Content uploaded by Christopher Ruhm
Author content
All content in this area was uploaded by Christopher Ruhm on Apr 17, 2014
Content may be subject to copyright.
/ 3008 0003 Mp 175 Thursday Oct 19 01:33 PM LP–JEP 0003
Journal of Economic Perspectives—Volume 11, Number 3—Summer 1997—Pages 175–186
Policy Watch
The Family and Medical Leave Act
Christopher J. Ruhm
This feature contains short articles on topics that are currently on the agendas
of policymakers, thus illustrating the role of economic analysis in illuminating cur-
rent debates. Suggestions for future columns and comments on past ones should
be sent to C. Eugene Steuerle, c/o Journal of Economic Perspectives, The Urban Insti-
tute, 2100 M Street NW, Washington, D.C. 20037.
Introduction
President Clinton’s first legislative action upon taking office in February 1993
was to sign the Family and Medical Leave Act (FMLA). The act is designed to
‘‘support families in their efforts to strike a workable balance between the compet-
ing demands of the workplace and the home’’ (Commission on Family and Medical
Leave, 1996, p. xii). Most notably, the FMLA is the first federal law requiring some
U.S. employers to offer maternity leave to women with qualifying employment his-
tories. Prior to its enactment, the United States was virtually the only industrialized
country that did not guarantee job-protected parental leave (Kamerman, 1991).
During the recent presidential campaign, President Clinton proposed expand-
ing the Family and Medical Leave Act by guaranteeing workers’ rights to up to
24 hours per year off work to participate in their children’s school activities or to
accompany relatives to medical appointments or related professional services. In-
䊏 Christopher J. Ruhm is Professor of Economics, University of North Carolina at Greensboro,
Greensboro, North Carolina, and Research Associate, National Bureau of Economic Research,
Cambridge, Massachusetts. His e-mail address is ruhmcj@iago.uncg.edu.
176 Journal of Economic Perspectives
/ 3008 0003 Mp 176 Thursday Oct 19 01:33 PM LP–JEP 0003
dividuals would also have the option to earn flex-time for overtime work of up to
80 hours per year, to be used for any purpose including family leave. These exten-
sions were strongly opposed by Senator Dole, as was the original FMLA by President
Bush during his term of office.
This article summarizes provisions of the FMLA, considers its possible effects
on labor markets and examines resulting changes in the ability of workers to take
leave. I conclude that the actual provisions of the act are quite modest and have
neither yielded large benefits to workers nor imposed significant costs on employ-
ers. One reason for this is that relatively few workers gained significant new rights
to time off work as a result of the law. These conclusions should be viewed as
tentative given the brief period the FMLA has been in effect and the dearth of
previous research on related legislation.
Provisions and Coverage of the FMLA
The FMLA, administered by the Wage and Hour Division of the U.S. Depart-
ment of Labor, took effect on August 5, 1993.
1
Under the act, eligible employees
are entitled to 12 weeks of job-protected leave in a 12-month period to care for
newborn or adopted children, relatives with serious medical conditions, or their
own health problems. The legislation covers private establishments employing 50
or more persons within 75 miles of the worksite during at least 20 weeks of the
current or previous year. Government employers are generally covered regardless
of size. Individuals are eligible for FMLA leave if they have been with a covered
employer for at least 12 months and worked for the employer 1,250 or more hours
during that time. The employer is not required to pay wages during the job absence
but must continue health insurance benefits on the same terms as if the worker
had not taken leave. Employees may be required to use accrued sick leave or va-
cation time to cover some or all of the work absence and are entitled to return to
their old jobs or an equivalent position, except that employers may sometimes re-
fuse to reinstate ‘‘key’’ employees (salaried workers who are among the highest
paid 10 percent of employees).
2
Compared to other industrialized countries and some states, the parental leave
guarantees of the FMLA are limited. All western European nations currently offer
at least three months of paid maternity benefits, and some nations (such as Finland,
Sweden and Germany) provide a year or more of paid leave (Organization of Eco-
nomic Cooperation and Development, 1995; Ruhm, 1997). However, the U.S. law
is relatively broad in that it allows eligible workers time off to care for sick rel-
1
For workers covered by ongoing collective bargaining agreements, the FMLA took effect at the end of
the agreement or on February 5, 1994, whichever was earlier.
