Consider a very important profession, say superhero. Having one superhero to stop the Joker is worth $1M. Having two superheros is worth $1.1M - Batman is sufficient to stop the very dangerous Joker, but two superheros can stop both the Joker and the far less dangerous Harleyquinn. The marginal productivity of an extra superhero is $100k even if the mean productivity is $550k. As a result, superheros get paid $100k because if any superhero quits, society loses only $100k rather than $550k.
There are professions where that's a good model, but it seems remarkably unrelated to the dynamics of the teaching market.
But let's say there are multiple employers competing for the superheroes, each with their own bad guys to hunt down... Then you can get paid more.
A monopsonistic situation can drive wages below the marginal product. A monopolistic situation (e.g., multiple school districts, 1 union) can drive it above the marginal product. What I described is simply the ideal situation.
Americans spend more on public schools than pretty much anyone else, by almost any measure you choose: http://www.huffingtonpost.com/2013/06/25/oecd-education-repo....
Chicago spends about the same per student as the wealthy suburb where I grew up. Teachers' salaries are also higher in Chicago than where I grew up, despite having comparable if not lower cost of living.
Half of all children in Chicago live in single-parent homes, and the poverty rate for families headed by a single mother are around 40%: http://articles.chicagotribune.com/2013-11-03/news/ct-edit-x.... That percentage is almost certainly much higher within the Chicago public school system, which skews poorer than the city as a whole.
You can't have functional schools with students who have to worry more about gang violence in their neighborhood than their grades. More money, better curricula, better teachers, none of that matters when the social structure outside the school is crumbling.
I imagine much of the dysfunction endemic to urban school systems is related to the Sisyphean nature of the task we give them. We focus almost single-mindedly on schools and education to solve all the ills of urban society. If only our schools were functional and our teachers competent, we say, these kids would be able to rise out of their condition and better themselves. But teachers who teach in these systems realize the absurdity of that charge. Stuck between a student population that needs help in ways that are not within their power to give, and a society that fails to realize the actual needs of these young people, teachers shift to a mode focused on self-preservation. Hence the dysfunction present in urban school systems.
Of course the new mayor of New York wants to stop their growth, but that's another complaint.
[1] http://www.successacademies.org/
[2] http://projects.nytimes.com/new-york-schools-test-scores/cou...
and
http://perdidostreetschool.blogspot.com/2013/08/bronx-succes...
Teacher salaries are often higher in urban districts than suburban ones (and often still result in teacher shortages) because the conditions in urban districts are often less attractive to qualified teachers. (Policies which equalize per student spending across states means this also results in lower non-teacher-salary spending in those districts, which results in worse working conditions and more pressure for teachers to reach into their own pockets to meet classroom needs, which makes teaching in those districts even less attractive.)
My sense is there are a few reasons for this:
1) It is very hard to measure the benefits of good teaching, and it tends to be on a much longer timeframe. (Do we value nice teachers, or the ones that force us to learn? And is there a good test for it? I still get lessons from my great high school teachers, but it took a long time to appreciate some of them.)
2) There is little competition for skills of high school teachers. In large cities, there is one employer, take it or leave it. Teachers get one salary - based on how long they've been there. By the time they're most beaten down and least effective, they're paid the most. Meanwhile, the academic literature suggests that after 5 years, there is little correlation between time in classroom and improved skills.
What can be done? Improve the imperfect measurements, and open up the system for competition. Make it easier for schools to compete for better teachers by paying them more. (In the end, parents know which teachers are better) Make it easier for parents to leave bad schools. Make it easier to close schools whose parents leave.
Public schools get plenty of money. It just doesn't go to the classroom teachers.
As a psychologist friend pointed out to me yesterday, jobs that are traditionally held by women tend to be underpaid compared to "equivalent" jobs that are traditionally male. Compare teacher salaries to police and fireman salaries in the same district.
That said, most of the (invariably obscenely overcompensated) superintendents I've heard of have been male.
It's not like I'm working harder or what I'm doing is more important. You just can't convince Americans to spend money on public schools.