Pulled Away or Pushed Out? Explaining the Decline of Teacher Aptitude in the United States

Introduction

It is only rational that there’s a link between a teacher’s value-added and her academic aptitude. However, for the United States, there has been a decline in aptitude since 1960.

Two main hypotheses for the decline in teacher aptitude:

1. Pull Hypothesis:

- Greater pay parity with men in nonteaching/alternative jobs have drawn women out of teaching

2. Push Hypothesis:

- Unionization may have compressed salary, benefits and nonmonetary returns to skills in teaching

The two hypotheses are not mutually exclusive, thus the goal is not to pick one but to distribute the blame.

The Decision to Go into Teaching

A.D. Roy’s model of occupational choice can expound on the hypothesis of declining teacher aptitude. The model, given a positive correlation of skills across occupation, shows that reduction of a job’s incentives for aptitude will push out the high aptitudes workers. On the other hand an increase in the job’s pay across the board for aptitude will pull in workers from other occupations, but will not essentially alter the distribution of skills across jobs. For example, if there’s an increase in opportunity in an alternative (to teaching) job for female college graduates of all skills, though fewer will teach, it does not necessarily follow that the teacher’s aptitude would decline.

Push and Pull Factors and the Decision to Become a Teacher[1]

From 1963 to 2000, the earning of the average female teacher has risen by 8 percent in real terms. The lowest aptitude teachers’ earning to mean teacher earnings increased, while for the highest aptitude teachers declined, thus by 2000, for all aptitude groups (lowest to highest), had earnings ratio near 1.

Another result shows that the ratio of female to male earnings in nonteaching jobs in the top 3 aptitude group has increased, thus female graduates belonging to the top 3 expect the ratio of their earnings to similar-aptitude-males earnings to rise from 0.77 in 1963 to .86 in 2000. However for lower-aptitude woman, the change in the female to male earnings ratio is significant.

Another finding suggests that for most male graduates, their real earnings in nonteaching occupations increased by around 31 percent from 1963 to 2000, except for those belonging in the top 5 percentile whose earnings rose by about 42 percent.

Lastly, the results show that the share of all teachers in the lowest aptitude group increased from 16 to 36 percent, while the share of all teachers in the highest aptitude group fell from 5 to 1 percent.

Why Teacher Aptitude Declined

First, the study shows that higher the ratio of teacher earnings for the graduate’s aptitude group to mean teacher earnings, the greater the chance one was to teach. Recalling the result stated above, the increase in the ratio for the lowest aptitude group and decrease for the highest aptitude, thus the pay compression actually increased the share of the female graduates in the lowest aptitude, who became teachers, however decreased the share of the female graduates in the highest aptitude who pursued teaching. 

Second, the observed increase in the real earning of the average teacher approximately caused by unionization, in effect gave rise to the share of female college graduates who became teachers. However, these results cannot clarify much of the decline in teacher aptitude since women across all aptitudes experience the same increase in mean pay.

Next, there is a less likely chance for women to teach given a higher ratio of female-to-male earnings in nonteaching occupations. Thus improvements in pay parity decreased the share of female teachers belonging in the top three aptitude groups, however caused little or no changes for the three lower aptitude groups. Another observation suggests that teaching and nonteaching pay likewise influence the decision to teach. On the other hand, pay parity in nonteaching jobs accounts much less for the decline in teacher aptitude than does the compression of teacher’s pay. This is due to the fact that the disparity between high aptitude and low aptitude female’s experience is smaller for pay parity than for pay compression.

The study also indicates that factors correlated with male earning decrease the share of females who pursued teaching across all aptitude groups, but these factors help mostly in explaining the decline in teaching among the highest aptitude groups. In addition, by accounting for the group size (given that they vary) there would actually be a slight rise in teacher aptitude since the lower aptitude groups are larger.

Apportioning the Blame

The decline in the share of teachers belonging in the highest aptitude group can be explained 80 percent by pay compression, about 9 percent by pay parity, 1 percent by the change in the mean teacher earnings, and another 19 percent is explained by male earnings at face value (if it is considered, however it almost definitely exaggerate the underlying effect).

The increase in the share of teachers in the lowest aptitude group can be explained by about 25 percent by pay compression, 6 percent by pay parity, 2 percent by the change in mean teacher earnings, and if considered at face value, 8 percent by male earnings. The rest is explained by the increase in the size of the lowest aptitude category (the number of female graduates in this aptitude).

In the beginning of the study the prior notion is that pay parity plays a bigger role than pay compression, without recognizing the significance that pay parity altered similarly for college women across all aptitudes thus making its minor role predictable.

Gains in the salaries of nonteaching occupations to college women are relatively the same for the high and low aptitude group. However, in teaching, there are significant losses for high aptitude women.

Source:

Caroline M. Hoxby and Andrew Leigh, “Pulled Away or Pushed Out? Explaining the Decline of Teacher Aptitude in the United States”, American Economic Review, Vol. 94, No. 2, (May, 2004), pp. 236-240.

---

[1] Respondents are divided into six groups based on the mean SAT scores of their respective colleges: those from colleges with SAT scores in the top 5 percentiles, the next 10, the next 15, the next 20, the next 25 and the bottom 25 percentiles. The aptitude groups are finer at the top distribution since previous study suggests that the top quartile reflects disproportionately for the decrease in teacher aptitude.