Part-Time Job Growth and the Labor Effects of Policy Responses

CRS Report for Congress
Part-Time Job Grow th and t he
Labor Effects o f Policy Responses
Upda ted Octobe r 28, 2003
LindaLevine
Specialist in Labor Economics
Domestic Social Policy Division

Congressional Research Service ˜ The Library of Congress

Part-Time J ob Growth and the
Labor Effects of P olicy Responses
Summary
The doubling o f p ersons who u sually work part-time (i.e., 1-34 hours p er week)
accounted for about one-fifth o f employm ent growth s ince 1969. T h e m ore rapid
increas e i n part-time versus full-time em ployment means t hat t oday 1 in 6 workers
h a s a part-time schedule, up quite modestly from 1 in 7 i n 1969. Thus, t h e
predominant work s chedule rem ai ns a full-time one.
The part-time labor force i s comprised of those who want short s chedules and
those who w a n t f u l l -time hours. Most part-timers still work few hours b y choice,
despite the l ong-term i ncrease i n p ersons involuntarily employed part-time. W ithin
the group of persons involuntar y e m p l o ye d p art-time, t here are t hose who usually
work full-time and t hose w h o u s ually work part-time. The increase i n i nvoluntary
part-time work has o ccu rred among those who usually have short workweek s, which
suggests t hat t heir prospect of obtai ning full-time jobs has diminished over time.
One ex planation for the great er use of alternative work arrangements (e.g., part-
time and t em porary emp l o ym en t ) is that they enable firms t o m ore effici ently
accommodate heightened competitiveness and variability in the m arketplace than if
they relied on t raditional (i.e., full-time, l ong-term) j o b s . A n other is that flex ible
work arrangem ent s enabl e fi rm s t o s ave o n l abor cost s, t h ereby m aki n g t hem m ore
competiti v e at what some believe is the ex pense of workers, thei r families, and
soci et y. A l ess widel y discussed ex planation i s t he possibility of a mismatch bet ween
t h e f ai r l y l ow skill qualifications of involuntary p art-timers (e.g., welfare m others
seeki n g w o r k a n d m en d i s pl aced from h i gh-wage fact ory j obs) and t h e h ei gh t ened
ski l l r e q u i rem ent s of a growi ng share o f j obs. Di s agreem ent over t he causes and
consequences of nonstandard jobs is likel y t o continue as lon g a s e m ployers t reat
them differently from t raditional j obs in terms o f j ob secu ri t y as well as
com p ensat i o n l evel s and pract i ces.
Some advocat e t hat policies, incl u ding t he public-private safety net (e.g.,
unemploym ent i nsurance, social security, p e n sion and h ealth benefits), should b e
reshaped so that they no longer are t ailored for traditional j o b s. Among other
changes, they have recommended amending the E qual P ay Act t o require employers
to provide equal hourly pay for equal work regardles s of full-time/part-time stat us.
The hourly pay d isparity between p ar t - t i m e and full-time workers i s unlikely t o be
much affected by such a change, however, b ecause most of the gap is due to
differences in the two group’s p ersonal and job characteristics. In addition, advocates
have proposed that employe r s b e required t o p rovide benefits (e.g., health and
retirement plans ) t o part-time em ployees . Employer m andates could produce winners
and l osers, however. The winners would i nclude involuntary p art-timers who obtain
full-time jobs and s ome part-timers who gai n new benefits. The losers would i nclude
part-timers who already are covered through other sources (e.g., a spouse’s h ealt h
plan) o r who prefer high er wa ge s o v e r t he new b enefits, as well as voluntary p art-
timers who accept full-time jobs or drop out of the l abor market due to a reduction
in part-time job opportunities.

Contents
WhoarePart-TimeWorkers? ........................................2
DefiningTerms ...............................................2
Their Demographic Breakdown ...................................3
TheTrend in Part-TimeEmployment ..................................5
VoluntaryPart-TimeEmployment .................................5
Involuntary P art-Time Employment ...............................6
Reasons for t he Growth in Involuntary P art-Time Employment ..............7
J o b Growth b y Industry .........................................8
ThePart-Time/Full-TimeWageGap ...............................9
Employment-BasedBenefits ....................................11
Di fferences i n R ecei pt and C ost o f Benefi t s ....................11
The Labor Market Effect of Mandating W orkplace Benefit
Coverage ...........................................13
Underemploymentand aSkill Mismatch ...........................15
ListofTables
Table 1 . P ercent Distribution o f All Employed Persons and o f P ersons
Employed Part-Time, b y Demographic C haracteristic, 2002 ............4
Table 2. Employed P ersons by Full-Time and P art-Time Stat us During Peak
YearsoftheBusiness Cycle ......................................5
Table 3 . Employed P ersons by Reason for W orking Part-Time during P eak
YearsoftheBusiness Cycle ......................................7
Table 4 . P ercent Distribution o f 25-64 Year Olds Employed Fu ll-Time
and P art-Time by Educational Attainment, 2002 .....................17

Part-Time J ob Growth and the
Labor Effects o f Policy R esponses
Part-time employment is one form of nontraditional work arrangements,¹ which
have generat e d i nt erest duri n g t he l ast few d ecades.² The t raditional work
arrangem ent t ypically is charact erized as a full-time, l ong-term j ob with fringe
benefits. The ex pansion o f a l t e r n atives to the s tandard arrangement h as prompted
concern about job s ecurity as well as the adequacy of earni n g s l e v e l s and benefit
coverage during i ndividuals’ work lives an d ex t endi ng i n t o t h ei r ret i rem ent years.
Some believe that the public-private soci al welfare s ys t e m, which i ncludes
unem p l o ym ent i nsurance (UI) and s oci a l s e c u ri t y as wel l as heal t h and p ensi on
benefits, h a s not changed s ufficiently to satisfactorily address t he needs o f t he
increased share o f all workers with more tenuous connections to their employers and
wi t h m o re vari ed em pl oyee- employer relationships.³ Initially, legislation was offered
t h at would h ave n arrowed t he hourly wage gap b etween part-time and full-tim e
workers and would h ave p romoted b enefit coverage of part-timers (e.g., H.R. 3657
and H.R. 3682 in the 104^th Congress). P roposals s ubsequently were introduced in the
105^th Congress to creat e a commission to study the impact of part-time em ployment
(S. 1453) and another t o s tudy a range of labor force i ssues including part-time work
(H.R. 2997). In t he 107^th Congress, interest continued in extending employer-
provided h ealth care coverage to part-time workers (S. 2639) and i n enabling j obless
persons seeking p art-time employment to receive UI benefits (H.R. 773). Legislation
concerning part-time work and UI b enefit eligibility has b een proposed in the 108^th
Congress (H.R. 1652) as well.
This report provides an overview of part-time em ployment, ex amining who and
how many part-time workers t here are as wel l as why thei r s hare of total employm ent
has i ncreas ed over time. The report analyzes t he potential effect on workers were
Congress to prohibit wage d iscrimination b ased on hours work e d a n d t o require

