...the purpose this article, overemployment occurs when a worker's desired hours of labor supply is exceeded by hours of labor demanded at their current pay rate. This article identifies empirically the demographic and job factors associated with being "overemployed," and the extent one may be willing to reduce hours of work at one's current (or suitable alternative) job for less income. Unlike previous studies of hours constraints, the focus here is less on underemployment--the desire for more hours and income--even though underemployment is more common and may be more adverse to worker welfare. (1) However, overemployment has considerable spillover (hidden) social costs. Facilitating a reduction in overemployment with appropriately targeted policy may potentially reduce the extent of underemployment, at least in sectors and workplaces where they co-exist. (2)

The research for this article relies on analysis of the May 2001 Supplement to the Current Population Survey (CPS). This Supplement queried workers directly (for the first time since a previous CPS in 1985) about their hypothetical choice between more income with more hours, fewer hours for less income, or same hours and income. The empirical findings can be contrasted to previous estimates of the "rate of overemployment" in the United States using the previous CPS or different instruments capturing the presence of "constrained hours." (See exhibit 1.) They also can be used to contrast the volume or rate of overemployment in comparable countries. (3) This article first sets the stage by considering the theoretical labor market and macroeconomic forces determining the overall rate and distribution of overemployment. Then, it discusses measurement issues pertaining to estimating the level of overemployment. Gauging the extent of overemployment has proven to be highly sensitive to survey question wording and range of options presented to respondents. The article then considers whether hours mismatches are widely shared or are more prevalent for certain types of workers. The empirical analyses test the null hypothesis that overemployment is distributed randomly among individuals against the alternative hypothesis that it is attributable entirely to workers' stage in their life cycle vis-a-vis the nature of jobs. There may be microeconomic, macroeconomic, and institutional reasons to expect that overemployment might be disproportionately associated with certain personal characteristics of workers, reflecting life cycle preferences such as being a parent. In addition, to the extent overemployment is also associated with certain occupations and industries, union coverage, longer usual workweeks, or inflexible daily work schedules, employer or workplace constraints may hold sway. There are few previous studies applying the type of disaggregated data needed to explore in depth the divergence of overemployment by workers' specific characteristics and the degree to which it is associated with either a greater or a lesser incidence of underemployment among a given type of worker. The conclusion section explores theoretical reasons why the overall rate has been stable since 1985 and will likely remain so, and derives implications for surveys and research.

Measuring overemployment

The overemployed are workers who state a preference to reduce hours of paid work even if to do this lessens their income. The most germane questions in surveys are those querying the employed if they are willing (but unable) to reduce hours at their current (or comparable) job in exchange for less current or future earnings or pay. Estimates of the rate of overemployment in the United States vary considerably, depending on the type of sample, instrument, wording, and context of the question from which it can be derived. Exhibit 1 summarizes available, recent estimates of the rate of overemployment from studies considering technical aspects of the survey attempting to measure hours preferences and the existence (or size) of a discrepancy with actual hours.

The CPS Supplement yields the lower bound while other surveys yield estimates of overemployment as high as 50 percent in the United States. Generally, in any survey that also presents an alternative option of obtaining higher income, the proportions of respondents indicating a preference for fewer hours are typically lower. (4) On one hand, if workers are presented exclusively with various hours and pay reduction options, the proportions indicating overemployment are higher. This leads some analysts to be skeptical whether workers' stated preferences would become revealed preferences. On the other hand, overemployment may be underestimated if the query provokes implicit assumptions among respondents about the current income foregone, amount and dimensions of hours reduced, and type of gains realized with time off. First, respondent openness to hours reduction is greatest when surveys do not explicitly state any direct tradeoff of lower income. (5) Second, rates will vary inversely with the extent to which respondents inherently believe they are unable in practice to change their own hours toward their truly preferred hours. Workers may perceive that hours reduction is either not permissible (for example, mandatory overtime), infeasible (under established organizational and job norms and rigidities), or penalized (no quality part-time or shorter standard workweek options). Furthermore, surveys find workers' inclination to forgo current income is considerably less than the willingness to forgo future income or raises. (6) Estimates of the proportion overemployed also tend to be greater if individuals are asked to specify how many hours they would have preferred to have worked in a given week, rather than just indicating fewer (or more) hours. Thus, survey questions regarding hours preferences are challenging not only because they are trickier to measure than actual working hours, but because it is often left unclear whether and how workers would get their "preferred" number of hours and whether they implicitly assume they would experience either more than proportional reduction in compensation, such as access to employee benefit coverage or premium pay. (7) Moreover, any preference for fewer hours might be suppressed if a worker anticipates being underemployed or unemployed in the future. (8) Finally, because survey questions do not address the intensity of work, respondents may be interpreting the "work less" question as implying not only less pay but also a consequently greater work pace or effort. Thus, it is likely that estimates of overemployment drawn from CPS-type survey questions may be biased downward, on balance.

