Arnold Kling  

Teacher Pay

About EconLog / Econlib... Organ Transplant Market?...

Peter Temin argues that teacher pay is too low. He says that in the past, limitied opportunities for women meant that schools were supplied with a pool of high-quality teachers willing to work for low pay. Today, however, talented women have other opportunities that pay better than teaching. He concludes,

The implication of this analysis is that it will be very hard to improve the quality
of American schools. We will need a concerted effort to break out of this stable
equilibrium, attract a new breed of teachers, and settle into a new higher paying
equilibrium. None of the current reforms even comes close to making that attempt. They are doomed to failure as a result, and the available empirical evidence suggests that they are in fact failing.

I once put together a back-of-the-envelope business model for a school in which teacher pay would be $90,000 a year. However, it requires trimming back on overhead.
For Discussion. If schooling were privatized, with a voucher system, do you think that competitive forces would lead to higher or lower teacher quality and pay?

Comments and Sharing

COMMENTS (18 to date)
achilles writes:

Seems like a very interesting paper, thanks, I will enjoy reading it.

Arnold, Iyou probably know the answer to the following question (or at least can give me a ballpark figure): what are the salaries for teachers at private and parochial schools, ranging from the elite ones like Andover Phillips to more generic Catholic schools? I have always been curious about how such salaries compare both to public school teachers and to college instructors, since I would expect elite private schools to screen very highly for quality and pay top dollar much like good business schools and econ departments do. I don't know anyone in that profession well enough to ask personally.

Arnold Kling writes:

I can't speak for elite private schools, but for typical private schools teacher pay is lower than at public schools.

The private schools maintain parity or better in quality because:

1. They can fire bad teachers.
2. For teachers who are motivated primarily to make a different in young people's lives, they offer a more supportive working environment.
3. In the public schools, there is a lot of pull toward adminstrative positions. Several of my daughter's best teachers left (in the middle of the semester no less) to become co-ordinator of this or administrator of that.

For what it's worth, I originally wanted to teach in public schools, but they did not have a way to accomodate a part-time, volunteer teacher without any experience or teaching certificate. So I teach at two local private schools.

Scott writes:

Here in Arizona the Catholic schools pay their teachers less than the public schools. However, it’s a better teaching environment for the teacher. Hence it can be seen as a tradeoff between salary and working conditions.

Teachers at charter schools here tend to get paid more than public school system teachers. It amounts to a few thousand dollars a year. They also have better working conditions, no union, and can get hired or fired fairly quickly.

Looking at the teacher pay issue from a productivity angle I think is instructive. For 20 years the teacher unions have fought for smaller class sizes and less class time for their members. They have been largely successful. However, average educational attainment doesn’t seem to have increased. If the output of the school system is supposed to be educated students, then that means teacher productivity has been falling. From an economic standpoint that means pay should be declining as well.

Jim Glass writes:

"The private schools maintain parity or better in quality because..."

4. They don't impose Byzantine Ed School certification requirements to limit the market of available teachers (and thus drive up salaries).

Anyone can teach in a private school if the those running the school deem them qualified to teach. Not so in the major unionized urban school systems. In fact, a "reverse filter" applies, since an Ed School degree is generally required for certification, and Ed School students as a group have the lower average academic qualifications (SAT scores etc.) than people in other University departments. So not only are great numbers of graduates in other fields steered away from teaching but those steered away have better average qualifications in terms of academic achievement.

Larger pool of better qualified job applicants = better employees at lower cost.

When PBS was running those "all big expert" roundables a few years back they did one on public education and an Econ nobelist, Becker IIRC, observed, "I can teach econ at any university and I can teach it a private high school yet I am deemed unqualified to teach econ at a public high school." To which the Education Commissioner of my home state, NY, replied in words that as a proud taxpayer of NY I shall always remember: "Nobody ever taught you how to conduct a birthday party in class".

There's a great book that reveals more about how the urban school systems actually work than you'll get from reading a hundred academic studies: _Shut Up and Let the Lady Teach_ by Emily Sachar, a former NYC public school teacher. It recounts her personal experiences.

She was a Stamford graduate in economics who had spent 10 years as a prize-winning journalist for Newsday, who decided altruistically to give it up for teaching -- *exactly* the sort of qualified achiever the public schools claim to want.

Her experience in getting hired was that she expected she would naturally teach English or Math, due to her work experience and education. But she was deemed unqualified to teach English because she hadn't studied English in college. (Her book on all this was nominated for a Pulitzer). And she was deemed unqualified to teach Math because she had no algebra courses on her college transcript, only calculus courses -- she'd finished studying algebra in high school, and that didn't count. So they assigned her to teach Social Studies, about which she knew nothing. She lasted one year -- the altruism beaten out of her.

It's really a very fine and disturbing book. A few years old and out of print, but well worth looking up in the library. (As I said, nominated for a Pulitzer.)

Jim Glass writes:

Maybe the most interesting issue in all this is the unexamined assumption that raising teacher pay will in fact actually raise teacher quality -- as Temin and so many others simply take for granted.