¹ T he t erms nontraditional, alternative, nonstandard, or f lexible work arrangements are used
interchangeably in this report.
² In addition t o part-timers, persons engaged i n alternative wo r k a r r angements include
employees of contract services firms, inde pendent contractors or c o n s u l t a nts, on-call
workers, leased employees, and temporary workers. For information on t emporary workers
specifically see CRS Report RL30072, Temporary Workers as M embers of the Contingent
Labor Force , by Linda Levi ne.
³ Richard S. Belous, “ T he Rise of t he Continge nt Workforce: Growth of T e mporary, Part-
T i me , a nd Subcontracted Employme nt,” Looking Ahead , vol. X IX , no. 1 ( J une 1997); a nd
Virginia L. duRivage, ed., New Policies f or the P art-Time and Contingent Workforce
(Armonk, NY: M.E. Sharpe, 1992). ( Hereafter c ited a s duRivage, New Policies f or the Part-
Time and Contingent Workforce .)

benefit eligibility of part-time workers. It cl oses by considering whether a mismatch
between the qualifications of involuntary p art-time workers and the h eigh tened s kill
requirements of j obs might ex plai n s o m e o f t h e long-term rise i n part-time
employment.
Who a re Part-Time W orkers?
DefiningTerms
Th e U .S. Bureau of Labor Statistics (BLS) derives dat a on part- and full-time
e m p l o yment from t he Current Population S urvey (CPS). The count of part-time
employment is the number o f i ndividuals working 1 -34 hours a week. The count of
full-time employment is the number of i ndividuals working 3 5 o r m ore hours a week.
In order t o reflect a w orker’s normal s che dule rather t han any aberration during t he
survey week, respondents are asked whether they usually work 1-34 hours (in which
cas e t hey are cl assified as part-time workers) or usually work at least 3 5 hours (in
which cas e t hey are cl assified as full-time workers).
Because the s tatistics are obtained fro m a s u r v ey of households rather than of
firms, the figures relate to part-time and full-ti m e w o rkers not jobs. It i s t hus
possible for individuals who hold m ultiple job s , o ne of which i s part-time, t o be
classified as full-time workers i f t heir hours t otal at least 35. As a result o f t he major
revision to the C PS implemented in J anuary 1994, information h as become available
on the p revalence o f p eople hold i ng multiple jobs and t heir usual hours i n t hose
jobs.⁴ Accordin g t o BLS , t he trend i n part-time em ployment (which is ex amined
shortly) would b e little changed b ased on a count of jobs rather than of workers.
Other changes made to the C PS in 1994 affect the consistency over time of the
part-time/full- time data series. The changes’ impact on trends are noted where
appropriate in the following pages.
Some individuals who work on a part-time basis choose t o do s o while others
would p refer l onger hours. The former are often referred t o as voluntary part-time
workers . They elect to work 1-34 hours p er week for what BLS considers t o b e
noneconomic reas ons, i ncluding problem s arrangi ng child care, other family or
personal obligations, health or m e d i cal l imitations, i n s chool or training, retired or
social security limit on earnings, vacation o r p ersonal d ay, l egal or religious holiday,

⁴ In 2002, 7.3 million workers moonlighted (i.e., held more than one j ob). T he most
common f orm of multiple j obholding involved full-time workers with a second part-time j ob
(3.9 million). Almost 1.6 million workers held primary and secondary j obs with each being
part-time, and 856,00 of these workers were employed at least 35 hours on t heir part-time
j obs combined. Another 1.2 million i ndivi duals who worked a full-time week held two j obs,
one of which had variable hours. T hus, 6 million workers who held one or more part-time
j obs were classified as full-time workers because the combined work week from t hose j obs
totaled at l east 35 hours. Part-time j obholders with full-time schedules more o f t e n are
prime-age worke r s ( 25-54 year olds) and married men compared t o either gr oup’s
prevalence among “official” part-time workers.

and weather-related curtailment. T he latter are often referred t o as involuntary part-
time workers or as bei ng em ployed part-time for economic reasons. T hey work l ess
than 35 hours a week due to slack work or business conditions, could only find p art-
time work, s eas onal work, and j ob started or ended during s urvey week.
Thei r Demogr aphi c Br eakdow n
W o men o f all ages, younger (16-24) and o lder (at l east 55) men, as well as white
workers m ak e u p l arger s hares o f workers voluntarily em ployed part-time than of al l
workers. (S ee Table 1.) Voluntary p art-timers in 2002 had an average work week
of 21.4 hours. Individuals most often gave “in school or training” and “other family⁵
or personal obligations” as t heir reason for choosing p art-time hours. Women
continue to disproportionately opt for p art-time schedules, whi ch probably reflects
their efforts t o accommodate family responsibilities; however, t hey h ave b ecome
increasingl y l ess likely over time to choose p art-time employment. In contrast, t he
rate of voluntary p art-time employment has i ncreased for younger and older m ale
workers.⁶
In contrast, young women (16-34 years old) and men (16-24 years o ld) as well
as black workers are over represented am ong involuntary p art-timers regardless of
whether t hey are usually employed 1-34 ho u r s o r a m i n imum o f 3 5 hours a week.⁷
Over time, however, t he incidence o f i nvol untary p art-time schedules among prime-
age m en (25-54) usually em ployed part-time has risen more so than am ong women.⁸
Economic part-timers average a l onger work week, at 23.0 hours i n 2002, compared
to voluntary p art-timers. Among persons em ployed part-time for economic reasons,
those who usually work full-time report l onger hours (24.1) compared to those who
usually work part-time (22.3). The most frequently o f f e r e d reason for b eing
involuntarily employed part-time in 2002 was “slack work or business conditions,”
but in more robust p eriods of economic growth , i t m o r e o ften is “could only find⁹

part-time work.”
⁵ BLS, Employment and Earnings , J une 2003, T a ble 20.
⁶ T homas Nardone, “ Pa r t -T ime Employment: Reasons, Demographics, a nd T r ends,”
Journ a l of Labor Research, vol. 16, no. 3 ( summe r 1995). ( Hereafter c ited a s Nardone,
Part-Time Employment.)
⁷ An example of s omeone who usually is employed full-time but w o r ks p a r t -time f or
economic reasons is a c o n s t r u c t i on worke r who had only t hree 10-hour days of work in a
week because one j ob e nded a nd another had not ye t begun.
⁸ Nardone, Part-Time Employment.
⁹ BLS, Employment and Earnings , J an. 2003, T a ble 20.

Table 1. Percent D istribution o f All Employed Persons
and o f Persons Employed Part-Time,
by Demographic C haracteristic, 2002
Involuntary part-time
UsuallyUsually
Tot a l Vol unt ary work work
Characteristic employed part-time Tot a l f ull-time part-time
Tot a l 100.0% 100.0% 100.0% 100.0% 100.0%
16-19 4.6 19.7 11.1 8.3 11.3
20-24 9.8 15.3 19.2 18.3 19.2
25-34 22.2 14.1 23.2 23.7 23.2
35-44 25.8 16.6 20.9 20.1 21.0
45-54 22.9 13.8 16.3 18.3 16.2
55+ 14.6 20.6 9.1 11.2 8.9
Men 53.4 30.9 45.2 47.3 45.1
16-19 2.3 8.9 5.5 4.7 5.6
20-24 5.1 6.2 9.4 8.3 9.4
25-34 12.1 3.0 10.5 11.2 10.4
35-44 13.9 2.2 8.9 9.5 8.9
45-54 12.0 2.3 6.3 8.3 6.2
55+ 7.6 8.3 4.5 5.3 4.5
Women 46.7 69.0 54.8 52.1 54.9
16-19 2.3 10.7 5.6 3.6 5.7
20-24 4.7 9.1 9.8 9.5 9.8
25-34 10.1 11.1 12.7 12.4 12.8
35-44 11.9 14.4 12.0 11.2 12.1
45-54 10.9 11.5 9.9 10.1 9.9
55+ 6.7 12.3 4.6 5.3 4.5
White 83.5 87.2 77.4 82.8 77.1
B l ack 10.9 7.9 16.2 11.2 16.5
^Source:^T^a^ble^c^reated^by^C^R^S^f^r^om^B^L^S^d^ata.

