Most public-school teachers' salaries are determined by years in the classroom and degrees held. But a new study from the Manhattan Institute shows that the premium we pay for teacher experience is far greater than is typically acknowledged.

In "Rewarding Experienced Teachers: How Much Do Schools Really Pay?", authors Josh McGee and Marcus Winters look at the nation's 10 largest school districts and find that the experience premium is understated because it doesn't take into account pension benefits, which are back-loaded. They also propose a way to better align compensation to teacher performance.

Research shows that teachers tend to steadily improve professionally at first, then plateau after five to seven years. But pension benefits are accrued on a very different schedule.

The authors base their calculations on a teacher who begins working in a school district at the age of 25 and holds a master's degree. In New York City, for example, teachers earn about $1,031 in overall retirement benefits during each of their first 15 years. That spikes to around $16,908 annually for the next 15 years.

The difference in pension compensation for a teacher with 35 years of experience, compared to a novice, is even greater than the salary differential. In year 35, a New York City teacher earns $42,729 more than a beginning teacher in annual take-home salary but in that one year alone also accumulates $45,419 more in projected future pension benefits.

Philadelphia teachers' pension benefits spike in their 25th and 35th years. In year 24, the teacher would accrue $7,716 in projected future benefits; in year 25, the number skyrockets to $102,975.

The authors recommend moving to a "cash-balance" plan, which guarantees a certain rate of return on both the employer and employee's pension contributions that would later be paid out in the form of an annuity. Overall retirement compensation would remain the same, but benefits would accrue as a constant percentage of salary throughout a teacher's career.

The approach has a couple of major advantages. The first is fairness. Instead of the bulk of pension benefits going to the one-quarter or fewer of teachers who spend all or most of their careers working in the same school district, benefits would be more evenly distributed across experience levels.

The plan would compensate those who stay for the initial five-to-seven-year period during which teacher quality tends to improve significantly while still treating career teachers fairly. It also would eliminate the incentive for those who might like to do something else for a living to stay long after their effectiveness has peaked.

A cash-balance plan doesn't directly link compensation to performance. But it does represent a way to reward teachers at the point in their career when they are most likely to be improving without the political trench warfare that surrounds issues like merit pay.