Of course that is the natural first thought -- but it assumes a functioning free market for teachers. And we know that in the highly unionized urban school districts, at least, this market is instead highly regulated in ways that look outright bizarre if one takes the trouble to examine them.

So does the natural "higher pay will improve teacher quality" effect actually survive through the regulation? There is peer-reviewed research on it, although the results aren't popular with those who make the assumption. E.g.:

"Can Public Schools Buy Better-Qualified Teachers?" By David Figlio, U of Flordia, in _Industrial and Labor Relations Review_, July 2002.

"This analysis ... uses panel data on new teachers in 188 public school districts ...

"For nonunion school districts, the author finds a positive, statistically significant relationship between a given district’s teacher salaries and that district’s probability of hiring well-qualified teachers.

"Several tests indicate that this relationship is *not* found in unionized school districts."

How could "higher pay leads to better quality" be true in non-unionized school districts, but not in unionized ones? Well, look at the typical union rules in urban school districts:

First, the higher pay always goes to *everyone*, not just to new hires or to teachers of specific subjects where there may be a shortage of teachers (like math and science). So every incumbent in the system gets a pay raise in the name of "improving quality" without their quality having been improved at all. Another name for this is "rent seeking".

Second, look at the effect of union rules on new hires -- the only people whose qualifications may be affected by higher pay. Take for example union lock-step seniority pay raise rules. Every teacher with the same amount of time in the system gets the same salary -- no matter what they teach or how good or bad they are at it.

Put yourself in the place of a top-achieving graduate from a top university who is drawn to teaching with the intention of making a personal difference in the lives of kids and having an impact on the system too, as achievers are wont to do! Then suddenly you realize that no matter how hard you work to achieve, you will never in all the coming years receive any pay raise above, or rise in the ranks any faster than, or have any more influence than the least able, most uncaring and slaggard fellow teacher who started working the same time you did. Your superior performance brings no reward.

What does that do to your morale? How long do you stay in that situation, when you can go somewhere else and actually make a personal difference, with your individual contributions being recognized in your paycheck as well as otherwise? How long did Ms. Sachar (of Stanford, not Stamford, sorry) remain teaching in NYC?

Another well-documented phenonema of unionized urban school districts is that turnover is highest among the most qualified new teachers -- they quit the fastest. So who is left collecting the pay raise?

OTOH, in non-union school districts labor market defects such as these do not exist or are much lesser. Pay raises can be targeted to make a difference -- e.g. on new hires, and on pay-for- performance. So one would naturally expect increased pay to have a larger impact on teacher qualifications than in urban unionized districts, and indeed they do.

Chris writes:

If there was no fundamental change in what parents expect from the school systems, things would get worse. In your hypothetical profitable school essay, you are assuming that all the teachers time is spent teaching. In fact, I think teachers spend most of their time on administrative stuff, discipline, socialization issues that should have been handles by the parents years ago, etc. In short, they spend very little time actually teaching, which is obviously related to your point that the schools have too much money tied up in overhead. I read somewhere once that the average school student directly interacts with a teacher for 2 minutes of a typical 50 minute class. The vast majority of a student's time is spent waiting for something else to happen.

That seems to say that in order for the current system to work, we need a lot more teachers.

A voucher system, which would encourage instutitional innovation, would likely provide good teachers with increased job satisfaction that could replace at least some monetary compensation. This should lead to an increase in teacher quality at the same salary levels, maintenance of teacher quality at lower salary levels, or some combination of the two.

That said, one danger with looking at the existing schism between public and private school salaries is that the private school teachers may be self-selecting for not only innovation, but also less 'need' for monetary compensation. Perhaps the typical private school teacher is part of the fraction of the workforce that has independent sources of income, either through a spouse's job or otherwise. If we were to dissolve the public school system and implement vouchers, it might turn out that many of the public school teachers (even the good ones) put a higher value on monetary compensation than current private school teachers.

Regardless, if we were to implement vouchers and then discovered that the per pupil voucher amount was not sufficient for the community's desired level of student achievement, for whatever reason (which could be easily assessed at that point by comparing schools that charged the just voucher amount vs. schools that charged the voucher amount plus an additional fee), then the community could increase the value of the voucher.

Blair writes:

One of the big drawbacks to any voucher scheme is the adverse selection problem. A market in a good that everyone is required to consume self-stratifies, as for example auto and medical insurance markets have done. Such a market in education would undermine two of the principal social goals of public education: (1) socialization and (2) implementation of a minimum standard. Put more concretely, a voucher system would make the educationally well-off better off, would make the poorly-off worse off, would Balkanize society, and would create an educational under-class. Vouchers might be nice in theory (at least when you assume away all the hard stuff), but they're terrible public education policy.

The simple fact is, we're under-funding public education, and no "free market" reshuffling of the boxes will make that go away. In fact, it could easily make it worse; public education is in many respects a public good, and it's a classic result that the free-market equilibrium in a public good is suboptimal.

Maybe if we made it a "national security" issue...