The Trend i n P art-Time E mployment
The doubling o f p ersons who u sually work part-time — from 11.3 million i n
1969 to 23.0 million in 2000 — accounted for one-fifth o f employm ent growth
duri n g t h e 3 0 -year period. (See Table 2.) The m ore rapid increas e i n part-time
(104%) t han i n full-time (71%) employm en t m eans t hat 1 in 6 wo r k e r s h a d a part-
time schedule i n 2000, up quite modestly over t he past few d ecades from 1 in 7 i n

1969. The p redominant work s chedule t hus remains a full-time one.

Table 2. Employed Persons by Full-Time and Part-Time Status
During Peak Years o f t he Business Cycle
^Part-ti^m^e^Part-ti^m^e
^for^eco-^for^eco-
^Full-^a^Part-^no^mⁱ^c^b^Full-^a^Part-^no^mⁱ^c^b
^To^t^a^l^ti^m^e^ti^m^e^reas^o^ns^To^t^a^l^ti^m^e^ti^m^e^reas^o^ns
^Yea^r^(Num^bersⁱⁿ^thous^a^nds⁾⁽^Percent^d^is^tribution)
¹⁹⁶⁹^77,902^66,596^11,306^2,056^100%^85.5%^14.5%^2.6%
¹⁹⁷⁹^98,824^82,654^16,171^3,577^100%^83.6%^16.4%^3.6%
¹⁹⁸⁹^117,342^97,369^19,973^4,894^100%^83.0%^17.0%^4.2%
²⁰⁰⁰^136,891^113,846^23,044^3,227^100%^83.2%^16.8%^2.4%
^So^urce^:^T^able^created^by^C^R^S^f^r^om^B^L^S^d^ata.
^No^te^:^D^at^a^f^or¹⁹⁹⁴^an^d^s^u^b^s^e^qu^en^t^y^ears^a^reⁿ^o^t^dⁱ^r^ect^l^y^com^p^arabl^e^wⁱ^t^h^dat^a^f^o^r¹⁹⁹³^an^d^earlⁱ^e^r
^y^ears^d^u^e^t^o^a^m^aj^or^redesⁱ^gⁿ^of^t^h^e^C^PS^.
^a^Bef^o^re^1994,^t^h^e^f^u^l^l^-^tⁱ^m^e^t^ot^alⁱⁿ^cl^u^d^es^pers^on^s^u^s^u^al^l^y^em^pl^oy^ed^1-³⁴^h^o^u^r^s^b^u^t^w^h^o^w^ork^e^d³⁵
^or^m^o^re^h^o^u^r^s^d^u^rⁱⁿ^g^t^h^e^ref^e^ren^c^e^w^eek^.^F^rom¹⁹⁹⁴^f^o^rw^ard,^s^u^ch^pers^on^s^a^reⁱⁿ^cl^u^d^edⁱⁿ^t^h^e
^part^-^tⁱ^m^e^t^ot^al^.^b
^In^clu^d^esⁱⁿ^divⁱ^d^u^a^l^s^e^m^p^loy^e^d¹^-³⁴^h^ou^rs^f^o^r^econ^o^mⁱ^c^r^eas^on^s^,^w^h^o^u^s^u^ally^w^o^rk^part-^tim^e^aⁿ^d
^w^h^o^u^s^u^ally^w^o^rk^f^u^ll-^tim^e^s^ch^edu^l^es^.^T^h^u^s^,^p^ers^oⁿ^s^em^ploy^ed^part-^tim^e^f^or^econ^o^mⁱ^c^r^eas^on^s
^are^not^a^s^ub^se^t^o^f^a^l^l^p^a^r^t^-tⁱ^m^e^w^o^r^ke^r^s⁽ⁱ^.^e^.^,^p^e^r^s^oⁿ^s^w^ho^usua^l^l^y^w^o^r^k¹^-³⁴^ho^ur^s^r^e^g^a^r^d^l^e^s^s
^of^reas^on^).
The pace of part-time em ployment growth appears t o have s lowed during t he
period under observation. This trend might be related t o t he heightened commitment
of wom en t o t he l abor force, as refl ect ed i n t h ei r d ecreased propensi t y t o l eave ful l -
time jobs and t heir i n creased propensity to move from part-time jobs or
nonparticipation t o full-time jobs.¹⁰
Voluntary P art-Time E mploym ent
Most part-timers — about 4 i n every 5 — choose s hort workweeks. In 2002, for
ex ample, BLS d ata s how that18.9 million out of 23.8 million p art-time workers opted
for l ess t han 3 5 hours o f w ork p er week on average.

¹⁰ Donald R. Williams, “Women’s Part-Time Employment: A Gross Flow Analysis,”
Monthly Labor Review, Apr. 1995.

Although t he share o f voluntary p art-time workers d ecreased in the 1970s (from
85.4% in 1969 to 79.7% in 1979) and t hen edged further downward i n t he 1980s (to

77.1% in 1989), t he trend m ay have reversed more recently (rising to 81.6% in 2000).

Some portion o f t he turnaround in the 1990s reflects changes in the C PS
questionnaire which, among other t hings , w a s reworded t o m ake i t easier t o
determine whether noneconomic/voluntar y o r economic/involuntary factors affected
the l engt h o f respondents’ workweeks.¹¹ Voluntary part-timers as a s hare of al l part-
timers jumped from 72.1% in 1993, based o n t he old questionnaire, to 75.7% in 1994
when the revised questionnaire was introduced. S ince this large 1 -year jump, t he
i n ci d e n ce of voluntary part-time em ployment has continued t o rise but at a m o r e
modestpace.
Involuntary Part-Time Employment
The m aj o r s t ory behind t he increas e over time in part-time em ployment
concerns those who work part-time but would p refer full-time ho u r s . Once the
busi n ess-cycl e effect¹² is eliminated by focusing on peak years o f economic activity,
it becomes clear that involuntary p art-time w o r k h a s g r o wn over t he long run. As
shown i n Table 2, all persons employed part-time for economic reasons numbered
more than 3.2 million i n 2000, which i s about 1½ times t he 1969 level. Involuntary
part-time employment accounted for 2.6% of total employm ent i n 1969. The
proportion s ubsequently rose, with a l arger i ncrease o ccu rring in the 1970s than
1980s. The involuntary p art-tim e e m p l o ym ent rate s ubsequently fell to 2.4% in
2000, with the s e e m ingl y reduced incidence partly due to the aforementioned C PS
revision.¹³
The l ong-term i ncrease i n i nvoluntary p art-time employment has o ccurred
among those who usually have short workweeks. As presented i n Table 3, l ess t han
one-half o f persons em ployed part-time for economic reasons in 1969 usually worked
part-time. T he share grew s ubstantially during t he 1970s and continued t o ex p and,