Underlying sources of overemployment

The conventional microeconomic model of the labor market suggests labor suppliers sort themselves or are matched into jobs that reflect their preferred work time in the long run. In the interim, they would receive a compensating wage differential. (9) If there were a persistent mismatch between desired and actual hours, even though it may be equilibrium, this is both individually and socially suboptimal. Hours mismatches are created when labor demand-side incentives lead employers to require longer hours than employees prefer in the context of human capital investment (the cost of training and screening or adverse selection), principal-agent, or efficiency wage models. (10) The labor market does not tend to offer "diverse durations" of shift lengths and instead may under-provide short-hour jobs). (11) An overemployment mismatch may exist and persist so long as: (a) employers perceive the costs of adjusting hours downward toward each employee's preference to exceed the benefits; (b) employers underestimate or discount the longer term indirect labor costs (for example, absences, tardiness, turnover, reduced labor productivity) they may incur with worker overemployment; (c) employees lack recourse or bargaining power to impose adverse cost consequences on employers who do not match preferences. (12)

The overall rate of overemployment also has macroeconomic sources, such as cyclical factors. When orders or customers are surging, the demand for hours per worker may rise faster than hours desired induced by rising wages (particularly when income effects dominate substitution effects on labor supply). Also, overemployment may be structural. This may result from skill upgrading or skill shortages and the rising quasi-fixed cost of health insurance contributions, or institutional factors such as deunionization and more noncompliance with Fair Labor Standards Act (FLSA) overtime hours and pay regulations. All of these factors would tend to increase average hours demanded. (13) Finally, frictional overemployment stems from the bundling of hours with pay in most employment contracts and from incomplete information regarding available jobs, hours, and scheduling arrangements and employee preferences. (14)

Hypotheses: overemployment distribution

An employee becomes overemployed when their employer's demand for hours per worker lengthens beyond the supply of hours employers can induce (with working conditions or pay) from employees. Alternatively, it occurs when workers cannot realize a new preference for reduced hours of paid market work because constraints in the workplace, such as minimum hours required to retain or perform a job preclude a commensurate downward adjustment of hours. Various theories of the labor market suggest that overemployment, all else constant, may be more prevalent among certain types of workers (as with unemployment and underemployment). Thus, overemployment (underemployment) is expected to be positively (negatively) associated with:

* Personal characteristics that are associated with relatively shorter preferred time in paid work activity, during certain life cycle stages. (15) This includes times when competing demands on time are greatest, especially in dual earner households, when household production, caregiving, or health needs are at a peak, (16) and personal characteristics associated with access to relatively higher relative wage rates, such as for whites and the higher educated, rather than disadvantaged minorities such as African-Americans and the lesser educated;

* Long average weekly hours, as this is associated with a desire to work less, particularly among full-time dual working spouses where at least one partner wishes to reduce hours; (17)

* High relative earnings per hour, where income effects may be stronger; (18)

* Occupations for which there are no legally required overtime pay premia for increasing hours, such as "exempt" jobs that tend to be paid by salary rather than hourly;

* Occupations and industries with workplaces or jobs that offer incentives that induce longer hours with the promise of future compensation rewards or enhance job security and/or penalize expressing preferences for shorter hours; (19)

* Occupations with structural economic constraints, such as high minimum hours requirements or little autonomy for workers to exert control over their own hours; (20)

* Industries where there is some productivity per worker gained while the additional wage cost is negligible, such as jobs compensated with salary rather than hourly wages; (21)

* Little bargaining leverage among workers to obtain arrangements for adjusting hours downward as needed when their preferences shift, such as younger or nonunion workers, and a paucity of alternative job opportunities; (22)

* Jobs with more flexible working options, such as flexitime scheduling and work at home, to the extent these may help alleviate chronic daily time conflicts associated with long workweeks, or lead workers to reciprocate with greater effort in the form of extra hours. (23)

Descriptive statistics

The descriptive statistics of the key variables used in the May 2001 cPS Supplement sample of more than 57,000 individuals appear in the appendix. (See page 37.) The key question asks: "If [you/name] had a choice [at your main job] would you/he/she prefer to: work fewer hours but earn less money. Work more hours but earn more money. Work the same number of hours and earn the same money?" Because proxy answers for this question were not allowed, just under 43,000 observations were collected. The distribution of hours mismatches by personal and work characteristics appear in tables I through 5. (24)

Table 1 shows that estimates of the overemployment rate using the CPS Supplement question on the willingness to trade income for reduced hours in 2001 was about 7 percent of all employed (7.4 percent among full-time workers), virtually the same as the 7.6 percent rate observed when last measured in 1985. (25) While a far greater proportion is either satisfied with their level of hours or seeks more hours to gain income, the share that is overemployed is not trivial. Indeed, it implies a growth overtime in structural or frictional overemployment, as there was presumably less cyclical overemployment in the midst of the 2001 recession year. The overemployment rates and overemployment ratios (overemployment over underemployment rate) are relatively higher among women and whites than among men and African-Americans. There is a clear pattern by age, with overemployment low among young workers but rising with age. There appears to be more interest in reduced hours in 2001 among the 55-64 age group, than had been in 1985, and somewhat more interest among 65 and older men as well. However, the overall rates by gender exhibited no discernable change over time.

Table 2 shows how the overemployment varies by level of work hours. The pattern by hours level is similar in both 2001 and 1985. Overemployment climbs steadily as hours lengthen, with the exception that overemployment and the overemployment ratio dip somewhat among those working exactly 40 hours, in both 2001 and 1985. A shift seems to have taken place over time where overemployment has become somewhat less disparate by hours. It has become less concentrated among those with very long hours, but is slightly more apparent among those with fewer than 30 hours. Thus, the small decline in overall overemployment rate observed since 1985 has...

NOTE: All illustrations and photos have been removed from this article.



Looking for additional articles?
Search our database of over 3 million articles.

Looking for more in-depth information on this industry?
Search our complete database of Industry & Market reports by text, subject, publication name or publication date.

About Goliath
Whether you're looking for sales prospects, competitive information, company analysis or best practices in managing your organization, Goliath can help you meet your business needs.

Our extensive business information databases empower business professionals with both the breadth and depth of credible, authoritative information they need to support their business goals. Whether it be strategic planning, sales prospecting, company research or defining management best practices - Goliath is your leading source for accurate information.