« What's going on with the economy? | Main | Cai Guo-Qiang, critic of collectivism »
An educational experiment with higher salaries
A New York City charter school set to open in 2009 in Washington Heights will test one of the most fundamental questions in education: Whether significantly higher pay for teachers is the key to improving schools.
The school, which will run from fifth to eighth grades, is promising to pay teachers $125,000, plus a potential bonus based on schoolwide performance. That is nearly twice as much as the average New York City public school teacher earns, roughly two and a half times the national average teacher salary and higher than the base salary of all but the most senior teachers in the most generous districts nationwide.
Here is the story, thanks to Kurt Muehmel for the pointer. For many years I've been telling some of my Ph.d. students that they should consider teaching in private high schools. None of them seem to listen but maybe this will have some impact.
Posted by Tyler Cowen on March 9, 2008 at 05:16 AM in Education | Permalink
Comments
I don't see how this would solve very much. Presumably, this school would have its pick of teachers in the NY metro area, practically guaranteeing that they can find the best (assuming they have some way of distinguishing quality). This won't tell us anything, however, about the potential effects of across-the-board pay increases -- are you drawing in higher quality people or simply paying the same underperformers more?
Posted by: Mario at Mar 9, 2008 6:14:06 AM
Presumably salary competition in the education sector would, over time, draw in talent from other professions which would increase the quality of the pool. Most white collar professionals wwould consider teaching for 125k a year but not for 60k. You'd think this expansion of the pie would be a good thing for students.
Posted by: N at Mar 9, 2008 7:12:19 AM
This seems a rather roundabout way of addressing the problem. Mightn't it be simpler to offer significantly higher pay to the students? Since they're currently working gratis, and (enterprising black marketeers excepted) seldom get more than the minimum wage on the job market, the chance of any kind of (shock horror) remuneration for (shock horror) schoolwork might be expected to act as a significant incentive.
Posted by: Helen DeWitt at Mar 9, 2008 7:40:23 AM
I think it's important to note (eg forestalling Mario's question) that they're not just offering significantly higher pay, but they also have an exceptionally rigorous interview process which eliminates many peopel from consideration (eg anyone without 90th %ile test scores and anyone without teaching experience). Actually they've thought about a lot of things that I, as a teacher, am very excited about -- things that speak to student learning but also to teacher morale. Things which experienced teachers might find as interesting as the salary, if not more so, in terms of compensation. (Inexperienced teachers or people outside the field would hear the compensation number first, but given that the interview process knocks people without teaching experience out of consideration, I don't think this actually expands the pie per N's comment, except insofar as it might bring former teachers who have left, back.)
I would also make sure your PhD students know that private schools do not in general pay 125K :P. Note that the school in the article is charter, not private. Privates typically pay less than the surrounding area (10% less is a good target to strive for), with the tradeoff being smaller classes, more motivated students, and less bureaucracy. (On the other hand, private schools also expect you to invest a lot more of your identity into them.) I've been teaching for five years and I don't make a third of what this charter is offering, a fact to which your students may be sensitive...
Posted by: Andromeda at Mar 9, 2008 9:19:49 AM
You've forgotten the hyperlink, Tyler.
There's also a good piece in the Times magazine today, a round table Q&A with some philanthropists and prominent administrators in education. The Times asks what they would do for schools with a 2 billion dollar gift.
It's interesting to compare the ideas of the Chancellor of NYC public schools, Klein, with those of Steve Barr, a charter school founder.
Klein seems to think targeting districts is more important than teachers or students:
I would look for the most promising individuals and make heavy investments in them. Let’s say you choose Michelle Rhee, the new schools chancellor in D.C. That school system has long been one of the worst-performing in the country, and Michelle wants to really overhaul it. I think our philanthropist could make an eight-year bet on her. It’s the same kind of thing I would have wanted to have happen to us when we started six years ago in New York. To start, I’d give her a couple of million to do some planning.
http://www.nytimes.com/2008/03/09/magazine/09roundtable-t.html?_r=1&ref=magazine&oref=slogin
And Ms. DeWitte, I enjoyed your novel, especially the first chapter.
Posted by: Rue Des Quatre Vents at Mar 9, 2008 10:16:06 AM
The problem I see with this school is that it is the brainchild of inexperience. The principal (and designer) of the school is a 31 year old yale grad who taught for three years in Teach for America at a middle school in Washington Heights. So, obviously... being bright and well educated, he figured out the problem quick.
