How a once-promising merit pay system led Denver teachers to the brink of a strike

Jeff Buck was the very first teacher to sign up for Denver’s revolutionary pay-for-performance system in 2005. More than 13 years later, he is one of a dozen teachers sitting in a conference room at an obscure school district building, trying to unravel what the system has become.

ProComp — “professional compensation” — was supposed to recognize and reward good teachers. It was hailed around the nation as a model for getting districts away from traditional pay structures that prize longevity more than results, and, remarkably, it had the support of Denver’s teachers union.

But educators eventually grew disenchanted. Many said their paychecks went up and down in ways that were hard to predict or even understand. All the while, living in Denver became less affordable. Beyond those pocketbook concerns, many came to see ProComp as part of a series of education reform policies that, in their view, damaged the school system they were intended to save.

Now Denver Public Schools is on the brink of its first teacher strike in 25 years. ProComp sits at the center of the dispute between the district and the teachers union, and Buck is part of the union negotiating team arguing for Denver to pay its teachers “like every other district does.”

Educators in West Virginia, Oklahoma, Arizona, and Los Angeles have gone on strike for higher pay in the last year, but Denver may be the first teachers strike to turn primarily on the question of pay for performance, a key policy in the education reform toolbox.

How did a pay system that once seemed to hold so much promise bring teachers to their breaking point?

“What we got given is a system of Wall Street-style bonuses that mask the erosion of our salaries,” said Buck, a high school math teacher who went door-to-door in 2005 to sell voters on the tax that funds ProComp. “People have finally figured that out. It’s why people are angry.”

Denver Superintendent Susana Cordova has acknowledged that ProComp needs to change, and the district’s pay proposal puts more money into base pay, reduces the size of many incentives, and aims to make the remaining ones easier to understand. She has drawn a line in the sand, though, on $2,500 incentives for teachers at high-poverty and “highest-priority” schools.

Cordova frames this as a fundamental issue of equity, with the bonuses a necessary tool to keep good teachers in challenging schools where students need them the most. The research behind this assertion is decidedly mixed, but for ProComp’s remaining defenders, this point is key. The compensation system enables the district to express its values, with the end goal of raising student achievement for the students of color and those from low-income families who make up the majority of Denver’s 93,000 students but who lag behind their white and middle-class peers on test scores, graduation rates, and college readiness.

The origins

According to a book written by some of the architects of ProComp, the idea to revolutionize how Denver teachers are paid can be traced back to a report called “What Matters Most: Teaching for America’s Future.” Compiled by a now-defunct advocacy organization, the report found that recruiting and retaining quality teachers were the best ways to improve student learning.

The finding made an impression on the leaders of the Denver Classroom Teachers Association and on a former Denver teacher named Phil Gonring, who in the mid-1990s was working on education projects for a local philanthropy called the Rose Community Foundation. (The Rose Community Foundation is a financial supporter of Chalkbeat.)

With grant money from Rose and support from union leaders, Denver Public Schools started a ProComp pilot program in 1999 that eventually included 16 schools. The pilot paid teachers bonuses for meeting student performance goals.

DCTA lead negotiator Rob Gould, middle, addresses Denver Superintendent Susana Cordova, not pictured, during continued negotiations between teachers and administrators.

Other districts around the country had experimented with merit pay, too. But Denver’s pilot stood out for having the support of the teachers union, if not the support of every teacher. Most unions vehemently oppose merit pay, favoring a “steps and lanes” structure that rewards experience and gives teachers raises for earning master’s degrees or doctorates.

That the Denver union was helping design a performance pay system made its leadership pariahs at meetings of the National Education Association, the country’s largest teachers union, a Denver school board member told the New York Times in 2005.

Margaret Bobb, a teacher at a ProComp pilot school, pushed her colleagues to sign on. Bobb did not love the idea of pay for performance. Rather, as a science teacher, she saw the pilot as a way to test whether it was viable. Plus, teachers got paid to do it.

