Teacher layoffs: An empirical illustration of seniority v. measures of effectiveness
School districts are confronting difficult choices in the aftermath of the financial crisis. In prior recessions, districts often muddled through by imposing a combination of tax increases and expenditure cuts that avoided involuntary personnel reductions. Today, the financial imbalance in many school districts is so large that there is no alternative to teacher layoffs. In nearly all school districts, layoffs are currently determined by some version of teacher seniority. Yet, alternative approaches to personnel reductions may substantially reduce the harm to students from staff reductions relative to layoffs based on seniority. First, because salaries of novice teachers are often much lower than those of veteran teachers, seniority-based layoffs lead to more teachers being laid off to meet any given budget deficit, with the associated implications for class size. Second, because teachers vary substantially in their effectiveness, staff-reduction polices that do not consider effectiveness likely will allow some ineffective teachers to continue teaching while some more effective teachers lose their jobs. Third, because many districts have redesigned human resource policies to place greater emphasis on the recruitment and retention of effective teachers, they may have hired disproportionately more effective teachers over the last several years than in prior years. In such cases, seniority-based layoffs will be even more detrimental for quality. Finally, if, as in many districts, novice teachers are concentrated in schools serving low-achieving students and students in poverty, then a seniority-based layoff approach will disproportionately affect the students in those schools. As a result, many school district leaders and other policymakers are raising important questions about whether other criteria, such as measures of teacher effectiveness, should inform layoffs.
This policy brief, a quick look at some aspects of the debate, illustrates the differences in New York City public schools that would result when layoffs are determined by seniority in comparison to a measure of teacher effectiveness. Due to data limitations and an interest in simplicity, our analysis employs the value-added of teachers using the 4th and 5th grade math and ELA achievement of their students. Unsurprisingly, we find that layoffs determined by a measure of teacher effectiveness result in a more effective workforce than would be the case with seniority-based layoffs. However, we were surprised by facets of the empirical results. First, assuming readily available measures of teacher effectiveness actually measure true teacher effectiveness, an assumption to which we return below, the differences between seniority and effectiveness-based layoffs are larger and more persistent than we anticipated. Second, even though seniority-based layoffs imply laying off more teachers, the differential effect on class size is very small in our simulations, though it would be larger for larger budget reductions. Third, there is somewhat greater school-level concentration of layoffs in a seniority-based system, though with a few notable exceptions, both methods result in fairly dispersed layoffs, with the vast majority of schools having no more than one layoff in grades four and five combined.
So where does this leave us? As a result of the limited applicability of teacher value-added measures to the full population of teachers as well as concerns about potential mis-measurement of effectiveness associated with using value-added measures even when available, neither seniority nor measures of value-added to student achievement should be the sole criterion determining layoffs. However, ignoring effectiveness measures completely as seniority-based systems do, is also problematic. Instead, the use of multiple measures of effectiveness for layoff decisions holds promise for softening the detrimental effect of layoffs.