His great plan (as per the NYT) is to attract the best and brightest by spending more money on them, and to make up the shortfall by taking only a $90,000 salary himself, having no assistant principals and no disciplinary deans, including only one or two social workers and no attendance people. He'll have a longer day and a longer year. The teachers will do all the admin work themselves and teach low performing and area students with 30 kids in a class. Basically, his idea is that if all the teachers were as bright and ambitious as he is, schools wouldn't need all those frills... like low class size, time to create curriculum, disciplinary deans to handle the raging, hormonal distractions that threaten to interrupt every day. Good luck with that.
Posted by: a at Mar 9, 2008 10:38:05 AM
Does not scale.
Posted by: josh at Mar 9, 2008 10:43:35 AM
Something strikes me as weird about this that I can't quite put my finger on. Perhaps it's the huge gap between teacher salaries. Perhaps this could bring some new blood into this one school. But if other schools followed, then there would be competition for the new teachers. Would there be enough good teachers to go around to maintain the same effect? Will the results be scientifically accurate if the effect cannot be duplicated? Will this give bad teachers more ammo to increase their salaries?
Such dramatic social experiments are almost always broken from the start. That's not to say I wouldn't apply for a position were I a New York teacher...
Posted by: infopractical at Mar 9, 2008 11:18:49 AM
The partial equilibrium effect: Massive flow of highly qualified professionals and teachers to high paying schools.
The general equilibrium effect: ???
What happens when more than one school embarks on this plan is much more difficult to predict. Permanently higher quality of teachers, and attraction of talented individuals into the field? A bankrupting of the school system? A further divide between rich and poor districts in the quality of schooling, and a final crushing blow to the health of the public school system in underprivileged districts? Danger, Will Rogers!
Posted by: Chris Blattman at Mar 9, 2008 12:36:13 PM
Some back of the envelope numbers:
There are about 4 million primary and secondary school teachers in the US
Average class size is about 21
Average teacher salary is about $40,000/yr
So if we increased average salaries by 50% (to about $60K) and increased class sizes by 33% (from 21 to say 28) the net cost would be about $20 billion per year. Perhaps there would be some savings if fewer administrators, social workers, etc. were needed to support the more skilled teachers that the higher salaries would attract (or perhaps as with the NY experiment, the teachers could be asked to work longer hours in exchange for the higher pay).
Is $20 billion a lot?
$20 billion is about 1/7 of 1% of US GDP. Or to put it another way, according to Joseph Stiglitz, the cost of the Iraq war is about $720 million or day, or over $20 billion per month
Would 50% higher pay make a difference?
Since the quality of teachers is apparently the number one determinant of educational outcomes, getting better teachers would seem to be a good strategy. The average salary in the US for a lawyer with 1-4 years of experience is about $64K. I suspect many of these people would prefer to be teachers if the salaries were comparable. Likewise for many other jobs like middle managers, etc. The influx of competition would be expected to increase the average quality of teacher entering the profession and staying there (selection effect) as well as strengthen incentives for doing a good job (to keep earning that salary). Indirect effects would be increased prestige for the profession, more interesting colleagues to work with, and over time, more micro and macro innovation in teaching, amplifying the benefits of this policy.
Would a 50% increase in teacher salary improve education quality enough to eventually grow GDP by at least 1/7 of 1 percent more than the status quo?
With about 50 millions students in the US each year, and the economy increasingly depended on knowledge work, the answer to that seems obvious to me.
p.s. Sorry, but I had to remove all the hyperlinks supporting the numbers above. TypePad's antispam filter tagged them as "potential comment spam". I think Tyler or Alex may have the option of posting the original version, with hyperlinks, if they override the filter.
Posted by: A student of economics at Mar 9, 2008 12:38:37 PM
I don't see why we should pay all teachers more. It's really the math and science teachers that are harder to find, which is not surprising. People who understand math and science have more outside options than people who study the humanities do.
Posted by: Curious at Mar 9, 2008 12:50:17 PM
Here is the original article. Within the text of the article, one will find the link to the charter school itself.