“I told my faculty … ‘I’ll take that extra money and put it away as my strike fund, so when we vote on ProComp, we’ll vote it down and I’ll have money to support a strike,’” said Bobb, who retired from the district last year. “I was naive. I did not understand … that when you do a pilot in education, it’s not a science experiment to see if it’s going to work or not. It’s just a little test to see how you’re going to implement it.”

The pilot program caught the eye of other philanthropic organizations, including the local Daniels Fund and the national Broad Foundation. Eventually, a task force of district staff and union teachers, including Buck, began designing a system that could be rolled out to all district schools. In addition to bonuses for student performance, it promised salary increases for teachers who completed training and applied their new skills in the classroom, and incentives for things like working in a hard-to-fill position or a high-poverty school.

To fund it, ProComp backers asked Denver voters to approve a $25 million annual tax increase, adjusted annually for inflation. (The tax is expected to generate $33 million this year.) Before the election, they sought union support.

Young teachers were excited about the chance to raise their pay by earning bonuses. Veteran teachers were wary of a new system.

It took a sophisticated campaign run by political consultants and funded by philanthropies to turn meager union support — early polling showed only 19 percent of teachers were in favor — into a victory. In the end, 59 percent of union members voted in favor of ProComp. In November 2005, Denver voters approved the tax increase by a similar margin.

ProComp was optional for teachers back then. By Dec. 31 of that year, the press reported that 735 of Denver’s more than 4,000 educators had signed up.

A big change

The first iteration of ProComp allowed teachers to earn permanent salary increases for meeting certain criteria. A 2005 Denver Post article gives some examples: a $999 raise if a teacher’s students made substantial progress on state reading and math tests, a $666 raise for taking a training course, a $333 raise for meeting two objectives a year tied to student achievement.

But in 2008, the system underwent a significant change that many teachers blame on then-superintendent Michael Bennet, who is now a U.S. senator. The change increased the size of one-time bonuses, but diminished permanent raises for teachers who’d been with the district 14 years or more — a move that favored newer teachers over more veteran ones.

The union reluctantly agreed to the changes.

“This was the best deal that we could get for our members,” the teachers union president at the time told the Rocky Mountain News. “There were compromises on both sides.”

The limit on salary building turned out to be a significant blow to ProComp’s reputation with teachers. Another qualm teachers have with the system is that the size of the bonuses can fluctuate from year to year, making their paychecks unpredictable.

For example, the bonus given to teachers at “top performing” schools with high test scores, and “high growth” schools where students showed progress, shrunk this year from $2,400 to $1,000.

The reason why illustrates another issue with ProComp. In the early days, the tax generated more money than the district spent, which built up a fund balance. But as the district grew and more teachers joined ProComp, the opposite became true. To get ProComp back in the black, district officials said they and the union agreed to shrink the size of the bonus for teachers at “top performing” and “high growth” schools.

Denver Superintendent Susana Cordova listens to a packed room of teachers during a brief resumption of negotiations between the district and the teachers union. (Helen H. Richardson/The Denver Post)

The criteria for earning incentives can also be frustrating for teachers. If a school’s demographics or rating change, even by a small amount, teachers can lose that money — something union leaders say is driving educators out of Denver Public Schools.

“Teachers don’t want money and incentives they could lose at any moment,” Jess Schneider, an English teacher at the district-run Noel Community Arts School, said at a union rally in December. “We want reliable, predictable, and sustainable pay.”

Special education teacher Maren Jorgensen told Chalkbeat last year that the loss of certain bonuses and incentives meant she was making less money in her eighth year of teaching than she had in her first. She said she planned to leave Denver Public Schools for a teaching job in Minnesota, where she’d be making $22,000 more per year.

The average Denver teacher salary this year is $62,095, according to the district. That includes base salary, ProComp incentives, and other stipends.

Not all teachers oppose the bonuses. Alison Corbett teaches at a high-poverty high school in far northeast Denver and earns several thousand dollars in bonuses each year, including a retention bonus for staying in her position. She said that while the extra money isn’t the reason she took the job, she agrees in principle that teachers who work at tougher schools, with students who are more likely to have experienced trauma, should be paid more.

“It feels like an acknowledgement, a nod of the head,” Corbett said.