2
Further information on FMLA provisions can be obtained from U.S. Department of Labor, Employment
Standards Administration, Fact Sheet No. ESA 93-24.
Christopher J. Ruhm 177
/ 3008 0003 Mp 177 Thursday Oct 19 01:33 PM LP–JEP 0003
atives or for their own serious medical conditions. In some situations, the work
absences can be taken in several intervals of time, rather than as a single episode,
or by reducing normal work hours. Conversely, the scope of the law is limited
by the exclusion of small private employers (those with fewer than 50 workers
in a 75-mile radius) and individuals not meeting the job tenure or work hours
requirements.
Much of the information currently available on the implementation of the
FMLA comes from surveys of employers and employees conducted in 1995 by
Westat Inc. (WESTAT) and the University of Michigan Institute of Survey Research
(ISR) at the request of the Commission on Family and Medical Leave, which was
created with the enactment of the FMLA. The Employer Survey contains a stratified
probability sample of private sector business establishments in the United States.
The Employee Survey attempts to provide a national random sample of adults living
in the continental United States with some paid employment during the previous
18 months.
Using data from these surveys, the Commission on Family and Medical Leave
(1996) estimates that between 60 and 66 percent of employees work for covered
employers.
3
The employee data further indicate that 83 percent of persons at these
establishments worked at least 1,250 hours during the previous 12 months. Some
of these individuals will not meet the one-year job tenure requirement and so will
not be eligible for FMLA leave. Thus, a maximum of 55 percent (.66 1 .83 Å .55)
of employed individuals are covered under the FMLA, with a more likely estimate
being that around half of workers are eligible.
A smaller percentage of women are eligible for FMLA maternity leave.
4
Kler-
man and Leibowitz (1994a) estimate that 63 percent of new mothers over the age
of 19 are employed during the 12 months prior to the birth of their children. Of
this group, just 52 percent meet the FMLA job tenure and work hours requirements,
and 59 percent of the remainder work for employers with 50 or more employees.
Taken together, this implies that just 19 percent of new mothers (.63 1 .52 1 .59
Å .19) are eligible for FMLA leave, as are but 31 percent (.52 1 .59 Å .31) of those
employed for one year before childbirth. Since the act was designed to be restricted
in scope, this limited coverage is not surprising. Nonetheless, it is important to keep
in mind when discussing its likely consequences.
3
The 60 percent estimate is probably low because the WESTAT survey excluded government employers,
who are generally covered regardless of size. The 66 percent estimate is almost certainly high because
the ISR slightly oversampled high-income individuals and included as eligible for FMLA leave persons
who worked for covered employers during any of the preceding 18 months, whether or not they contin-
ued to do so at the time of the survey. Further information on the two surveys can be obtained from
Cantor et al. (1995) and McGonagle et al. (1995).
4
Women constituted 46 percent of all employees, but 58 percent of leave takers (Commission on Family
and Medical Leave, 1996). Other groups disproportionately using leave, including 25- to 34-year-olds,
African-Americans, Latinos, high school dropouts, hourly employees, workers with children and those
with low family incomes.
178 Journal of Economic Perspectives
/ 3008 0003 Mp 178 Thursday Oct 19 01:33 PM LP–JEP 0003
Consequences of the FMLA
In a competitive spot labor market with perfect information and no external-
ities, family leave mandates reduce economic efficiency by limiting the ability of
employers and workers to negotiate the optimal compensation package. Instead,
firms are required to allow job absences even when the costs exceed the benefits
of doing so. Since most of these expenses are likely to be passed on to workers, the
decrease in effective compensation results in a deadweight loss.
However, there are situations under which efficiency might be increased. If
large employers set compensation policies according to the preferences of the me-
dian worker, who may receive little benefit from rights to time off work, family and
medical leave may be underprovided to a substantial minority of workers. Similarly,
if individuals have better information than employers regarding the probability of
using leave, companies voluntarily providing the benefit will attract a dispropor-
tionate share of ‘‘high-risk’’ workers and so will be forced to pay lower wages. Per-
sons less likely to need time at home will shun these firms.
