The $320,000 kindergarten teacher

Very good kindergarten teachers raise students’ future earnings by $320,000 a year, estimate a team of Harvard economists. They looked at 12,000 Tennessee adults, now about 30 years old, who participated in Project Star in elementary school.

While Project Star was designed to show the impact of small  classes (13 to 17 students) in the early grades, the data also showed some teachers were much more effective than others.

Some teachers’ students earned higher test scores for awhile, but the effect faded out in junior high school. But Raj Chetty and colleagues discovered other effects that didn’t fade. Dave Leonhardt writes in the New York Times:

Students who had learned much more in kindergarten were more likely to go to college than students with otherwise similar backgrounds. Students who learned more were also less likely to become single parents. As adults, they were more likely to be saving for retirement. Perhaps most striking, they were earning more.

. . . A student who went from average to the 60th percentile — a typical jump for a 5-year-old with a good teacher — could expect to make about $1,000 more a year at age 27 than a student who remained at the average. Over time, the effect seems to grow, too.

The Harvard researchers “estimate that a standout kindergarten teacher is worth about $320,000 a year,” based on “the additional money that a full class of students can expect to earn over their careers,” Leonhardt writes.  He guesses that children learned soft skills — “patience, discipline, manners, perseverance” — that improved their life prospects but not their test scores.

Like Core Knowledge Blog, I wonder why these skills — especially discipline and perseverance — wouldn’t raise reading and math scores over the long haul.

One of my daughter’s very best teachers was her kindergarten teacher, Janet Rose. It was a very good year.

Update: Robert VerBruggen is skeptical about the study.

About Joanne


  1. As I recall, the effects of reducing class size in the Tennessee project weren’t repeatable elsewhere. Maybe because so many changes were made at once that it was hard to know which change or combination of changes led to greater achievement. So it’s interesting to see that the project led to other findings that certainly should be repeatable– have good teachers, starting in kindergarten.

  2. These would be the same economists that told us the economy is fine and there is no housing bubble.

  3. How then does this dolt explain the fact that when I was in school in the 50’s and 60’s, the average class size in NY was 35.5? Somehow we managed to survive that, and as statistics will back up – there were more high school grads back in the 60’s who were honor students, and national merit scholarship winners than in the 90’s and up when class sizes were lowered.

  4. tim-10-ber says:

    I live in Tennessee and everything I have ever heard was how Project Star was a failure…how did they track these people…

    Yes we know kids must have strong elementary teachers (academically strong elementary teachers) to be successful in middle and high school…if this is true about the kindergarten teachers what does it mean about the caliber of teachers in pre-school and head start programs…shouldn’t they be degreed too? Interesting…time to raise the caliber of elementary teachers the most critical link in the chain and today one of the weakest…

    It will be interesting to see what comes of this information

  5. This is such nonsense. Teachers are irrelevant unless you had a rilly rilly awesome lady who taught you how to stand in line and share milk and cookies? Come on. It’s absurd.

  6. tim-10-ber says:

    think i just answered by own question…next step is college ready standards for pre-school…what are we suppose to do…hand the kids over to the schools at birth…

  7. palisadesk says:

    There is solid data to support the existence of very powerful teacher effects, especially at the earliest levels.

    See the account I summarized on Kitchen Table Math of a landmark study published in Harvard Educational Review:

    Better yet, get a copy of the paper (unfortunately, it is not online anywhere, but the data are very comprehensive). I got my district’s professional library staff to get me a copy through interlibrary loan.

  8. Looks like a very interesting study, which deserves an in-depth reading. I am impressed by the fact that the researchers understand the concept of Net Present Value (ie, that a dollar in the future is worth less than a dollar now, and that this effect can be accounted for mathematically.) Studies of this type often fail to adjust for this point.

  9. (Downes): “These would be the same economists that told us the economy is fine and there is no housing bubble.”

    Really? Names? Or is this reflexive opposition to standard economic analysis?

    (Joanne): “The Harvard researchers ‘estimate that a standout kindergarten teacher is worth about $320,000 a year,’ based on ‘the additional money that a full class of students can expect to earn over their careers’, Leonhardt writes.”

    Since homeschooled kids stomp conventionally schooled kids on usual measures of academic performance, I wonder what a homeschooling mom is worth. $320,000 = 20 x $16,000. Looks like we should pay homeschooling moms $16,000 per child per year, and that’s not taking into account the reduced exposure to legal liability that homeschoolers provide to taxpayers.

  10. Joanne,

    I think you’ve made a few critical mistakes on this one.

    1) You trusted the reporter to get the findings of a complicated academic paper right. I don’t have access to the original paper, and I have not spoken to the researchers myself, but the pieces makes claims tht are unsupported by the powerpoint summary that the article links to.

    2) You didn’t notice that they are projecting lifetime earnings for people who were in kindergarten in the 1980’s. If the increase in scores that the Star study found did not persist over time, why should be presume that that increase in early lifetime earning persists over time? I would think that particularly with THIS data set that researchers ought to be very careful with projecting future outcomes based only on early outcomes.

    3) There is no explaination of how teacher quality was measured. Rather, the slides admit that it is impossible to separate peer effects from teacher quality in kindergarten, given their data. Heck, they can’t even show kindergarten growth, so we don’t know how much gain to split up among different potential contributors, anyway.

    Any time someone talks or writes about teacher quality, we should ask them how they are measuring it, and how they re disentangling it from all the other factors.

  11. Cranberry says:

    Ceolaf, if we’re looking at the same slides, there are graphs showing an increase in later wages correlated with teacher experience. This would be an argument to attempt to retain as many experienced teachers as possible (experienced meaning more than 9 years of experience.) However, in the real world, a school’s workforce is not that stable. Teachers get old, and retire, just like everyone else. In order to have a supply of experienced teachers, there must also be a number of rookies learning the ropes. Some students must sit in those classes.

    Perhaps the suburban districts tend to have a more experienced teacher workforce. I know my district tends to hire experienced teachers, because they have more applicants than openings. The experienced, reliable teachers are not evenly distributed. That doesn’t mean that one can produce an education system in which all teachers have more than 10 years of experience.

    One way around it would be to encourage team-teaching, in which a younger teacher and a veteran teacher would teach the equivalent of 2 classes.

    There is also the question I’m surprised no one has posed. If a stellar teacher were to increase later earnings, does a very bad teacher decrease earnings?

  12. I learned a long time ago in Elementary Ed classes at USF to be skeptical of any educational “research”. Too many variables is one problem. It is too easy to skew the results to an intended outcome.

    Having said that, I do agree with the premise that great teachers have a tremendous effect on ‘some’ students. Typically great parents have the greatest effect and while in the past almost all children went to public schools, today I see many of the best parents homeschooling their children.


  1. […] this reminds me of the $320,000 kindergarten teacher, another argument that vvery good early education can teach behavioral and emotional skills that […]