Seniority-Based Layoffs, Can't We Do Better?
Note: Dan Goldhaber, an economist and professor at the University of Washington, is guest-posting this week.
Today I'm joined by my co-author, Roddy Theobald, and we are going to do a little self-promotion in the context of talking about teacher layoffs.
Few recent issues in education have garnered more mainstream calls for reform than teacher layoffs. This is partly a matter of timing. The recent economic crisis has forced the education community to revisit an issue it was largely able to ignore for the better part of three decades. But layoffs also bring home to the public the consequences of seniority-provisions in a very tangible way. It's personal when a favorite teacher loses a job. Seniority-driven layoffs--the system that exists in most districts--also strike a nerve as they exemplify for many observers the prioritization of "what's good for adults" over "what's good for students." Calls to reform the seniority-based system are becoming commonplace in the mainstream press, national centers, and even class-action lawsuits.
Seniority as a factor in determining layoffs is, of course, not unique to public education. As a system it has some virtues, being both transparent to employees and easy to administer. But the calls for reform focus on four main problems with the seniority-driven layoff system. First, since less-senior teachers make less money, the seniority-based system necessitates more layoffs to achieve the same budgetary savings. Second, the seniority-based system is by definition quality-blind, so districts are forced to let go of some promising young teachers before at least some less effective teachers who have greater seniority. Third, since less-senior teachers are more likely to teach high-demand subjects like math and special education, a strict seniority-based system would results in more layoffs in the very areas that are hardest to staff. Finally, since less-senior teachers within many districts are disproportionately in low-income, high-minority schools, the seniority-based system results in an inequitable distribution of layoffs across a district.
These four issues, however, come with many questions, and these questions are crucial for administrators weighing the costs and benefits of taking on the seniority-driven layoff system. Just how much money could we save with a seniority-blind system? What are the real consequences of our current system to student achievement? What are the actual distributional consequences of the current system? And if we moved to a layoff system based on teacher effectiveness, how much would we gain?
We investigate these questions in a recent study using data from Washington State on which teachers received layoff notices (most of these teachers were ultimately not laid off as their jobs were protected by stimulus funds, but we believe this is unlikely to play out this way in the future). This allows us both to investigate the determinants of teacher layoffs (e.g., what teacher and school characteristics predict which teachers received layoff notices) and the implications of these layoffs. (Our study builds off of the fine work by researchers Donald Boyd, Hamilton Lankford, Susanna Loeb, and James Wyckoff who simulate the consequences of seniority-based layoffs using data from New York City.)
Our findings on the determinants of teacher layoffs are unlikely to surprise anyone familiar with the layoff provisions existing in many collective bargaining agreements. Seniority is the primary determinant of layoff decisions; all else equal, a first-year teacher is over twice as likely to receive a reduction-in-force (RIF) notice than a teacher in her 4th-6th years in the district, and teachers with more than 6 years of seniority have close to a zero probability of receiving a layoff notice. Subject area is also predictive of which teachers are targeted for layoffs, but the protections that appear to accrue to those teachers holding an endorsement in a "high-needs area" (math, science, or special education) are dwarfed by the impact of seniority. Finally, we find no evidence that teacher effectiveness is considered in determining which teachers are placed on the chopping block.
We go on to simulate an alternative system where layoffs are driven by value-added estimates of teacher effectiveness. Here our findings mirror many of the results of the Boyd, et al. paper. First, we find effectiveness-based layoffs would result in a very different population of laid off teachers: the overlap in teachers under the two different systems is only 16 percent. And, an effectiveness-driven versus seniority-based system would result in at least 10 percent fewer teacher job losses.
We also estimate pretty profound effects on student achievement. Student achievement in effected classrooms is estimated to be on the order of magnitude of 2.5 to 3.5 months of student learning better under a system that considers teacher effectiveness than one that is driven by seniority. Lastly, we find that seniority-driven layoffs have a disproportionate effect on various student sub-groups. Because of the distribution of teacher seniority, African-American students are far more likely to be in a classroom with a teacher who receives a layoff notice than are white students.
There are doubtlessly good reasons to consider seniority in layoff decisions, but it's nearly impossible to make the argument that relying strictly on seniority to drive decisions is in the best interests of student achievement.