¹¹ Nardone, Part-Time Employment.
¹² In 1979, at the peak of a business cycle, persons involuntarily employed less than 35 hours
per week totaled almost 3.6 million. A f ew years l ater, around the time of t he 1981-1982
recession, their number climbed t o 6.3 million. Af ter t apering off during t he 1980s
recovery, t he level t urned up again — r eachin g 6 . 5 mi l lion — f ollowi ng the 1990-1991
recession. T he i nfluence of cyclical fluctuations in aggr egate demand on t hose u n a b l e t o
get as many hours of work per w e e k a s t h e y would like i s evi dent from t his pattern and
demonstrated empirically in Ronald A. Ratti, “Involuntary Part-Time Employment: Cyclical
Behavi or and T rend Over 1968-1987,” Economic Letters, 1991.
¹³ Between 1993 and 1994, when the r evised questionnaire was implemented, the number
of part-timers who preferred l onger hours d r o p ped from 6,481,000 to 4,625,000, which
could r esult i n at l east a 1 percentage point d ecrease i n t h e involuntary employment r ate.
Note: Before t he 1994 CPS r evision, interviewers inferred whether persons employed part-
time f or economic reasons wanted and were available f or full-time j obs. From 1994 onward,
respondents have been asked explicitly about their desire and availability for f ull-time work.
“[T ] he reduced number of i nvoluntary part-time employees results almost entirely from t he
direct question about desire for f ull-time work; the question on availability has little affect.”
Nardone, Part-Time Employment, p. 289.

but at a considerably diminished p ace, during t he 1980s. W hile the s hare declined
during t he 1990s, i t remains well above its 1969 level. The d ata s u ggest that the
prospect of moving from part-time to full-time em ployment has diminished over t he
years, thereby m aking part-time work a m ore permanent stat us for t hose who would
prefer full-time work.
Table 3. Employed Persons by Reason f or Working Part-Time
during Peak Years o f t he Business Cycle
(numbers in thousands)
Personsat
work <35 1969 1979 1989 2000
hoursper
w eek Number P ercent Number P ercent Number P ercent Number P ercent
Part-time
for 2,056 100.0 3,577 100.0 4,894 100.0 3,227 100.0
economic
reasons
Usually
work part 963 46.8 2,102 58.8 3,164 64.7 1,894 58.7
time
^Source:^T^a^ble^c^reated^by^C^R^S^f^r^om^B^L^S^d^ata.
^No^te:^D^a^t^a^f^o^r¹⁹⁹⁴^an^d^s^u^b^s^e^qu^en^t^y^ears^a^reⁿ^o^t^dⁱ^r^ect^l^y^com^p^arabl^e^wⁱ^t^h^dat^a^f^o^r¹⁹⁹³^an^d^earlⁱ^e^r
^y^ears^d^u^e^t^o^a^m^aj^or^redesⁱ^gⁿ^of^t^h^e^C^PS^.
^a^T^o^t^a^l^w^ho^usua^l^l^y^w^o^r^k^p^a^r^t^tⁱ^m^eⁱ^nc^l^u^d^e^s^t^ho^se^usua^l^l^y^e^m^p^l^o^y^e^d¹^-³⁴^ho^ur^s^p^e^r^w^e^e^k^b^u^t⁽¹⁾^w^h^o
^w^e^re^abs^eⁿ^t^f^r^om^w^o^rk^f^o^r^t^h^e^en^tⁱ^r^e^r^ef^eren^ce^w^eek^or⁽²⁾^f^or¹⁹⁹⁴^an^d^a^f^t^er,^w^h^o^w^o^rk^ed³⁵
^o^r^m^o^r^e^ho^ur^s^d^urⁱⁿ^g^t^he^r^e^fe^r^e^nc^e^w^e^e^k^.^A^s^t^he^se^gr^o^u^p^s^a^r^eⁿ^o^t^sho^wⁿ^s^e^p^a^r^a^t^e^l^y^,^t^h^e^s^um
^o^f^th^e^p^ar^ts^is^less^th^an^th^e^t^o^t^al.
Reasons for the Grow th in
Involuntary P art-Time E mployment
The l ong-run i ncreas e i n part-time em ployment has occurred among those who
want to work full-ti me hours, which likely m eans t hat d emand h as outpaced the
supply o f voluntary p art-time workers. Therefore, ex planations of the t rend typically
have focused o n t he demand side, t hat i s, on em ployers’ motivation for favoring part-
time over full-time job creation.
The changing economy i s one ex planati o n c o m monly o ffered for business’
greater use o f nontraditional work arrangem ents, i ncluding part-time and t emporary
workers, l eased em pl oyees, and em pl oyees of cont ract servi ces fi rm s. S u ch fact ors
as deregu lation and interna tionaliz ation o f p roduct m arkets, i t i s asserted, h ave m ade
t h e m ark e t p l ace i n creasingly competitive and vari ab l e . S o m e c o n t en d t h a t c o m b i n i n g
di fferent work arrangem ent s al l o ws fi rm s t o m ore effi ci ent l y accom m odat e changi ng
or fluctuating p atterns o f d emand for goods and s ervices than if they relied s olely o n
traditional (full-time, l o n g-t erm ) jobs. Great er flex ibility in staffing has been
achieved t hrough s uch s trategies as m aint aining a core workforce augm ented b y (1)

calling o n workers directly or through t em porary agencies when p roduction m ust b e
increase d t o f i l l the m ore s poradic orders of customers who no longer want to
maintain sizeable inventories, and by (2) scheduling part-time workers to ensure
coverage during s tore hours which have been ex tended t o m eet the n eeds o f t oday’s
dual-earner families.
Another l eading ex planation for the i ncreas e i n alternative work arrangem ents
is labor cost minimization. The s ource of lower costs is twofold: from reduction i n
paid non-productive time (i.e., having “just-in-time employment” rather t h an a
constant staff l evel over t he course of a d ay, week, o r year ) and from t he relatively
low wages and limited b enefits of nonstandard jobs. S ome oppose t he creation o f a
workforce v ariously described as t wo-tiered, disposable, o r m arginal which they
believe depresses m oral e due to unequal t reatment o f employees and d ampens
product i v i t y growt h due t o reduced em pl oyer-p rovi ded t rai n i n g a s w el l as diminishe d
reasons to innovate. In t heir view, t he increas ed competitiveness o f U.S. firms
achieved t hrough t he proliferation o f flex i ble wor k a r r angements has come —
literally — at t he ex pense of workers and, ultimately, of s ociety to the ex t ent t hat
m o re i ndi vi dual s rel y on publ i c assi st ance (e.g., wel fare and m edi cai d) because of t h e
“low quality” of nontraditional j obs.
A l ess widel y discussed ex planation c oncerns the skill composition of
involuntary p art-time workers and the n ature o f j ob growth. W ith many more jobs
t oday t han i n t he recent p ast requi ri ng fai rl y hi gh educat i onal at t ai n m ent (i .e., at l east
some postsecondary schooling), employers may h ave found that a growing share o f
workers do no t possess the s kill levels they are s eeking t o fill full-time long-term
positions. In o ther words, persons involunt arily employed part-time might have been
on t h e ri s e over t he l ong run b ecause of a m i s m at ch b et ween t h ei r qual i fi cat i ons and
the requirements o f m any “high quality” j ob opportunities. Fo r t he same reason,
firms m ay not have converted as many part -time to full-time jobs as might have been
ex pected during t he tight labor mark ets t hat p revailed i n t he late 1990s.
Job G r ow t h by I ndustr y
The greater incidence o f p art-time work might be related t o above-average job
gr o wth in industries t hat have histori cally relied o n p art-time workers o r t o a n
increased rate of part-time scheduling with in industries. The former ex p lanation
reflects s hifts i n customer d emand for goods and s erv i c e s a m ong industries, and
hence, in the i ndustrial d istribution o f e m p loym ent. The l atter (within-industry)
ex planation reflect s a change in the staffi n g strategy of firms, which might be
motivat ed by either labor flex ibility or labor cost considerations.
Rapid employm ent gains in industries t hat are historically heavy u sers of part-
time workers were estimated t o account for the entire i ncrease i n t he ratio of part-
time to full-time employment during t he 1980s and i nto t he early 1990s.¹⁴ Part-time
intensive i ndustries i nclude services (e.g., business and repair, p ersonal, medical