5
A government mandate
eliminates the incentive for this type of sorting behavior and has the potential to
increase efficiency. Finally, some workers may bargain for too little leave because
they have incomplete information regarding the advantages of staying home with
infants.
Klerman and Leibowitz (forthcoming) provide a model suggesting that the
FMLA may have an ambiguous impact on the amount of time new parents will be
away from work. Figure 1 helps to illustrate their argument. The key assumptions
are that the reservation wage (W
R
) decreases over time (as the infant gets older)
and that compensation on the old job (W
O
) is higher than that in alternative em-
ployment (W
A
). If unconstrained, the individual would return to work at time L
*
,
when W
O
Å W
R
. However, if the maximum leave allowed by the employer (L
O
)is
less than L
*
, the employee must choose between returning to the old firm at L
O
or
staying out of work until L
A
, but then switching jobs and receiving lower future
compensation. Now assume the government passes legislation requiring firms to
supply more leave than L
O
(but less than L
*
). Workers who would have previously
picked L
O
increase their leave to the amount guaranteed under the new law. Con-
versely, some persons who would have otherwise chosen L
A
find it worthwhile to
reduce their work absences to remain with their previous employers. This occurs
because the gap between the desired and available (with the old firm) amount of
leave decreases. If this effect dominates, new parents may work more rather than
less on average when leave is mandated.
Next consider how leave entitlements affect employment and wages. The man-
5
Rothschild and Stiglitz (1976) make analogous arguments for market failure in insurance markets.
Aghion and Hermalin (1990) show that, under some situations, socially optimal leave benefits might not
be voluntarily provided to any workers.
Policy Watch: The Family and Medical Leave Act 179
/ 3008 0003 Mp 179 Thursday Oct 19 01:33 PM LP–JEP 0003
Figure 1
Determinants of Leave Duration and Reemployment Decisions
date raises labor costs and so shifts the labor demand curve to the left (from D
1
to
D
2
in Figure 2, where C is the cost to employers of the leave guarantee). Since
employees receive at least some benefit from the entitlement, the supply curve
simultaneously shifts to the right (from S
1
to S
2
). Thus equilibrium wages (but not
necessarily total compensation) fall while employment may either increase or de-
crease, depending on the relative size of the shifts—the figure illustrates the case
where the expenses to employers exceed the benefits to workers.
There could be additional ‘‘dynamic’’ effects. For instance, if family leave raises
firm-specific human capital, by allowing workers to return to their old jobs, the
marginal product of labor will rise, causing the demand curve to shift to the right
(from D
2
to D
3
in Figure 2), attenuating the wage reduction or actually leading to
a rise in earnings. Indeed, proponents of leave mandates frequently use this rea-
soning to anticipate increases in both wages and employment for the groups most
likely to use leave.
6
Previous research examining the economic consequences of leave entitlements
is quite limited. The best available evidence pertains to parental leave benefits.
Studies examining the effects of maternity leave voluntarily supplied by employers
typically find that women receiving leave or returning to their old jobs are out of
work less time, receive higher wages and have steeper wage profiles (largely due to
preservation of job tenure) than those whose employers do not supply maternity
leave (Dalto, 1989; Spalter-Roth and Hartmann, 1990; Waldfogel, 1994, forthcom-
ing). However, the advantages may result from nonrandom selection into jobs of-
fering the benefits, rather than because of the leave itself, and efforts to purge the
effects of the unobserved heterogeneity have not been entirely successful.
6
Summers (1989) and Mitchell (1990) provide detailed discussions of the economics of mandated
benefits.
180 Journal of Economic Perspectives
/ 3008 0003 Mp 180 Thursday Oct 19 01:33 PM LP–JEP 0003
Figure 2
Consequences of Leave Mandates
Four recent studies provide evidence on legislated parental leave rights. These
investigations, briefly summarized in Table 1, suggest that the entitlements modestly
increase the employment of women while either having no effect on or (at dura-
tions exceeding three months) lead to slight reductions in their wages. These find-
ings, which should be viewed as preliminary since the results are sometimes sensitive
to the choice of models or specifications, are consistent with a relatively large out-
ward shift in the labor supply curve combined with a smaller ceteris paribus reduc-
tion in labor demand.