Does ProComp work?

On its face, it seems as though doling out more money to good teachers who have harder jobs should attract and keep them in those roles. But University of Colorado education Professor Allison Atteberry, who has extensively studied ProComp, said it’s not that simple.

“There’s these, basically, economics-based theories about why this should help, and then there’s the reality of what happens on the ground — and they’re not really meshing,” she said.

In recent research, Atteberry found that student achievement increased more in Denver Public Schools than in neighboring districts since ProComp started — a phenomenon she said can’t be explained away by demographic shifts within the district or other factors.

Other research has established a link between effective teachers and higher student achievement. Atteberry found that under ProComp, Denver did a better job attracting and retaining teachers deemed effective — as measured by their students’ year-over-year academic growth — than other districts. That is evidence that ProComp worked by at least one metric.

But Atteberry could find no evidence that Denver teachers were more likely to work in high-poverty schools under the system. Her paper is still undergoing peer review.

Earlier research by Eleanor Fulbeck of the American Institutes for Research found ProComp bonuses were associated with fewer teachers leaving the district. But the effect was more modest for high-poverty schools — a 7 percent reduction in turnover compared with a 30 percent reduction for the district as a whole — which, she wrote, suggests the extra money does not offset “more substantial issues,” such as the working climate in those schools.

That is significant because Denver Public Schools officials have argued that the high-poverty incentive is critical to closing test score gaps between more and less privileged students, and to keeping good teachers in challenging schools. Yet research finds less impact for those teachers.

Sharyn Kline, a 2nd grade teacher at Park Hill Elementary school wears pins on her lanyard as she sits in a packed room to listen to continued negotiations between DPS teachers and administrators. (Helen H. Richardson/The Denver Post)

More recently, the district has pointed to findings that retention is up 4 percent at schools where teachers receive a “highest-priority incentive,” and that principals feel the bonuses are important recruiting tools. Research in other states shows pay incentives do improve retention, but that modest incentives produce only modest improvements.

Atteberry noted that although studies show ProComp has had some positive effects, it doesn’t mean teachers’ frustration with their pay is unwarranted.

“Somehow all of that money has left teachers feeling undervalued instead of more valued,” Atteberry said of the bonuses and incentives. “It’s a shame.”

A proposed shift

Both the district and the union are now proposing shifting the way ProComp works, while still honoring the intent of the tax increase voters approved in 2005. The limits on salary-building for teachers with more than 14 years of experience would disappear, and the size and number of bonuses would shrink to put more money into teachers’ base pay.

Both sides want to go back to a traditional “steps and lanes” system of raises based on teacher longevity and education. Currently, the district uses “steps and lanes” only to set teachers’ starting salary, with raises dependent on completing certain training and other factors.

Cordova, who became superintendent last month after 30 years as a teacher, principal, and administrator in Denver, has acknowledged that ProComp needs to change.

“I think ProComp was incredibly innovative and groundbreaking at the time — and it’s complicated, it’s hard to follow, and I think it’s hard for teachers to understand what their salary is going to be like,” she said at a forum in December.

But while the district and the union agree on the concept, they differ sharply on the details. The union’s proposal has more “lanes” than the district’s proposal, which means teachers would have more opportunities for raises. They’d also have more ways to earn those raises, because the union wants training courses to count toward lane changes.

The district’s proposal reserves more money for incentives, including a $2,500 retention bonus for teachers who return to work at one of 30 schools deemed “highest-priority.” The union wants to spend that money to increase the base salaries of all teachers.

An estimate released Monday by the Colorado Office of State Planning and Budgeting found the union’s proposal would cost about $8 million more than the district’s in the first year the new system would be in effect. But the labor dispute is about more than money or even philosophy. It’s rooted in a lack of trust that teachers say stems from more than a decade of school reform experiments, of which ProComp is an especially tangible example.

“They keep moving the cheese,” said Bobb, the recently retired teacher who participated in the ProComp pilot. “The cheese gets moved every freaking year. That’s why teachers have just reached this breaking point of, ‘Quit yanking us around.’”

Erica Meltzer contributed to this report.