¹⁴ Chris T illy, Half a J ob: Bad and Good Part-Tim e J obs in a Changing Labor Market,
(Philadelphia, PA: T emple University Press, 1996). ( Hereafter cited as T illy, Hal f a J ob .);
and Nardone, Part-Time Employment.

ex cluding hospitals, and professional); ret ai l t rade; and fi nance, i n surance, and real
estate. T hus, i t appears t hat t he differential rate o f j ob growth across i nd u s t ries —
which reflects nothing more invidious than c h a n gi ng consumer demand for goods
and s ervices — h as fueled the growt h o f p art-time employment since 1980.
During the 1970s, however, t he increased rate of part-time employment within
industries d id account for a substantial s hare of the h eigh tened i ncidence of part-time
employment.¹⁵ The retail t rade and s ervices industry groups in particular stepped u p
their h iring o f p art-time as compared to full-time workers i n t his d ecade.
Taken t ogether, these findings s uggest fi rm s h ad changed t hei r i n t ernal st affi ng
strategi es and achieved what t hey consider to be a m ore effici ent mix of part-time
and full-time jobs by 1980. But, debate about whether l abor flex ibility or labor costs
motivated the higher ratio of part-time to full-time jobs, and about the consequences
of alternative work arrangem ents, i s likel y t o c o n tinue as long as “s tandard and
nonstandard jobs [ d o not pay] similar wages to people with similar characteristics,
[ d o not] p rovide ... equal fringe b en e f i t s , [ d o not] allow ... equal opportunity for
career advancem ent l adders, a n d [ d o not ] p rovide ... an equivalent level o f j ob
securi t y.”¹⁶
The Par t-Ti m e/ Ful l - Ti me Wage Gap
The wages of part-timers are l ower than those of full-timers, but the s ize of t he
gap h a s been fairly stable for d ecades. W hile the growth i n t he part-time/full-time
wage gap t hus cannot ex plain t he upward t rend in involuntary part-time employment,
the gap’s very ex istence migh t have induced firms t o i ncreas e t heir relative use of
part-time workers.¹⁷
In 2003, private s ector firms p aid p art-time employees 45% less than full-time
employees. The former earned $9.96, and t he latter $18.02, per hour worked.¹⁸
However, the hourly wage gap n either accura tely reflects t he cost savings employers
might gain by using part-time rather than full-time workers, nor the pay disadvantage
part-timers might suffer s olely f rom working short hours. “Much o f t he pay
discrepancy bet ween full-time and part-time workers can be attributed to who t hey
are and what jobs they hold,”¹⁹ that is, t o variati o n s i n t he distribution of part-time
and full-time workers across d emographic, occupational, and i ndustrial groups.

¹⁵ See T illy Hal f a J ob , a nd Nardone, Part-Time Employment.
¹⁶ Economic Policy Institute (EPI) and Women’s Research & Education Institute (WREI) ,
Nonstandard Work, Substandard J obs , ( Washington, D . C . : E PI, 1997), p. 5. ( Hereafter
citedasEPIandWREI,Nonstandard Work .)
¹⁷ Tilly, Hal f a J ob .
¹⁸ BLS, “Employe r Costs for Employe e Compensation — M arch 2003,” news release, J une

11, 2003. (Hereafter c ited a s BLS, Employer Costs f or Employee Compensation.)

¹⁹ Sar A. Levitan a nd Elizabeth A. Conway, “ Part -T i mers: Livi ng on Half-Rations,”
Challenge , M ay-J une 1988, p. 13. See a lso M arvi n K osters and Deirdre McCullough, “Does
Part-T ime Work Pay?,” The American Enterprise, Nov.-Dec. 1994.

The results of empirical studies c o n f i r m t hat d ifferences other t han hours
worked account for t he great majority of part-time workers’ relatively l ow hourly
pay. Once they are t aken into account, t he wage gap b etween the t wo groups narrows
subst ant i al l y. Aft er adj u st i n g for com pos i t i onal d i fferences i n sex , race, educat i on,
and ex perience bet ween part-timers and full-timers, one anal ys is es timated t hat t he
former earned 29% less than the l atter. When the concentration of part-time workers
in l o w - p a i d industries and occupations was t aken into account as well, the gap
narrowed further to perhaps 10%.²⁰ A n egat i v e rel at i onshi p al s o w as di scerned
between part-time status and earnings i n 40 out of the 46 i ndustries analyzed
individually in another s tudy. After controlling for human capital and several other
factors known t o i nfluence wages, the adjusted wage gap averaged about 13%.²¹ A
third s tudy found that “regular” p art-time status depressed t he hourly wage of women
by 5%, and men by 10%, compared t o full-time workers wi t h s i milar personal and
job charact eristics.²²
The unex plained portion of t he pay differential bet ween part-time and full-time
workers reflects s ome combination o f unmeasur e d , unmeasurable or imprecisely
specified variables and wage discrimin ation b ased o n h ours worked. Because
selection bias²³ could affect estimation o f t he adjusted hourly wage gap, one analys is
developed a model t o correct for it. After m aking t his correction, the s tudy fou n d
that part-time stat us did not depres s t he wages of women generally, b u t w o men
involuntarily employed part-time and all male part-timers di d i ncur an hourly wage
penalty.²⁴
The s mall adjusted wage gap i ndicates th at differences in hourly rates o f p ay for
part-time and full-time workers are respo n s ible for little of the earnings disparity
between the t wo. Ex p ansion of the current federal requirement under t he Equal P ay
Act o f 1963, t h at firms p rovide e qual p ay for equal work, to equal hourly pay for
equal work might thus have less impact on part-timers’ wage and retirem ent i ncome
levels than anticipated by those w ho have supported s uch a proposal.

²⁰ J ohn D. Owe n, Working Hours: An Economic Analysis, ( Lexington, MA: D.C. Heath a nd
Company, 1979). ( Hereafter c ited a s Owen, Working Hours .)
²¹ Ronald G. Ehrenberg, Pamela Rosenberg and J eanne Li, “ Part-T ime Employment i n t he
United States” in Robert A. Hart, ed., Employme nt, Unemployment and Labor Utilization ,
(Boston, MA: Unwin Hyma n, Inc., 1988). ( Hereafter c ited a s Ehrenberg, e t a l., Part - Ti me
Employment.)
²² EPI a nd WREI, Nonstandard Work . Note: “Regular” part-time employment i n t h i s
analys is means e mployees who worke d l ess t han 3 5 hours and who were not in any other
nonstandard work arrangement.
²³ In the i nstant case, selection bias would occur i f i ndivi duals choose t o enter the l abor force
or choose part-time over f ull-time j obs due to factors t hat are not explicitly accounted for
in the estimation procedure and that affect wages i ndependently of part-time status.
²⁴ Rebecca M . B lank, “ Are Part-T i me J obs Bad J obs?” in Gary Burtless, ed., A Future of
Lousy J obs? (Washington, D. C. : T he Br ookings Institution, 1990). ( Hereafter cited as
Bl a n k, Are Part-Time Jobs Bad J obs ?) Note: T he EPI a nd WREI analys is in Nonstandard
Wo r k takes a different approach to adj ust for pot ential systematic differences between part-
time and full-time workers not explicitly included i n t he estimation procedure. It found that
selection bias accounts f or only a small portion of t he part-time/full-time wage gap.