Neither theory nor the available empirical evidence provides unambiguous
predictions regarding the effects of the FMLA on economic efficiency or labor
market outcomes. However, based on the previous research and given that the leave
periods provided for in the act are quite short, it may be reasonable to anticipate
that they will have small positive effects on employment and little effect on the
wages of the groups most likely to use them.
There is scant reason to believe that these mild benefits impose large costs on
employers. Well over 90 percent of covered establishments responding to the
WESTAT Employer survey stated that the FMLA had no noticeable effect on busi-
ness performance or growth (Commission on Family and Medical Leave, 1996).
Furthermore, larger percentages of employers indicated positive rather than neg-
ative effects over a number of dimensions of employee performance. Thus, 13 per-
cent mentioned a positive effect on employee productivity versus 5 percent claiming
a negative impact. Beneficial effects were also more frequently spoken of with re-
gards to employee turnover (5 percent vs. 0 percent), career advancement (8 per-
Christopher J. Ruhm 181
/ 3008 0003 Mp 181 Thursday Oct 19 01:33 PM LP–JEP 0003
Table 1
Results of Studies Examining the Labor Market Effects of Parental Leave
Legislation
Author Type of Leave/Data Source Main Results Comments and Caveats
Klerman & Leibowitz
(forthcoming)
Examines state maternity
leave mandates using
Census data from 1980
and 1990.
The employment and
leave taking of
mothers increase
after the enactment
of state maternity
leave mandates.
Higher employment may
occur because states
with booming
economies are more
likely to enact
maternity leave
mandates. The
employment of
women with infants
sometimes increases
less than that of
mothers with slightly
older children.
Ruhm & Teague
(forthcoming)
Investigates national
parental leave laws in
17 OECD countries
over the 1969–1988
time period.
Parental leave durations
are positively
correlated with
employment-to-
population ratios,
labor force
participation rates
and, except at very
long durations, per
capital GDP.
The models may not
fully account for the
endogeneity of
parental leave. In
particular, the labor
market effects are not
concentrated among
women, who are most
likely to use the leave.
Ruhm (1997) Studies national parental
leave laws in 9
European nations over
the 1969–1993 period.
Durations of paid
parental leave are
positively related to
the relative EP ratios
of women but, except
at short durations,
are negatively related
to their earnings.
Men or older women
are used as a
comparison group,
under the assumption
that the effects of
parental leave are
concentrated among
women of
childbearing age.
Waldfogel (1996) Examines the FMLA using
data from the 1992–
1995 March Current
Population Surveys.
The FMLA has had no
impact on wages but
has modestly
increased the
employment of
women.
The estimated
employment effect is
sensitive to the choice
of models. Some
specifications indicate
negative or
inconsistent impacts
(e.g. the employment
of mothers with
infants grew faster,
after passage of the
FMLA, in states with
than in those without
maternity leave
mandates).
182 Journal of Economic Perspectives
/ 3008 0003 Mp 182 Thursday Oct 19 01:33 PM LP–JEP 0003
cent vs. 1 percent) and the ability of employees to care for their families (34 percent
vs. 0 percent).
7
Effect of the FMLA on Leave Taking
Enactment of the FMLA has led to substantial changes in formal employer
leave policies, but it has had a more modest effect on the actual use of family and
medical leave. Data from the WESTAT survey indicates that two-thirds of covered
establishments altered their leave policies to comply with the FMLA (Commission
on Family and Medical Leave, 1996). Among these employers, the most common
changes were allowing leaves for more reasons (76 percent), permitting males to
use leave (69 percent), extending the period of work absence (66 percent) and
guaranteeing job reinstatement (55 percent). Since the changes were dispropor-
tionately concentrated among relatively small (covered) employers, they affect a
smaller percentage of employees than they do establishments. Also, a large fraction
of individuals do not work at covered employers or, as discussed, do not meet the
work history requirements for eligibility. Finally, as shown below, firms modifying
policies to achieve technical compliance with the act often already provided mech-
anisms by which workers could take parental or medical leave.