Posted by: Yancey Ward at Mar 9, 2008 12:59:15 PM
Given time, higher pay will draw in a higher quality teacher pool- this is simple economics. The novel thing about this particular proposal is that it reallocates resources from the bureaucratic support and the technology (computers, for example), and redirects it towards higher salaries. Teacher unions are not going to like this system, however, since it will weed out poor teachers, and if such teachers become unionized, the benefit that arises from the weeding process will be lost, and higher salaries will not fix it. I call this institutional rot. I think avoiding this rot will be Mr. Vanderhoek's real challenge, should he lead the school longer than the four years he has promised.
There are two problems with the way we run education in this country, in my opinion. The biggest problem is the students and their parents/guardians. I don't see a systemic way to fix this within the education system- these are deep cultural problems. The other big issue I see is the high level of overhead in the education system. This arises from teachers' desire to be promoted out of teaching- to be promoted up into the bureaucracy where the job is less taxing and higher paying. Every public and private education system I have seen seems to be structured this way- except, now, for the proposed one. Paying the teachers more and making the job of teaching the primary and premier job within the schools may have a lot of merit. I think Mr. Vanderhoek is onto something.
Posted by: Yancey Ward at Mar 9, 2008 1:22:26 PM
This is nothing like a controlled experiment on the value of paying teachers more.
I think we already know from private schools that their main advantage is not how much they pay their teachers--they actually get less than their public school counterparts. A more likely source of advantage is better classroom discipline, maintained by the power to cull disruptive students.
Nevertheless, this (uncontrolled) experiment may produce modestly better results because of initial selection advantages, absence of union work rules, etc. Then the unions will say, "See, we need our pay doubled to see an improvement in outcomes (but let's please still not try to measure outcomes)." Then we'll see their pay skyrocket. And there will be no improvement.
Posted by: M. Hodak at Mar 9, 2008 2:42:12 PM
Student of economics: STUDENT quality, not teacher quality, is the number one determinant of educational outcome. As far as teacher quality goes, outcomes behave like an S-curve relative to quality. There is a continuum of bad teachers. A good teacher (near the top of the S-curve) is effective. A great teacher is only marginally more effective than a good one. Once your teachers are adequate, you hit the point of diminishing returns very quickly.
(The hypothesized reason for this is the following. Each student has some "innate potential". A bad teacher will not help them reach this potential. A good teacher will come close (say 90%). A great teacher may reach 95% or 99%, but never higher than 100%.)
Also, are you accounting for the opportunity cost of skilled workers elsewhere in the economy?
Posted by: mathgeek@mathsciences.edu at Mar 9, 2008 2:56:47 PM
interesting. I agree with "N" above in that higher wages should draw workers from other professions. It continues to be a factor for me, personally. I have considered teaching but work as an engineer instead because the pay is so much better.
Posted by: jason at Mar 9, 2008 3:03:40 PM
Mathgeek:
1. You may be right about good vs. great teachers. In any event, my suggestion of $60K/yr is probably aimed more at shifting the quality from bad to good, than all the way to great (except for those great teachers willing to work for a lot less than they could probably make elsewhere).
2. You are right about the opportunity cost of skilled workers elsewhere. It's my subjective assessment that shifting someone from being say, a lawyer or marketing manager, to being a teacher would be a net plus for the economy. It's less clear for engineers (like jason), but my guess is that teaching has more positive externalities than most professions.
Chris Blattman's concern about inequality across school districts suggest that any such plan should be implemented nationally (or at least statewide) rather than by individual districts (although experiments are always welcome).
I would agree that the extra pay would be wasted if it didn't lead to a change in teacher quality. Hence, it would be important to measure outcomes carefully and be aggressive about selection when hiring rather than just hiring the same people as before. I suspect most school districts and principals do prefer good to bad teachers, and would have the right incentives on the margin, as long as they were relatively free to make these decisions. This is especially true if educational outcomes are measured and well-publicized.
Posted by: A student of economics at Mar 9, 2008 3:26:16 PM
This is a great idea; and while we're at it, we should pay engineers 3x their current salary as well. Just think how great they'll be! Somehow, this will deal with the problem of the 50% of engineers that are below average -- they just won't make the cut!
And why stop there -- we'll pay lawyers 3x as much too. If it works to marginalize bad teachers, surely it will eliminate the problem of bad lawyers. Then politicians -- just think how much more effective and efficient government will be when real CEOs and investment bankers can afford to be civil servants!