Because of t h e num ber o f wom en i n t h e p a r t - t i m e l abor force as wel l as t he
number of part-time emp l o yees i n rel a t i v el y l o w -p ai d f em al e-d o m i n ated occupations,
adoption o f comparable worth as national policy h as sometimes b een advocated in
connection with the part-time worker issue.²⁵ (Comparable worth would ex t end t he
current equal p ay mandate to equal p ay for equivalent jobs within a firm.) S imilarly,
the considerable repres entation of women and youth a m ong both part-time and
relatively l ow-paid workers has m a d e rai si ng t h e federal m i n i m u m wage anot her
policy option s ometimes m entioned i n connection with the p art-time worker issue.²⁶
Employment-Based Benefits
Over t h e years, em pl oyer cost s for em pl oyee benefits have grown substantially.
Employer contributions for l egally required s ocial i nsurance²⁷ rose from $100 million
in 1929, into billions of dollars following enactment of social security and o ther
Depression-era p rograms. Employer paym en ts for m andated benefits have continued
t o i n crease over t he years, accordi n g t o U.S . Depart m ent of C o m m erce d at a, as bot h
the l abor force grew and Congress ex panded coverage, rai s ed wage ceilings , and
i n creased t ax rat es. W i t h em pl oyer ex pendi t u r e s o n d i s cret i onary benefi t s (e.g.,
vacat i on, hol i d ay, and si ck l eave; rest peri ods; p ension and p rofit-sharing p lans; and
health, disability, and life insurance) also increasing, total employee benefit costs
topped $1 t rillion by 1990.²⁸
W i t h t h e rat e o f b enefi t i n creases oft en ex ceedi n g t hat o f w age i ncreases over
time, non-wage compensation t oday consumes a greater share o f em p loyers’ t otal
l abor cost s. Most recent l y, t he em pl oyer port i o n o f b enefi t ex penses com p ri sed
27.8% of total compensation for employees in the private sector: discretionary
benefits accounted for 19.4% and m a n d ated benefits, 8.4%, o f t otal compensation
costs at p rivate sector firms i n 2003.²⁹
Diffe rences in Receipt and Cost of Benefits. High er quasi-fix ed l abor
cost s are ex pect ed t o l ead fi rm s t o reduce thei r d em and for part-time compared to

²⁵ For i nforma tion on t he potential l abor ma rket impact of this policy s ee CRS Report 98-
278, The Gender Wage Gap and Pay Equity: I s Compa r a b l e Wo r t h t he Next Step?,by
LindaLevine.
²⁶ For i nformation on t he labor market effect of the f ederal mi nimum wage s ee S u s a n N .
Houseman, “T he Effects of Employer M andates” in Ri chard B. Freeman and P eter
Gottschalk, eds., Generating Jobs: How to Increase Demand for Less-Skilled Wo r k e r s ,
(N.Y.: Russell Sage Found a t i on, 1998). ( Hereafter c ited a s Houseman, The Effects of
Employer Mandates.)
²⁷ T hese payments r epresent employer contributions to federal f unds (i.e., old-age, survivors,
disability, and hospital i nsurance; unemployment i nsurance; federal employee and railroad
retirement; veterans’ life i nsurance; military medial insurance; and workers’ compensation)
a n d t o s t a t e / l ocal funds (i.e., s t a t e / l o c a l e mp l o ye e r e t i r e me n t , t e mp o r a r y d i s a b ility insurance,
and workers’ compensation).
²⁸ U.S. Chambe r of Comme rce, Employee Benefits, ( Washington, D.C.:U.S. Chamber
Research,1991).
²⁹ BLS, Employer Costs f or Employee Compensation.

full-time workers. Employers i ncur certain costs for each employee regardless of the
number o f hours t he employee works (e.g. , for health and o ther insurance; mandated
benefits with low wage ceilings ; and recruitment, s upervision, and t raining); t hey are
per-employee (fix ed) rather than per-hour (var i a b l e) costs. All else b eing equal,
hi gh er fi x ed l abor cost s m ake i t rel at i v el y l ess ex p ensi ve for fi rm s t o em pl oy ful l -
time workers b ecause pe r -employee ex penditures are spread over m ore hours o r
recouped m ore quickly in the c a s e o f t raining for ex ample.³⁰ More speci fi cal l y,
unless b enefits can be prorated based o n hours worked o r earnings, part-timers who
receive benefits will cost firms more per hour to employ than full-timers.
However, part-time em ployees less often have acces s t o or are eligible for
p a rt i c ipation i n employer benefit plans t han are full-time em ployees. Many f ew er
part-timers than full-timers in the p rivate sector have access t o p aid l eave (e.g. , for
holidays, jury duty, military service, and v acations).³¹ If firms o ffer h ealth or pension
benefits to their employees, employees who work f a i r l y f e w hours m ay find it
d i ffi cult to meet length of service requirements or m ay be legally ex cl uded from
coverage.³² When part-timers are eligible, t he amount of the firm’s contribution m ay
be based o n hours worked or earnings which might make the employee’s contribution
suffici ently ex pensive t o cause them to forgo coverage. Other eligible part-timers
m i ght opt not t o part i ci p at e b ecause t h ey al ready are covered t hrough o t h er sources.
For t hese various reas ons, only 18% of part-time em ployees compared to 58% of full-
time em ployees in the private sect or participat ed in thei r employers’ retirem ent plans
in 2003; for p articipation i n m edical care benefits, t he proportions were 9% and 56%,
respect i v el y.³³
Factors o ther than part-time stat u s cannot fully ex plain t he difference i n
discretionary benefit receipt between part -time and full-time workers. After t aking
into account such variables as age, education, firm siz e, o ccupation, and union status,
part-timers remain sign ifi cantly less likely t han full-timers to receiv e e m p l oyer-

³⁰ Owe n, Wo rking Hours . Note: One a nalysis c hallenges the l ong-held c onception t hat
many voluntary benefits (e.g., paid leave and pensions) r epresent quasi-f ixed co s t s. It
concludes t hat only health insurance i s, at least i n part, a f ixed cost o f employment. See
Michael K. Lettau, “Comparing Benefit Costs for Full- and Part-T i me Workers,” Monthly
Labor Review, M ar. 1999.
³¹ BLS, “Employee Benefits in Private Industry, 2003,” new r elease, Sept. 2003. (Hereafter
cited as BLS, Employee Benefits in Private I ndustry.)
³² Federal l aw requires t hat only part-time e mployees who work more t han 1,000 hours per
year ( i .e., about 20 hours per week) must be c overed under a firm’s pension plan. See
earlier pages of this report f or the average weekly hours of part-time workers.
³³ BLS, Employee Benefits in Private I ndustry.

