Has education reform finally come?

This extensive piece in the NY Times Magazine discusses the federal Race to the Top education funding that provides cash in exchange for schools coupling teacher compensation and tenure with student performance. 

My gut reaction is that this is a good thing, that there should be an element of accountability for performance for teachers.  But at the same time, there are many factors out of a teacher's control. 

The article does discuss a fascinating anecdote, where a charter school and a public school in Harlem share the same building, but the charter school's very different style (and performance-linked pay for teachers) is getting much better results:

A building on 118th Street is one reason that the parents who are Perkins’s constituents know that charters can work. On one side there’s the Harlem Success Academy, a kindergarten-through-fourth-grade charter with 508 students. On the other side, there’s a regular public school, P.S. 149, with 438 pre-K to 8th-grade students. They are separated only by a fire door in the middle; they share a gym and cafeteria.

On the charter side, the children are quiet, dressed in uniforms, hard at work — and typically performing at or above grade level. Their progress in a variety of areas is tracked every six weeks, and teachers are held accountable for it. They are paid about 5 to 10 percent more than union teachers with their levels of experience. The teachers work longer than those represented by the union: school starts at 7:45 a.m., ends at 4:30 to 5:30 and begins in August. The teachers have three periods for lesson preparation, and they must be available by cellphone (supplied by the school) for parent consultations, as must the principal. They are reimbursed for taking a car service home if they stay late into the evening to work with students. There are special instruction sessions on Saturday mornings. The assumption that every child will succeed is so ingrained that (in a flourish borrowed from the Knowledge Is Power Program, or KIPP, a national charter network) each classroom is labeled with the college name of its teacher and the year these children are expected to graduate (as in “Yale 2026” for one kindergarten class I recently visited). The charter side of the building spends $18,378 per student per year...

Although the city’s records on spending per student generally and in any particular school are difficult to pin down because of all of the accounting intricacies, the best estimate is that it costs at least $19,358 per year to educate each student on the public side of the building, or $980 more than on the charter side.

But while the public side spends more, it produces less...

To take one representative example, 51 percent of the third-grade students in the public school last year were reading at grade level, 49 percent were reading below grade level and none were reading above. In the charter, 72 percent were at grade level, 5 percent were reading below level and 23 percent were reading above level. In math, the charter third graders tied for top performing school in the state, surpassing such high-end public school districts as Scarsdale.
Same building. Same community. Sometimes even the same parents. And the classrooms have almost exactly the same number of students...

Dan Goldhaber, an education researcher at the University of Washington, reported, “The effect of increases in teacher quality swamps the impact of any other educational investment, such as reductions in class size.”

Is tying teacher compensation to performance the answer?  Or does it require a much more comprehensive effort (note the much longer hours of the charter school teachers, among other requirements)?  In reality, the charter teachers probably get paid less per hour.  It'd be interesting to know if their job satisfaction is better...