Self-selection in the market of teachers

Abstract

Public school teachers are usually paid according to centralized earning schedules, in which their income depends mainly on experience. By contrast, in private schools, there is high wage dispersion, and salaries correspond mainly to teachers’ performance. That dichotomous labour regulation encourages teachers with better unobservable skills to self-select into private schools because the likelihood of earning higher wages is higher than in public schools. The other side of the coin is the self-selection of ‘bad’ teachers into public schools. Using a representative sample of Chilean teachers, we estimate a two-sector Roy model to test self-selection. We find evidence of negative self-selection of teachers into public schools.

Keywords:

• education
• human capital
• self-selection

JEL Classification:

• I21
• I28
• J24

Notes

¹ Ballou and Podgursky (2002) find that returns to seniority are relatively higher for public school teachers in the United States. They posit that this phenomenon cannot be explained by teachers’ human capital or the imperfect monitoring of their performance. They find that unions seek to reward senior teachers with ‘backloaded increases in pay’ and junior teachers with fewer steps in their salary schedules, allowing them ‘to reach high levels of pay more quickly’. Heutel (2009) shows that higher teacher salaries in unionized districts in the United States can be explained by a tournament model. He states that high administrator-to-teacher salary ratios lure teachers to compete to be promoted as administrators. Like Ballou and Podgursky (2002), Heutel (2009) also rejects the imperfect monitoring hypothesis as a cause of high seniority returns. He argues that ‘teachers are among the most difficult of all employees to fire’. The findings of Ballou and Podgursky (2002) and Heutel (2009) present different explanations consistent with the idea that tenure protections challenge public schools to use alternative models to fire shirking teachers, rather than traditional models of imperfect monitoring.

² McEwan (2001) finds that there is no consistent difference between student achievement in public and nonreligious private voucher schools, although fee-paying private schools and Catholic private voucher schools have higher achievement levels than public schools. Mizala and Romaguera (2001) and Sapelli and Vial (2002, 2005) find that students who attended private voucher schools have higher educational outcomes than those from public schools. More recently, Anand et al. (2009) find that students in fee-charging private voucher schools score higher than students in public schools. However, they find no difference in the academic achievement of students in fee-charging private voucher schools relative to their counterparts in free private voucher schools. Their results suggest that low-income students who typically attend public schools can benefit from attending private voucher schools. Bravo et al. (2010) use the 2002 and 2004 Social Protection Surveys (Encuesta de Protección Social) to estimate a dynamic model of schooling and working decisions. They conclude that the school voucher program induced individuals affected by the program to attend private voucher schools at a higher rate, to achieve higher educational attainment, to participate more in the labour force and to earn higher wages. Finally, Lara et al. (2011) use a number of propensity-score-based econometric techniques and changes-in-changes estimation methods and find that private voucher education leads to small differences in academic performance.

³ Among those schools, there exist free semiprivate schools that do not charge co-payments to parents as well as fee-charging semiprivate schools.

⁴ Public, semiprivate and private schools represented 40.71%, 50.73% and 8.56%, respectively, of the total enrollment in primary and secondary education in 2010.

⁵ The maximum is set at 100% of the BNMS for a teacher with more than 30 years of experience.

⁶ PSU stands for Prueba de Selección Universitaria.

⁷ For instance, teachers who work in public and semiprivate schools can receive special monetary benefits according to their performance in government programs designed to measure teachers’ achievement (such as Asignación por Excelencia Docente and Asignación por Excelencia Pedagógica).