Even prior to the FMLA, the Pregnancy Discrimination Act of 1978 (an amend-
ment to Title VII of the Civil Rights Act of 1964) required employers to treat disabil-
ity resulting from pregnancy or childbirth in the same manner as any other disability
(Trzcinski and Alpert, 1994). As a result, companies allowing leave for temporary
disabilities had to include those caused by pregnancy and childbirth. Data from the
U.S. Department of Labor’s Employee Benefits Surveys indicate that 86 percent of
full-time workers in private establishments with 100 or more employees had short-
term disability coverage in 1991, as did 96 percent of state and local government
employees in 1992 (Silverman et al., 1995). Hence, most women working for me-
dium or large employers could take some maternity leave (often paid) before the
act was passed. Short-term disability benefits were also reasonably prevalent among
private firms with fewer than 100 employees—64 percent of full-time workers with
these companies had coverage in 1992. Admittedly, employees of very small firms
and part-time workers were much less likely to have disability benefits, but these
individuals are also generally not eligible for FMLA leave.
Temporary disability insurance legislation in five states (California, Hawaii,
New York, New Jersey and Rhode Island) and Puerto Rico extends beyond the
7
Noncovered worksites expect much higher costs from the FMLA than those experienced by companies
actually covered (for example, 47 percent anticipate a decline in business performance and profitability).
This could reflect undue pessimism concerning compliance and administrative costs or, alternatively,
might indicate that the act would actually be relatively much more expensive for currently uncovered
employers, perhaps because of their small size.
Policy Watch: The Family and Medical Leave Act 183
/ 3008 0003 Mp 183 Thursday Oct 19 01:33 PM LP–JEP 0003
Pregnancy Discrimination Act by requiring firms with two or more employees to
provide temporary disability benefits, including partial replacement of lost wages,
to workers who have been with the employer for a minimum amount of time (gen-
erally 14 to 20 weeks). Nonwork related disabilities pertaining to pregnancy and
childbirth are included under these laws. Typically, the leave period ranges from
four weeks to four months, and the cash benefit replaces about half the wages of
the average employee.
Finally, 12 states and the District of Columbia enacted maternity leave legisla-
tion prior to the FMLA. The leave periods varied from six to 17 weeks, with all states
exempting small employers and many excluding medium-sized firms (Klerman and
Leibowitz, forthcoming). Several other states required maternity leave for govern-
ment (but not private) workers or mandated rights to limited job absences but
without guaranteeing reinstatement. By 1992, 25 states had enacted some type of
parental leave statute (Trzcinski and Alpert, 1994).
These laws, in combination with employer sick leave and vacation policies,
imply that a substantial number of workers eligible for FMLA leave would have
been able to take some time off work in its absence, and such leaves were relatively
common prior to its enactment. For example, during the 1986-88 period, 73 per-
cent of ‘‘employed’’ women with one-month-old infants were on leave (and 41
percent on paid leave) rather than working, as were 41 percent (16 percent) of
those with two-month-old babies (Klerman and Leibowitz, 1994b). Most women
were able to return to their old jobs following childbirth, if they so desired. Thus,
in 1990, 85 percent of new mothers who held full-time positions one year earlier
and were also working (either full- or part-time) six months after the childbirth had
returned to the same employer, as had 79 percent of those employed 18 months
after the birth (Klerman and Leibowitz, 1994a).
Data from the ISR Employee Survey directly suggests the modest effect of the
FMLA on leave taking. Over the 18-month period ending in the summer of 1995,
17 percent of workers aged 18 or over reported using leave for a reason included
in the FMLA. However, just 7 percent of this group (or 1.2 percent of all employees)
claimed that the leave was taken under the federal law. With over 125 million work-
ers in the United States, this implies that over one million additional individuals
received family and medical leave as a result of the act, with a larger number of
persons likely to use FMLA leave over more extended periods of time. However,
even if the fraction of individuals saying they took time off work under the law is
understated by a factor of two or three—perhaps because many did not realize
their leave was due to the legislation—it still represents only a small fraction of job
absences for reasons covered by the FMLA. By this criteria, the act has had a rela-
tively small impact.