"public education is in many respects a public good, and it's a classic result that the free-market equilibrium in a public good is suboptimal."

It's not a free market equilibrium when government is paying for the service. Just as competitive outsourcing may be the most efficient method of ensuring other public goods, such as clean streets, vouchers may well be the most effective way to implement publicly-financed education.

The so-called 'adverse selection' problem of vouchers is overblown - a realistic implementation of a voucher system would not dissolve the public schools, but rather give parents the option of sending children to the public school (thus ensuring the 'minimum standard') or to one of the available private schools. It wouldn't take much of an outflow for the public school to start looking for ways to become more competitive. The children of passive (or uninformed) parents would therefore benefit from the decisions and actions of more informed and knowledgeable consumers, just like in other markets.

And as for working "in theory (at least when you assume away all the hard stuff)" - I thought that *was* central planning...

Patrick R. Sullivan writes:

" a voucher system would make the educationally well-off better off, would make the poorly-off worse off, would Balkanize society, and would create an educational under-class."

All of the above problems exist in the current public school regime. Vouchers would give the poor, stuck in lousy schools a fighting chance for an education (if they want it). It's not the people living in good school districts who want vouchers, it's those who don't.

BTW, education is clearly NOT a public good, as it flunks both the tests of non-excludability and non-rivalrous consumption.

Scott writes:

Here in Arizona charter schools are providing some competition for the public school systems. The new superintendent of the Scottsdale school system came into office vowing to win back the students that had left for the charter schools.

Competition is a good thing.

Scott writes:

Just who is it that supports “socialization” as one of the objectives of the public school system. The liberal politicians certainly don’t. They send all of their kids to private schools. So do an astounding number of public school teachers.

People vote with their feet and it’s apparent that supporting socialization isn’t high on parents’ agendas.

Patrick R. Sullivan writes:

Regarding Scott's comments, Caroline Hoxby has the data that reinforces his points, both about competition generally, and Arizona specifically at:

And Robert Maranto documents the Arizona Charter School successes in this amusing article:

From which:

" The Roosevelt district, an inner-city
district in Phoenix that was hit hard by charters, sent letters to local charter parents
asking why they had left district schools and explaining how the district would serve
them better. Teachers and administrators in the nearby Isaac district visited parents in
their homes. The largest district in the state, Mesa Unified, responded with particular
vigor. Mesa sends policy staff members to the state department of education to study
charter proposals, checking up on the competition. Since 1996, Mesa has conducted
customer-service training for staff, developed a “Red Carpet Treatment” for reintegrating
charter parents into district schools, and expanded all-day kindergarten. Mesa officials
initially denied, but later confirmed, that their actions were motivated by competition
from charters."

steven kyle writes:

Patrick Sullivan is wrong when he says public education flunks the test of a public good. Yes, its true you can't consume my education but you sure can benefit from it. It's a good thing for us all when everyone can read and understand current events and are all able to exercise intelligently their functions as a citizen in a democracy. It is even useful to capitalist rightwingers to have a large well educated workforce who can understand the instructions they are given and read the directions for whatever job they do. Then they dont have to spend their time and money searching for good employees. I'm sure you can think of others

Patrick R. Sullivan writes:

The traditional test of a public good is that it is both non-rival and non-excludable. Education is neither. What Steven Kyle is talking about is a positive externality, but, as David Friedman is fond of pointing out, there are also negative externalities associated with education.

Jack Trow writes:

I have been an HR Manager for the last 28 years in business (ExxonMobil,COSCO and others) - and I come from a family of public school teachers in NYS. My daughter just graduated with a Teachers Degree and - amongst stiff competition - was lucky enough to land a 6th Grade Inclusion Class here in the Hudson Valley. Union dues $600 up front and pay will be about $39k.

Teacher pay may be low in some areas - cities cannot attract - suburban America has no problem. But what totally astonishes - coming from the business world - is the draconian pay structure for Teachers - we have very comprehensive merit (pay for performance - motivitation through merit) models built in - teachers are hung up with step up, fixed pay, tenure, cash payout of unused sick time ( I've seen the $60K checks made out upon retirement for unpaid sick time,etc) - there are so many dymanics in performane that your article didn't address! As in any business, we must look at total cost and not total pay. As a Labor Negotiator, I could speak forever about motivation and productivity. Thanks.
Jack Trow

Kate writes:

I am a public school teacher (Vermont). I would like to know if there are public schools in which teachers are not unionized. I am particularly curious to know how such teachers are contracted and paid. We use collective bargaining and reach agreement with the school board on a multiyear contract which includes conditions of employment and salary scales. My personal interest stems from a desire to understand the implications of a "pay-for-performance" arrangement. I do not know of any public schools that use any method other than years of service to determine pay. I would very much appreciate any information regarding this. Thanks. Kate

Sean writes:

I don't think that "throwing money at the problem" will solve the problem of a decrease in the marginal productivity of teachers. I don't think teachers go wake up and go to work with the thought, "I am not going to teach my best today, b/c I am not paid enough."

Comments for this entry have been closed
Return to top