Teacher Exit and Educational Opportunity: Lessons From Career and Technical Education

Ensuring a stable pool of teachers is critical to building a pipeline of future workers, especially in career and technical education (CTE), where programming can lead to immediate post-school employment or postsecondary enrollment. We use longitudinal state data with unemployment insurance records to document workforce dynamics among CTE teachers. We find that teachers in hard-to-staff CTE areas are more likely to leave teaching and are difficult to replace, creating net reductions in the number of students who can be served. We demonstrate that teachers with the greatest likelihood of leaving are also those who earn the most money in their post-teaching employment, suggesting actionable dimensions for policies to help retain or recruit teachers in these areas.Ensuring a steady and stable pool of high-quality teachers is critical to building a pipeline of future workers and professionals to support economic stability, especially in growing and high-wage fields. This is particularly true in career and technical education (CTE) where programming in high school can lead to immediate postschool employment or enrollment in aligned postsecondary programs. Increased policy focus on CTE has expanded program offerings over the last 15 years, and new federal requirements that programs align with in-demand occupations and industries have heightened potential tensions between the private and public employment opportunities for current or potential CTE teachers (Perkins V, 2018). Even before the new federal mandate, CTE teachers were a hard-to-staff group, with school administrators citing higher turnover and difficulty hiring (Hensley et al., 2017). Though the quantitative research on CTE teachers is limited, evidence does suggest that teachers who score higher on content-specific certification exams produce graduates who earn more money in their early years of post-high school employment, bolstering the claim that these teachers are important (Chen et al., 2023).