The FMLA could have little effect on the overall use of leave while having a
stronger impact for selected groups, such as women with young children. Waldfogel
(1996) recently examined whether the legislation increased the extent of maternity
leave among full-time workers. Her findings suggest a modest increase in the avail-
184 Journal of Economic Perspectives
/ 3008 0003 Mp 184 Thursday Oct 19 01:33 PM LP–JEP 0003
ability of maternity leave. In particular, leave taking by women with infants in-
creased, after passage of the FMLA, relative to that of other women or men. How-
ever, this conclusion is qualified because: 1) the coefficients are frequently esti-
mated imprecisely and exhibit substantial instability across post-FMLA years; 2) the
growth in maternity leave is generally greater for persons working for large (500 or
more employees) than medium (100–499 employees) employers, even though the
smaller companies less frequently voluntarily provide leave benefits; and 3) there
is no consistent indication that the FMLA had a larger effect in states without than
in those with pre-existing maternity leave mandates.
Data from the ISR Survey of Employees suggests that overall patterns of leave
taking have been largely unaffected by the FMLA. Among persons working for
covered employers, 60 percent of job absences were for the employee’s own health
problem other than maternity disability, 4 percent for maternity disability, 13 per-
cent to care for a newborn or adopted child, and 23 percent to care for an ill relative
(Commission on Family and Medical Leave, 1996). For respondents holding posi-
tions in exempt establishments, the corresponding percentages were 56 percent,
7 percent, 15 percent and 22 percent. This may not be a perfect comparison since
the statistics include all leaves for reasons covered by the act, whether or not they
represent FMLA leave, and health or family status could differ systematically with
firm size (and so coverage under the law). However, the findings provide little
indication that FMLA eligibility has led to relative increases in work absences for
maternity or to care for sick family members.
To summarize, the small effects of the FMLA reflect the limited scope of the
legislation and the existence of other mechanisms by which many employees of
medium and large firms could take time off the job, even prior to its passage into
law.
Concluding Thoughts
The enactment of the Family and Medical Leave Act has considerable symbolic
importance by demonstrating the emergence of a widespread consensus regarding
the need to balance the competing interests of work and family. However, the actual
provisions of the act are modest. The leaves are short and unpaid. Only about half
the workforce is eligible under the law, and most of them would have been able to
take at least some time off the job even without the legislation. Neither economic
theory nor the available empirical evidence provides a clear indication of the effects
of family leave mandates on employment or earnings.
Given its limited scope, the FMLA has probably not had a strong impact on
either workers or employers. More substantial effects might be expected, how-
ever, if the legislation were extended to cover more workers, to require payment
during the job absence, or to lengthen the duration of the job-protected leave.
It is also possible that the usage of FMLA leave will increase over time. My pro-
Christopher J. Ruhm 185
/ 3008 0003 Mp 185 Thursday Oct 19 01:33 PM LP–JEP 0003
visional belief is that the advantages of a limited extension of the act—like cov-
ering employers with 25–49 employees or requiring partial payment during a
portion of the leave—would probably outweigh the costs. Conversely, I antici-
pate that the costs would exceed the benefits if the extremely generous leave
provisions of many European countries (like rights to a year or more of leave
with pay) were adopted.
References
Aghion, Phillipe, and Benjamin Hermalin,
‘‘Legal Restrictions on Private Contracts Can En-
hance Efficiency,’’ Journal of Law, Economics, and
Organization, Fall 1990, 6, 381–409.
Cantor, David, et al., ‘‘The Impact of the Fam-
ily and Medical Leave Act: A Survey of Employ-
ers,’’ mimeo, Westat, Inc., Rockville, Md., Octo-
ber 2, 1995.
Commission on Family and Medical Leave, A
Workable Balance: Report to Congress on Family and
Medical Leave Policies. Washington, D.C.: U.S. De-
partment of Labor, 1996.
Dalto, Guy C., ‘‘A Structural Approach to
Women’s Hometime and Experience-Earnings
Profiles: Maternity Leave and Public Policy,’’ Pop-
ulation Research and Public Policy Review, Septem-
ber 1989, 8, 247–66.
Kamerman, Sheila B., ‘‘Childcare Policies and
Programs: An International Overview,’’ Journal of
Social Issues, 1991, 47:2, 179–96.