sponsored health or p e n s i o n b enefits and l eave.³⁴ Involuntary p art-timers are even
l ess l i k el y t o recei ve heal t h or pensi o n b enefi t s from t hei r em pl oyers.³⁵
The l ower incidence o f b enefit receipt by part-time workers generally has m ade
them less ex pensive t o employ t han fu ll-time workers. In 2003, for ex ample, p art-
timers cost private s ecto r em p l o yers an average of only $2.45 in benefits per hour
worked, while full-timers co s t an average o f $7.36.³⁶ Benefits added 25% to the
average hourly wage costs o f p art-time w o r k ers at p rivate firms. They added a
considerably high er 41% to the average hourly wage of full-time employees . By
increasing t he ratio of part-time to full-time workers on t heir payrolls, firms have
been able to minimize t he increas e o v er time in benefit ex penditures and likel y
contributes to em ployers’ deci sion to use part-time and other fl ex ible staffing
arrangem ent s .³⁷
The Labor Market Effect of Ma ndati ng Workplace Be nefi t Coverage.
Given t he growth of part-time and o ther nonstandar d w o r k arrangements, s ome
advocat e t hat policies need to be reshaped so they no longer are t ailored for full-time,
long-term j obs with benefits. The ex tension o f t he public-private social safety net t o
nontraditional j obholders has b een urged b y s ome as a humane m eans o f easing t he
ad justment from a more rigi d t o a more flex ible work environmen t.³⁸ One analyst has
commented tha t “ A lthough economists tend to focus o n efficiency grounds for
employer mandates, achieving equity is arguably the m ore important political
motivation for legi slating employer m andates.”³⁹
Economic theory suggests t hat requiring wo rk-based benefit p rovision for p art-
time em ployees could adversel y affect thei r wages or em ployment. Unles s firms
covered b y t he m andat e coul d t rade-off t h e b enefi t i n crease agai n st a w age d ecrease,
so that thei r compensation costs do not rise, t he aggregat e dem and for part-time labor
i s ex pect ed t o fal l .

³⁴ Bernard E. Ichniowski a nd Anne E. Preston, “New Trends in Part-T ime E mp l o yme nt,”
Proceedings of the Thirty-Ei ghth Annual Meeting of t he Industrial Relations Research
Association, Dec. 28-30, 1985; and Ehrenberg, e t a l., Part-Time Employment i n t he United
States.
³⁵ Bl a n k, Are Part-Time Jobs Bad J obs? .
³⁶ BLS, Employer Costs f or Employee Compensation. N o t e : Because of the l ower
prevalence of benefit coverage among part-time t han f ull-time workers, t he series tends to
overstate t he benefit c ost gap between the t wo.
³⁷ Susan N. Houseman, “Why Employe r s Use Flexible Staffing Arrangements: Evi dence
from an Establishment Survey,” Industrial and Labor Relations Review, vol. 55, no. 1 ( Oct.

2001).

³⁸ In addition to pr oposals requiring that firms or t he government provide certain benefits
to part-time workers, other suggested changes i nclude lowering the a nnual hours t hreshold
for coverage under t he Family and M edical Leave Act, pension portability, and changi ng the
unemployment compensation s ys tem s o t hat workers with low earnings or who are s eeking
part-time work are more often eligible. See duRivage, New Policies f or the Part-Ti me and
Contingent Workforce ; a nd EPI a nd WREI, Nonstandard Work .
³⁹ Houseman, The Effects of Employer Mandates.

Whereas it mi ght be s upposed that e mp l o ye r s will respond to legi slated
augmentation of mandatory benefits by reducing t he wages of all employees to
compensate for t he added burden, our findings s ugge st that part of their r esponse
mi ght be, in s t e a d , t o r e duce t heir hiring of part-time workers. If t hey do, the⁴⁰
labor market for t hose seeking part-time employment will shrink.
Wa ges. Stat utorily set minimum wage rat es might constrai n how much firms
can cut o r s low t he growth rate of part-time workers’ wages i n response t o a benefit
mandate. Employers would be able t o offset little, i f any, of t he benefit i ncreas e
through a pay cut if part-timers’ wages were cl o s e t o t he minimum wage. The
smaller t he gap b etween minimum and part -time wages, the great er the likelihood
that firms would adjust t o a benefit requirement by curbing part-time em ployment.⁴¹
As benefit m andates, both t hose i n effect (e.g., the Family and M edical Leave Act )
and t hose t hat h ave b een offered (e.g. , h ealth care p roposals during t he Clinton
Administration), t ypically ex em pt some part-timers, i t i s likel y t hat employers will
“shift low w age workers from work s ch edules just above the m andated hours⁴²
threshold t o j ust below it, in order t o avoid t he cost of the m andate.”
If em ployers were able t o l ower the wages of part-time workers t o compensat e
for t he increas e i n benefits, t hen t he economic well-being of s ome part-timers would
suffer. As previously noted, m any workers who d o not have health benefits through
thei r part-time jobs are i nsured through other sources ; i f t heir wages were reduced,
they would not ex perience any attendant ga in from t h e benefit requirement. Other
part-time workers might not value t he additional benefits as much as they value t heir
forgone wages. In this cas e, as w e l l , the imposition of a benefit package would
diminish t he workers’ economic well-being.⁴³
Employment. A reduction i n part-time jobs might have the s al utary effect of
b ringi ng demand closer to the s upply o f voluntary p art-time workers. Fi rms cou l d
maintain thei r l evel of output by em ploying m ore full-time workers whom the benefit
mandate has m ade l ess ex p ensive to use. T h e opportunity then would i ncrease for
involuntary p art-timers to obtain t he full-time jobs they prefer. Alternatively, firms
in some instances could s ubstitute capital for the now more cos t l y labor input and
consequently cutback their t otal employment.
Th e e x t en t t o which imposition of a workplace benefit m andate increas es the
relative cost of utilizing part-time workers would depend on whet h er fi rms m ust
make the s ame p ayment for each employee regardless of hours worked o r earnings
level. If the employer’s contribution i s a fix ed (per-employee) sum, t h e m an date

⁴⁰ Mark Montgome ry and J ames Cosgrove , “ T he Effect of Employee Benefits o n t h e
Demand for Part-T i me Workers,” Industrial and Lab o r R e l a t i o ns Review, vol. 47, no. 1
(Oct. 1993), p. 96.
⁴¹ Olivia S. Mitchell, “T he Effects of M andating Benefit Packages” i n Laurie J . Bassi and
D a vi d L. Crawford, eds., Research in Labor Economics, vol. 11, (Greenwich, Conn.: J A I
Press Inc., 1990). ( Hereafter cited as M itchell, The Effects of M andating Benefit Packages.)
⁴² T homas C. Buchmueller, “Fringe Benefits and t he Demand for Part-T i me Worke r s , ”
Applied Economics , vol. 31 ( 1999).
⁴³ Mitchell, The Effects of M andating Benefit Packages.

would i ncreas e t he total compensation of part-timers by a l arger percentage than if
the contribution v aried b y t otal wages o r hours worked. While the costlier approach
w o u l d b e m ore likely t o open u p full-time jobs for i nvoluntarily employed part-
timers, i t also could s o reduce part-time demand that those who want jobs with short
workweek s are unable t o get them . As t he vast majority of part-timers prefer short
schedules, “many of them would b e worse off i f forced to transfer into a permanent
full-time job” or to drop out of the l ab o r fo rce.⁴⁴ Alternativel y, the l es s ex pensive
approach might result in low l evels o f retirement i ncome, little accumulated l eave t o
care for ones e l f an d dependents, or few additional part-time workers with health
i n surance b ecause t h ey coul d not afford t h ei r s hare of t h e p rem i u m .⁴⁵
Im position of a work-bas ed benefit requirement al so could change l abor costs
across groups of workers and af f e ct thei r j ob opportunities i n unintended ways.⁴⁶
Women and older workers are t wo large com ponents o f t he part-time labor force. If,
for ex ampl e , f irms had t o ex t end health benefits to thei r part-time em ployees and
they believe that coverage of women and older part-timers would raise group
premiums, firms might replace them with part-timers thought to be lower risks (e.g.,

16-24yearolds).

