Compensation in the Nonprot Sector
Christopher J. Ruhm Carey Borkoski
a b s t r a c t
We investigate the determinants of pay in the nonpro t sector using data for 25– 55 year olds from the 1994– 88 Current Population Survey Out-
going Rotation Groups. Our results are consistent with the hypothesis that compensation is primarily determined in competitive markets without
‘‘labor donations’ ’ to nonpro t employers. One implication is that non- prot workers receive virtually the same wages as observationally equiva-
lent employees in similar positions with prot-seeking enterprises. We can- not rule out the possibility of nonpro t penalties or premiums for selected
groups; however, the differentials are generally small and competition ap- pears to play a dominant role in nonpro t wage setting.
I. Introduction
Nonprot enterprises are an increasingly important part of the Ameri- can economy. The number of nonprot associations grew 54 percent between 1980
and 1997 U.S. Census Bureau 1998, Table 1286 and the fraction of GDP accounted for by them rose from 2.9 to 4.3 percent Bureau of Economic Analysis 1998. Non-
prots utilize the majority of volunteer labor and are responsible for a substantial proportion of paid employment in some industries. Despite this growing signicance,
Christopher J. Ruhm is a professor of economics at the University of North Carolina at Greensboro, Greensboro, NC 27402-6165 and a research associate at the National Bureau of Economic Research.
Carey Borkoski is a Ph.D. student in the policy science program at the University of Maryland, Balti- more County. The authors thank Peter Bearse, Richard Frank, H. E. Frech III, Dan Rosenbaum, Doug
Staiger, and Burton Weisbrod for helpful comments and William Black for research assistance. Ruhm gratefully acknowledges nancial support from the National Science Foundation SES-9876511 and
from an NBER faculty fellowship for the study of nonpro t institutions. The data used in this article can be obtained beginning April 2003 through March 2007 from Ruhm.
[Submitted August 2000; July 2002] ISSN 022-166X
Ó 2003 by the Board of Regents of the University of Wisconsin System T H E J O U R N A L O F H U M A N R E S O U R C E S X X X V I I I 4
compensation in the nonprot sector remains poorly understood. There is little ques- tion that nonprot workers earn less than observably similar employees of for-prot
rms.
1
However, the distribution of jobs and worker characteristics varies markedly, raising the possibility that the disparities reect compensating differentials or indi-
vidual heterogeneity not accounted for in standard earnings regressions. We address these issues through a detailed analysis of the determinants of pay in
the nonprot sector. Our goal is to ascertain how the earnings of individuals em- ployed by nonprot enterprises compare to those of identical workers in similar
jobs with prot-seeking rms.
2
We use several complementary approaches including: analyzing the size and pattern of the cross-sectional wage differentials with and
without controls for job characteristics, estimating how earnings change when work- ers shift between nonprot and for-prot jobs, and examining the disparities in wage
levels and growth rates for workers in narrowly dened industries or occupations with a substantial mix of nonprot and for-prot employment.
Our results generally support the hypothesis that nonprot workers are paid in competitive labor markets and do not ‘‘donate’’ labor to their employers by accepting
lower wages. What this means is that, after controlling for limited set of job charac- teristics, persons in nonprots earn approximately the same amount as if they were
employed in equivalent positions with prot-seeking rms. This is true even though the wages of nonprot employees average 11 percent less than those of their counter-
parts with similar observed attributes. The reason for the lower earnings is that non- prot jobs require fewer hours and are concentrated in a small number of industries
that offer relatively low pay but are probably also desirable places in which to work. Our evidence does not rule out the possibility of wage penalties or premiums for
selected groups. However, the magnitudes of the differentials are generally small and do not detract from the dominant role that competition appears to play in setting
nonprot wages.
II. Relative Earnings in the Nonprot Sector