Why are professors “poorly paid”?

As an aside, his analysis of American Time Use Survey diary data demonstrates those lazy professors work almost 50 hours a week during the academic year, and 40 hours a week during the summer (see his Table 2).

I have documented a large pay disadvantage of academics behind
otherwise identical doctorate-holders. Part of this disadvantage is a
compensating pay differential arising from the more equal distribution
of work time across days of the week among academics; but in a simple
model of utility this difference alone can account for no more than one-fourth
of the disadvantage. This small proportion is not surprising: The
expressed preferences of a sample of academics who, while mostly not
in elite schools, are nearly all in at least partly research-based institutions,
suggest that the advantages of work-timing in academe are far
from its most important attraction.

The unexplained part of the pay disadvantage in academe might be
accounted for by the job security offered by academic tenure. This is
consistent with evidence for young economists (Ehrenberg, Pieper, &
Willis, 1998) that entry salaries are lower where the chance of obtaining
a tenured position are greater. But with only seven percent of
respondents to the survey of social scientists stating that job security
matters to them; and with this characteristic being unrelated to experience
in academe, the role of job security does not seem likely to
explain much of the remaining pay differential between doctorate holders
in academe and those elsewhere.

The pay disadvantage might arise from self-selection due to differences
in preferences of those who choose this sector compared to others
(as in Goddeeris, 1988, examining attorneys’ choices, and Stern, 2004,
examining biologists’ choices). It may also result, even if workers’
preferences for job characteristics are homogeneous, from the greater
importance of other aspects of the academic life unrelated to work
timing. That respondents to the survey of social scientists noted that
such things as independence and thinking about new ideas, and
spending time with students, are more important to them than flexibility
of time use suggests that this might be the case. Indeed, isolating
the role of independence from being supervised by others in determining
compensation would be a useful avenue for future study of
academic labor markets and indeed of professional occupations more