Is there any sector that can't be cured by massive pay increases? [ugh]
Posted by: st4rbux at Mar 9, 2008 4:46:49 PM
If school wasn't a government monopoly, I wouldn't know or care how much the teachers were paid & I wouldn't have to.
In every other facet of my life where I depend on people to meet my needs, their salaries are completely a private matter between them & their employees. If they aren't providing satisfactory service, then they work it out. I have never held a meeting with my neighbors to discuss whether we should change the pay structure at the local Century 21 office in order to make a satisfactory realty market. The idea would be absurd.
To the extent that we're even having this conversation, I think we're not seeing the forest for the trees here.
Posted by: kebko at Mar 9, 2008 5:19:51 PM
If school wasn't a government monopoly, I wouldn't know or care how much the teachers were paid & I wouldn't have to.
In every other facet of my life where I depend on people to meet my needs, their salaries are completely a private matter between them & their employees. If they aren't providing satisfactory service, then they work it out. I have never held a meeting with my neighbors to discuss whether we should change the pay structure at the local Century 21 office in order to make a satisfactory realty market. The idea would be absurd.
To the extent that we're even having this conversation, I think we're not seeing the forest for the trees here.
Posted by: kebko at Mar 9, 2008 5:19:56 PM
I'll watch this, just like I watch every other alternative approach to education, and hope that it adds something to the discussion. As an engineering manager, and trustee of a private school, I'm skeptical ... it really doesn't seem at all novel. Highly paid engineers (or other technical professionals) rarely have what it takes to be good (much less exceptinal) teachers, and great teachers (although typically underpaid) don't do it for the money. Teaching, just like managing, engineering, or even sports, is a skill set and to the extent that they are able to attract a greater % of the most skilled, they'll out perform other institutions with less talent.
Assuming competent teachers, student quality and parental involvement seem to be the biggest drivers of performance; couple that with the fact that this doesn't scale except over the long-term as it brings wins greater numbers of high-potential people away from other professions.
Posted by: Mike at Mar 9, 2008 6:50:19 PM
---one of the most fundamental questions in education
well, I think that proves how unsound education is.
They have more than enough data to find the answer to whether salary matters, when controlling for other variables. That they don't know that is the isssue. That they can't do those metastudies, that they can't understand controlling for variables, that really shows how immature their discipline is.
Posted by: mouse at Mar 10, 2008 12:36:24 AM
The school will see significantly "better" results from its students in terms of standardized test scores and grades if the program attracts better students.
The number one factor in determining how well a school or classroom performs is the intelligence and innate aptitude for acquiring knowledge of its students. You cannot socially engineer intelligence, regardless of how much we would like to believe otherwise.
Posted by: Mike B. at Mar 10, 2008 3:30:27 AM
Mouse - the lack of research into basic problems of education is generally indicative of the kind of person who pursues graduate studies in education (at least at the M.Ed. level). Your typical M.Ed. is only interested in getting their Master's because it puts them higher on the pay grid when they return to work. That may be a broad generalization but at my library it is a topic that comes up frequently amongst the librarians - M.Ed. students are generally speaking the only grad students that don't really have a passion for advancing their discipline.
I think it would be interesting to do a study to see where the majority of ed. PhD's started their academic journey, that may also explain a lot about their discipline.
As for paying the teachers more - that is part of the problem, but if I were to ask my friends who are teachers why they are teaching money would be way down the list. I think, as others have said, student quality is a big part of the equation.
Posted by: Ian at Mar 10, 2008 6:46:00 AM
Student quality and family engagement are of course huge factors, but they're not the only factors. Most teachers don't have good economic incentives to do a good job, honestly. Saying that increasing pay won't help because most teachers you know aren't doing it for the pay is just using a self-selected sample. There are probably a good number of people out there who would be good teachers but are not doing it partially for lack of pay. Bureaucracy, the difficulty of the job, and so on are of course additional factors.
The thing that interests me the most about this debate is the analogy with CEO pay. If you believe that paying teachers more is valuable, does that give you pause about the argument for paying CEO's enormous amounts because of the scarce talent pool? Likewise, if you believe that current CEO pay is justified (compared with historical figures, current pay is often much more than 3X), why shouldn't higher teacher pay be justified?
Posted by: Greg at Mar 10, 2008 7:42:59 AM
