teachers in some schools within a large district do not have a negative impact on teaching performance.



Although I'm sure researchers are aware, but this article does not mention the shallow conclusion that higher test standardized test scores are an accurate measure of teacher effectiveness or quality.

The other major flaw in some studies is that the test scores used to compare did not measure the same subjects with the same group of students from year to year. For example, you can't compare World History scores for Student A one year with US History scores for that Student A the next year. Subject matter is much too different. Student maturation levels change. Personal circumstances, familial, financial, etc. change and affect outcomes. You also can't compare World History scores for Student A last year with Student B this year. Researchers are often not controlling their variables well.

Like a veteran teacher friend of mine said recently, "Too often the discussion is about what public schools are producing, rather than the increase in problems kids are coming into our schools with. We see more and more kids struggling in ways that we did not see 10 or 20 years ago. More and more kids we teach today are less equipped to learn and produce the way their parents were when they attended our schools."

Are schools failing kids, or are our schools a reflection of society's tribulations and their effects upon the next generation?


Kevin Johnson

Well-put and the new problems coming in are things I try to anticipate. I also find it valuable to get to know students that have parents that are teachers, and get to know them closer, so they can fill me in on things that a teacher would not normally think or see--I choose those students because I usually know their parents well and they know me, and that they see me as a normal person, not locked in a box all summer and during weekends.

I have constantly had to change my practice. My first students would recognize what I do now, aside from the same content. The hardest part is getting rid of things that I worked hard on that I thought would succeed, but did not achieve what I expected.

Your point I have seen put thus: do students fail school or school fail students? With that in mind, I am trying to change how I teach and how its assessed by using gamification (another major topic for another day) but I do not think many teachers out there are that willing to part with what they have been doing.



I live in a state where teacher's salaries vary widely from school to school. Teacher's at the school's with the highest salaries never leave. Promotions are based on years of service not performance. Take a few Master's level classes in the summer break, and you are guaranteed to max out your salary. The result is that the school is filled with experienced 30+ teachers just coasting their way to retirement. Are they good teachers? Well, yes, the majority are good - because they started out being good. The school's with the highest salary have very few job openings because, as I said, no one leaves except to retire. As a result they have their pick of the best candidates. Does experience make them better? Maybe for a few years, but the system both fails to motivate improvement and doesn't punish those who lose interest and just coast for the rest of their career. Knowing that they can have their pick of the brightest new graduates such schools could definitely be improved by getting rid of some of the dead wood.

How this would work in the low salary schools I can't guess; are the long term teachers in those schools the dedicated ones who stayed, or did they move on leaving only those who couldnt get a job at a higher paying school.


Stewart Herring

This sentence

Alternatively, if productivity is negatively correlated with disutility from teaching, the teachers who choose to take up the ERI may be those that are least productive.

Shows that there is a need fof a "plain english campaign"

Jim s

The automobiles that roll off of today's assembly lines are far better than those of thirty years ago. Each generation incorporates the improvements of its fore bearers. So now we have better quality, better mileage for the same cost. We have gotten better at making cars. In education, however, each generation of teacher essentially starts out from scratch. Even though a new teacher may be armed with the latest pedagogy there are still many soft skills that are developed only through practice and experience. Unlike our automobiles, our education system is not getting better with each generation. When a teacher retires all those years learning how to identify learning styles for each individual student is lost. The challenge I see is to figure a way to capture the hundreds of thousands of hours of effective teaching that is done every year and make that information available to new teachers in such a way that does not make more work for them. I dont know what this will look like, but here is a possibility. Consider the advances in big data and machine learning. Although it my sound like big brother, I can imagine a classroom where each student is monitored intimately...how they answer questions, facial expressions, speed of response, who they are friends with for example. From this information a big data system could recommend appropriate pedagogy that has been found to work for this profile. And I am not talking about kids sitting around on computers all day. Our data collection techniques could get quite advanced and subtle(think google searches, voice recognition, and cameras everywhere). The teacher would focus more on relationships, motivation and community building and rely on proven cirriculum. What a waste to have an algebra teacher develop a customized lesson plan every day from year to year when it has already been done by thousands of teachers. Algebra doesnt change and the way kids learn probably falls into 10 effective strategies. Surely we can capture this and apply it appropriately and individually. Bottom line is that we need to shake up our thinking about how to give people the skills and knowledge they need to be effective in this world. Our current paradigm is old and not as effective as it could be.



Please don't start confusing people with measurements or facts. Liberals want to feel good about their work and believe they are producing results. Just like the idea that coddling and learning the history of certain groups will get them educated and pull them out of poverty. Laughable.

Eric M. Jones

Unfortunately...and I really mean UNFORTUNATELY!, K-12 teachers need great personal energy to communicate to all those little ruffians. Older teacher might be good for college, but for the earlier grades, students are more likely listen to people with high energy...and that usually means younger teachers.

teach now

I am glad that you said usually Eric. But that means schools will need to treat teachers as individuals and that will take a bit of a change of heart. I do have reason to think that alot of teacher energy is lost to all the rules and regulations that they must follow that get in the way of real teaching. and there are styles of teaching so matching student to teacher can be time consuming but definately a worthy endeavor. Showing students that you care, I found, means alot to them and it does not matter who they are, where they come from or what their differences are so clearly one needs not to interrogate such inequities that exist, but ask the right questions of them so as to enhance success at achieving what they are aiming to acommplish whether they know it or not.


It strikes me as odd that Illinois is used as the example in this story. Just check the headlines to find out how well the ERO (Early Retirement Option) is working for the taxpayers, the government, retired teachers, and career teachers like myself who have inherited one HUGE mess.


Early retirement incentives waste valuable resources.
I worked with several teachers who were forced to retire early (in Oregon, some teachers made more retired than they did working). After retiring, they just moved to another state and resumed teaching. One teacher I worked with just retired from teaching for the third time--that's three pensions from three different states!


Yeah... you lost me at 'improves test scores'. As if that was an unquestionable, proven standard of student learning and success.

All it is is proof that someone (maybe not the students) filled in correct answers on a test.

Let's try replacing the word 'teacher' with another profession. 'Is getting rid of experienced doctors going to negatively affect peoples' health?' 'Is getting rid of experienced detectives going to improve the case closure rate?' 'Is getting rid of experienced computer programmers going to lead to faster releases of software?'

Not all of these scenarios are equivalent, but the level of complete disrespect for the teaching profession evinced by even suggesting that experience -- in classroom management, in dealing with dozens of students of varying ability, at enticing kids to learn, in a million other day to day things that I don't know about because I'm not a teacher -- is not important?

About the only 'experienced' group of professionals we could maybe use less of are hedge fund managers.


Doug Clerk

Please tell me that you don't believe that students' test scores are a valid measure of teacher effectiveness - ever heard of teaching to the test? (Of course you have - I read your book!)