Underemploym ent a nd a S kill M i smatch
Involuntary p art-time empl o ym e n t is of concern t o s ome observers not only
because “lost” hours impose a cost o n workers and t heir families i n t erms of forgone
compensation, but also on the economy i n t erms of forgone production o f goods and
services. J ust as unemploym ent i s one measure o f t h e underutilization o f human
resources, s o t oo is involuntary p art-time employment.
The l ong-run i ncrease i n workers supplyi ng fewer hours o f l abor per week than
the y w i s h m eans t hat t he ex tent of underemploym ent o r p artial unemploym ent h as
spread. E ven m ore s o t han i n t he past, t h e n , i t c o u l d b e argued t hat t he official
unemploym ent rate overstates t he degree of tightness in the l abor market . According
to this perspective, there i s m ore room for output and emplo ym ent growth without
accel erat i n g i nfl at i o n t han i s apparent from t he l evel o f t he unem p l o ym ent rat e.
In light of the s carcity of labor that ex isted not too l ong ago, some have
wondered why firms d id not tak e greater advantage o f t hese underutilized workers
and o ffer m ore o f t hem t he full-time hours t hey want. Perhaps t here is something o n
t h e s uppl y side, rather than the dem and s ide, that makes workers involuntarily
em ployed part-time less-than-attractive can didates for full-time jobs. Indeed,
according t o one empirical analys is , t he ex pansion o f t he 1990s produced an
inconsequential decreas e i n i n v oluntary part-time em ployment compared to the
ex pansion o f t he 1980s. The researcher suggests t hat firms might have become more
reluctant t o h ire f r o m t h e p ool of economic part-timers because of its altered

⁴⁴ Rebecca M . Blank, “ Continge nt Work in a Changing Labor Market,” in Freeman and
Gottschalk, Generating J obs , p. 285.
⁴⁵ Mitchell, The Effects of M andating Benefit Packages.
⁴⁶ Ib i d .

composition: “state and l ocal welfare ref orms in the m i d - 1990s disproportionately
increased the s upply o f l ow-skilled femal es who d esired to work more hours” and t he
United S tates ex perienced a “relative s urge in legal and illegal immigration” during
the d ecade; that is to say, those w ho remained involuntarily employed part-time
despite the ex t remely tight labor market that characteriz ed the l ate 1990s “may have
been m o re l i k el y t o possess i n feri or charact eri s t i cs (or fi rm s h ad t h e p ercept i o n t hat
thes e were l ower-quality workers).”⁴⁷
As shown i n Table 4, relatively m ore workers i nvoluntarily employed part-time
have not completed h igh s chool compared to either persons voluntarily working p art-
time or persons in full-time jobs. This “suggests t hat t heir inability to get as m uch
work as they desire may be due t o a l ack of skills rather than simply a l ack of full-⁴⁸
time jobs.” Given t he disparity in skill composition and t h e great er benefit costs
that firms co u l d i n c u r were they to switch workers from part-time to full-time
schedules , e m p loyers might prefer to cope with short-run tightness in the l abor
market by lengthening t he hours o f full-time employees already o n t heir payrolls.⁴⁹
Over t h e l ong run, f i r m s m i ght have part i al l y accom m odat ed any m i s m at ch
between the qualifications of workers and the hei ghtened s kill requirements of a
growing s hare of jobs by favoring t he creation of part-time over full-time long-term
jobs. S ince 1983, when the o ccupational classification s ys tem was revised, h i gh er
skilled j obs (i.e., those requiring some postsecondary education at a minimum) have
recorded the rel ativel y great es t gai ns. Higher s killed j obs have increas ed to the point
where t hey now account for m ore t han one-half o f t otal employment. As s hown i n
Table 4, however, j ust 42% of involuntary p art-time workers possessed at l east s ome
postsecondary schooling i n contrast wi t h 60% of full-time workers. It is thus
possible t hat both s upply and demand have been factors i n t he long-run i ncrease i n
persons involuntarily employed 1-34 hours a week.
The i ncreased supply o f l ower skilled w orkers might ex tend beyond new groups
entering the l abor force as s ugge s t e d above. It h as been hypothesiz ed t hat l ower
skilled men, who in particular faced falling real wage opportunities, added t o t he
supply o f i nvoluntary p art-time wor k e r s : because men d isplaced from h igh-wage
fact ory j obs, for ex am ple, were not readily able to obtai n comparably pai d full-time
positions, t hey opted for part-time em ployment rather than unemploym ent or
withdrawal from t he labor force. The employm ent constraint thus was o n t he wage
side rather than the hours s ide, whic h implies t hat t hese workers m ay have been
misclassified as involuntary p art-timers . Instead of an undersupply o f full-time work

⁴⁷ Mark D. Partridge, “Part-T i me Workers a nd Economic Expansion: Comparing t he 1980s
and 1990s with U.S. State Data,” Papers in Regional Science, vol. 82 ( 2003), p. 65.
⁴⁸ Nardone, Part-Time Employment, p. 283.
⁴⁹ For more i nformation on t he role of benefit c osts in firms’ hours-employment decision,
see CRS Report 97-884, Longer Overtime Hours: The Effect of the Rise i n Benefit Costs,
byLindaLevine.

per s e, t his a n a l ys i s implies t hat t here may h ave b een an undersupply o f “good”
jobs.⁵⁰
Table 4. Percent D istribution o f 25-64 Year Olds Employed
Full-Time and Part-Time b y Educational Attainment, 2002
Full-time Voluntary part- Involuntary part-
Characteristic w orkers time w orkers time w orkers
Tot a l 100.0 100.0 100.0
8^th gr ade or l ess 3.2 3.1 8.9
some high school 5.8 5.9 13.6
high school gr aduate 30.5 30.4 36.2
some college 27.6 30.8 23.9
college gr aduate or more 32.8 29.7 17.4
Men 100.0 100.0 100.0
8^th gr ade or l ess 3.8 5.2 9.4
some high school 6.6 8.1 14.3
high school gr aduate 30.6 27.5 36.4
some college 26.0 27.8 22.7
college gr aduate or more 33.0 31.3 7.3
Women 100.0 100.0 100.0
8^th gr ade or l ess 2.3 2.6 8.4
some high school 4.8 5.4 13.0
high school gr aduate 30.4 31.1 36.0
some college 29.9 31.6 25.1
college gr aduate or more 32.6 29.3 17.5
^Source:^T^a^b^u^lated^b^y^CRS^f^r^o^m^th^e^C^P^S^.

⁵⁰ Alec Leve nson, “Long-Run T r ends in Part-T ime a nd T e mporary Employment: T oward
an Understanding,” in Davi d Neuma rk, e d., On the J ob: Is Long-Term Employment a Thing
of the Past?, ( NY: Russell Sage Foundation, 2000).