THE ROLE OF EMPLOYER AND WORKPLACE SIZE IN THE US FEDERAL SECTOR JOB QUEUE†
| DOI | http://doi.org/10.1111/j.1468-0084.1994.mp56002004.x |
| Published date | 01 May 1994 |
| Author | Madhu S. Mohanty,John S. Heywood |
| Date | 01 May 1994 |
OXFORD BULLETIN OF ECONOMICS AND STATISTICS. 56,2(1994)
0305-9049
THE ROLE OF EMPLOYER AND WORKPLACE
SIZE IN THE US FEDERAL SECTOR JOB
QUEUEt
John S. Heywood and Madhu S. Mohanty
I. INTRODUCTION
Following Poirier's (1980) pioneering work on partial observability in
hivariate probit models, several applications have been made to the labour
market. Abowd and Farber (1982) used the techniques to find evidence of a
queue for US union sector jobs as did Farber (1983) with a different specifi-
cation. Venti (1987), Heywood and Mohanty (1990) and Mohanty (1992)
have extended the examination to the US federal government confirming a
queue of workers for federal sector jobs as well. The existence of a queue of
workers waiting for federal jobs fits with a long line of research demonstrat-
ing that US federal workers experience wage premiums which cannot be fully
explained by differences in human capital and other personal characteristics
(Ehrenberg and Schwarz, 1986). Krueger (1988) reports that the number of
job applications per federal worker hired rises as the federal-private dif-
ferential increases. Smith (1976, 1977), Quinn (1979) and Moulton (1990)
report that both male and female workers in the federal government appear
over-compensated relative to their otherwise equal private sector counter-
parts. This, combined with generous federal sector fringe benefits (Heywood,
1991), presents a logical explanation for the presence of workers queuing for
federal jobs.
Typical examinations of the federal wage premium make no effort to hold
constant the size of the employer and the size of the workplace. These
variables are known to be important wage determinants within the private
sector. At least as early as Lester (1967) we have known that workers in
larger plants (workplaces) receive wages that are higher. More recently,
Brown and Medoff (1989) found that in each of the more than one dozen
studies they reviewed plant size and/or firm size emerged as a significant
determinant of wages despite long lists of control variables. Indeed, their own
work confirms the role of both plant and firm size as determinants of higher
t The authors thank Dale Belman, Robert Drago, the participants in the UWM applied
economics seminar and the reviewer for helpful suggestions.
© Basil Blackwell Ltd. 1994. Published by Blackwell Publishers, 108 Cowley Road, Oxford 0X4 uF,
UK & 238 Main Street, Cambridge, MA 02142, USA.
172 BULLETIN
wages. Findings of Idson and Feaster (1990), Mellow (1982) and Pearce
(1990) confirm that the employees of a large firm enjoy positive wage
premiums over their smaller firm counterparts.' Similar evidence from Long
and Link (1983) and Heywood (1988) suggests that firm and plant size may
also increase the prevalence and generosity of fringe benefits.
Findings on private sector size effects should be incorporated into
estimates of the federal wage premium because we know the federal sector
evidences no such size effects. Belman and Heywood (1989) demonstrate
that there is no premium associated with larger workplaces within the federal
government and, obviously, as a single employer there also exist no employer
size premiums. Thus, as a statistical matter, if there exist variables which
positively influence private sector wages but do not influence federal wages,
their exclusion will bias upward the true federal differential. This bias is even
more pronounced as most federal jobs are located in large establishments
(see Moore and Raisan, 1991). Indeed, Belman and Heywood (1990) have
shown that after inclusion of a full set of size variables in the typical Oaxaca
decomposition the federal premium drops to 2 or 3 percent, less than one-
quarter of its size without those variables, and loses statistical significance.
Such previous work on size effects sets the stage for our current investiga-
tion. The existence of a queue for federal sector jobs should be conditioned
on the possibility that workplace and employer size are characteristics which
workers consider when choosing a job. Similarly, any model of the hiring
decision should recognize that different employee characteristics may be
required for workplaces and employers of different sizes. The first point is
well grounded in the theory of hedonic wage differences (See Dunn, 1984).
High wages may be required to compensate for unpleasant characteristics
associated with large plants and employers. The second point has received
recent empirical support from Evans and Leighton (1988) who find that the
vast majority of the employer size effect in the private sector reflects worker
heterogeneity that the typical variables in a modern labour economics data
set caimot directly measure. They suggest that this results because larger
employers seek particular worker characteristics which they find more
valuable than would a small employer. Hence, employer size, although
considered job-descriptive by some authors (Linneman and Wachter, 1990),
may proxy valuable omitted worker characteristics in wage equations.
The federal government is clearly a large employer and is also comprised
disproportionately of large workplaces.2 Workers seeking a large workplace
or employer may thus seek a federal sector job. If large private sector
workplaces have characteristics which must be compensated for with higher
wages, it seems logical that the federal government must also offer such wages
to compete in the labour market. For example, higher private sector wages
Schmidt and Zimmermann (1991) find similar evidence in the context of German govern-
ment employment.
See Belman and Heywood (1990) and Moore and Raisan (1991) for confirmation of this
point.
© Basil Blackweil Ltd. 1994.
To continue reading
Request your trial