Teacher Labor Markets and the Perils of Using Hedonics to Estimate Compensating Differentials in the Public Sector

---

August 2007
Dan Goldhaber, Kate Destler, Dan Player

Download Full Report (PDF: 469 K)

SFRP Working Paper 17

Policymakers and researchers alike have expressed concern about a teacher quality gap between schools with affluent student populations and the more disadvantaged ones. Teachers in high-poverty schools have, on average, fewer years of education and lower scores on credentialing exams than do teachers at low-poverty schools. There appears to be growing policy interest in using financial incentives to influence the distribution of teachers across schools.

A significant body of empirical research shows that compensation and working conditions both play important roles in individuals’ labor market decisions. One methodology for calculating compensating salary differentials (“combat pay”) is to use hedonic modeling, a statistical methodology that assigns dollar weights to the factors (both teacher-specific and location-specific) that determine individual teachers’ salaries. However, the hedonic approach may not generate consistent or accurate results in the public sector. The lack of competitiveness in the teacher labor market, salaries set on a district-wide level, and unmeasured differences in teacher quality may all obscure the importance of working conditions in determining a school’s desirability to teachers.

The private school market may offer a better context for estimating the wage premiums necessary to induce teachers to work in high-poverty schools. Adequate assessment of teacher quality remains a challenge in the private sector, but private schools are more likely than public schools to recognize and reward differences in teacher characteristics in a way that is consistent with the reward structure in the labor market as a whole, and thus, in theory, should better reflect teacher preferences for various school contexts.

This study uses teacher and school-level data from the NCES Schools and Staffing Survey combined with census-level information about community characteristics to build hedonic wage models for both public and private schools. The key variable of interest was the percentage of students qualifying for free/reduced-price lunch, but other student characteristics, such as academic ability and race, were also taken into account.

Both public and private sector schools appeared to compensate teachers for some working conditions, such as class size, contract hours, and community-level cost-of-living. Statistically significant differences between the two models, however, suggest that that public and private schools offer different economic returns to working with disadvantaged populations. Controlling for other student, school, and community characteristics, private schools with a larger proportion of poor students appeared to pay a wage premium; in comparison, public schools with similar economic demographics seemed to pay teachers less. In the case of a female teacher choosing between schools in the 25th and 75th percentiles of student income, the difference in “combat pay” between public and private schools is substantial—almost $2,500.

It is possible that private school teachers have preferences that are distinct from public school teachers. However, a more plausible explanation is that the salary setting processes in public schools tend to distort estimates of compensating differentials in this context. This in turn implies that hedonic wage models may not be appropriate for determining the wage differentials necessary to induce teachers to teach in disadvantaged schools.

Policymakers seeking to determine the “right” salary differential are thus in a bind. In our conclusion, we present four options: 1) use private school differentials as a basis for determining public school premiums; 2) estimate salary differentials by studying teacher mobility; 3) employ alternate estimation techniques; and 4) consider the direct effects of breaking apart the single district salary scale to allow schools to compensate for different working conditions. None of these options is simple, however. Each presents considerable political and logistical challenges for even the most reform-minded district.