Klerman, Jacob A., and Arleen Leibowitz,
‘‘Employment Continuity Among New Moth-
ers.’’ National Longitudinal Surveys Discussion
Paper No. 95–22, Washington D.C., U.S. Depart-
ment of Labor, Bureau of Labor Statistics, March
1994a.
Klerman, Jacob A., and Arleen Leibowitz,
‘‘The Work-Employment Decision Among New
Mothers,’’ Journal of Human Resources, Spring
1994b, 29, 277–303.
Klerman, Jacob A., and Arleen Leibowitz, ‘‘La-
bor Supply Effects of State Maternity Leave Leg-
islation.’’ In Blau, Francine D., and Ronald Eh-
renberg, eds. Gender and Family Issues in the Work-
place. New York: Russell Sage Foundation Press,
forthcoming.
McGonagle, Katherine A., et al., ‘‘Commission
on Leave Survey of Employees on the Impact of the
Family and Medical Leave Act,’’ mimeo, Institute
for Social Research, Survey Research Center, Uni-
versity of Michigan, Ann Arbor, October 13, 1995.
Mitchell, Olivia, ‘‘The Effect of Mandatory
Benefits Packages.’’ In Bassi, Lori J., David L.
Crawford, and Ronald G. Ehrenberg, eds., Re-
search in Labor Economics. Vol. 11, Greenwich,
Conn.: JAI Press, 1990, pp. 297–320.
Organization for Economic Cooperation and
Development, ‘‘Long-Term Leave for Parents in
OECD Countries.’’ In Employment Outlook: July
1995. Paris: OECD Department of Economics
and Statistics, 1995, pp. 171–202.
Rothschild, Michael, and Joseph Stiglitz,
‘‘Equilibrium in Competitive Insurance Markets:
An Essay on the Economics of Imperfect Infor-
mation,’’ Quarterly Journal of Economics, Novem-
ber 1976, 90, 629–50.
Ruhm, Christopher J., ‘‘The Economic Con-
sequences of Parental Leave Mandates: Lessons
From Europe,’’ mimeo, University of North Car-
olina, Greensboro, January 1997.
Ruhm, Christopher J., and Jackqueline L.
Teague, ‘‘Parental Leave Policies in Europe and
North America.’’ In Blau, Francine D., and Ron-
ald Ehrenberg, eds. Gender and Family Issues in the
Workplace. New York: Russell Sage Foundation
Press, forthcoming.
Silverman, Celia, et al., EBRI Databook on Em-
ployee Benefits. 3rd ed., Washington, D.C.: Em-
186 Journal of Economic Perspectives
/ 3008 0003 Mp 186 Thursday Oct 19 01:33 PM LP–JEP 0003
ployee Benefits Research Institute, 1995.
Spalter-Roth, Roberta A., and Heidi I. Hart-
mann, Unnecessary Losses: Costs to Americans of the
Lack of Family and Medical Leave. Washington,
D.C.: Institute for Women’s Policy Research,
1990.
Summers, Lawrence, ‘‘Some Simple Econom-
ics of Mandated Benefits,’’ American Economic Re-
view, May 1989, 79, 177–83.
Trzcinski, Eileen, and William T. Alpert,
‘‘Pregnancy and Parental Leave Benefits in the
United States and Canada: Judicial Decisions and
Legislation,’’ Journal of Human Resources, Spring
1994, 29, 535–54.
Waldfogel, Jane, ‘‘The Family Gap for Young
Women in the U.S. and U.K.: Can Maternity
Leave Make a Difference?,’’ mimeo, Harvard
University, October 1994.
Waldfogel, Jane, ‘‘The Impact of the Family
and Medical Leave Act on Coverage, Leave-Tak-
ing, Employment, and Earnings,’’ mimeo, Co-
lumbia University, New York, May 1996.
Waldfogel, Jane, ‘‘Working Mothers Then and
Now: A Cross-Cohort Analysis of the Effects of
Maternity Leave on Women’s Pay.’’ In Blau,
Francine D., and Ronald Ehrenberg, eds. Gender
and Family Issues in the Workplace. New York: Rus-
sell Sage Foundation Press, forthcoming.
