III. LABOR OVER-SUPPLY

Overemployed Workers in the U.S. Labor Market

Lonnie Golden
      Penn State University

Abstract

     The article estimates the current rate and distribution of "overemployment"--workers expressing a willingness to decrease hours of work at their job even if it means lower income. I explore the theoretical causes determining the level and trend of overemployment and considers measurement issues. The May 2001 Current Population Survey (CPS) Supplement indicates an overemployment rate under 7 percent among full-time workers. However, the rate is measurably higher among women, whites, parents of young children, workers with long workweeks, occupational classifications like manager/administrators, scientists, engineers, and some technicians, and in industries such as health care, utilities, and transportation. I then draw implications for refining Fair Labor Standards Act reforms.

Introduction

    Much of the research and policy discussion in the U.S. concerning trends in working hours and regulation occurs without much reference to how many and which specific types of workers tend to be overemployed, defined as working beyond their preferred number of hours. While such preferences are not observable, restrictions on individuals' choice of hours of work are viewed as central feature of the labor market (Kaufman 1999; Altonji and Oldham 2003). Most models recognize that workers often face binding constraints imposed by employers who set fixed or minimum shift lengths (Gunderson and Weiemair 1988; Contensou and Vranceanu 2000) and an under-provision of short-hour jobs (Rebitzer and Taylor 1995). Labor market institutions, such as the fixed cost of employee benefits, non-compliance with Fair Labor Standard Act (FLSA) overtime regulations, and weakening of labor unions lessen restraints on the hours demanded per worker (see Belman and Belzer 1998; Hamermesh and Trejo 2000; Altman and Golden 2004).

Measuring Overemployment

    Overemployment exists when there are workers employed who are willing but unable to reduce their hours of paid work at their current (or a comparable) job even if they are prepared to accept proportionately lower current or future income. Estimates of the aggregate level of overemployment vary by the type of sample and instrument, since these estimates are highly sensitive to survey question wording and the options that are presented. The preference for fewer hours depends on the implicit assumptions provoked about the foregone income, the dimensions of hours reduced, and the type of time-off gains realized. When questions include an option of obtaining higher income via more hours of work, such as in the May 1985 Current Population and General Social Survey, estimates of overemployment are as low as 6 to 10 percent (Lang and Kahn 2001; Reynolds 2003), but also up to 30 percent (Heldrich Center for Workforce Development 1999). If respondents are presented exclusively with options for reducing hours and pay, the proportion of workers that would give up at least one half day's pay for at least one half day of work less per week (or more free time or family time), ranges from 28 to over 50 percent (Jacobs and Gerson 2001; Feather and Shaw 2000; Schor 1995, 2001; Friedman and Casner-Lotto 2003).

    The May 2001 Current Population Survey (CPS) Supplement, for the first time since 1985, queried if individuals, "given the choice, (would) opt for more income and more hours, less income and fewer hours, or the same income and hours?" Table 1 shows an overemployment rate no higher than the 1985 estimate of 7.6 percent (Shank 1986). There are several reasons why the rate might remain stable. The May 2001 survey was during a recession and 1985 was a period of expansion. In addition, preferred hours may become endogenous. Overemployed workers may upwardly adjust their target income (Altman 2001; Altman and Golden 2004), under loss aversion (Dunn 1996), shifting preferences toward more time-saving, purchased goods and services (Rothschild 1982). Moreover, the climate of rising earnings inequality motivates workers to longer hours as a signaling tactic (Landers et al. 1996; Eastman 1998; Bluestone and Rose 1998; Bell 2000; Brett and Stroh 2002) and positional consumption (Schor 1995). Workplace amenities, rewards, or flexibility, diminish resistance to longer hours (Hochschild 1997; Kaufman 1999; Golden 2003).

Hypotheses

    A worker is overemployed if actual hours (h) exceed desired hours (h*) at their current wage and job:

This gap may occur if some employers are not induced to adjust h downward toward h*. If,
the gap can persist so long as any of these six conditions exist:

1) the administrative costs () to constantly adjust h toward each employee's h* are large;
2) the size of the gap between h and h* is not overly large;
3) long-term risks on the organization of overemployment are small or discounted ( is one);
4) employees lack bargaining leverage in the workplace to impose adverse cost consequences (e.g., absences, tardiness, turnover, or reduced average hourly labor productivity) on employers;
5) employees are induced to discount the cumulative, long-term consequences of overwork;
6) it is less costly for firms to induce employees to adjust h* upward than to reduce actual hours.

Overemployment is hypothesized to be more prevalent among workers:

•who have personal characteristics associated with lower h*, such as parents with children at home;
• with spouses present in the household (see Clarkberg and Moen 2001);
•at jobs whose hours are relatively longer (for either cyclical or structural reasons);
•with bargaining leverage insufficient to adjust their own hours downward when h* shifts;
•in occupations and industries with insufficient job autonomy to exert control over hours; and
•in industries where the added wage cost of overtime is negligible, such as salaried jobs.

Data, Model and Estimation Results

    From the May 2001 CPS sample (over fifty-seven thousand respondents), Tables 1 and 2 display overemployment rates by fifty-two (and forty-six detailed) occupations and industries that are above the overall average of 6.3 percent. Whether an individual reports being willing to reduce hours and income would depend on a worker's personal () as well as job () characteristics, including work hours or shifts, occupation and industry employed in, self-employment or union membership status, and flexibility of their job. The likelihood an individual (i) responds affirmatively to the option of reducing both hours and income is the respective vectors of estimated coefficients, X and Y:


    The model is estimated using probit analysis. The dependent variable is bi-variate, taking on a value of one if an employed individual reports a preference for fewer hours and less income. The coefficients are derivatives of the probit estimates, representing the marginal probabilities that an individual possessing a given personal or job characteristic is overemployed.

     The estimation results in Table 3 show that female workers are more at risk of being overemployed than their male counterparts. Conversely, African American workers are significantly less likely to be overemployed. This finding probably owes to the lower average wage rates of these workers. Being married raises the likelihood of overemployment, relative to the (omitted) reference group of single workers. Having children in the household (relative to having either no or fully-grown children) displays nuanced effects. When the youngest child in the household is younger than three years old, this raises the likelihood of feeling overemployed. Having preschool children aged three through five has a somewhat weaker effect than the presence of younger children. When the youngest child present reaches the age of fourteen, the effect of having children is reversed. This finding suggests that when children are young there is a greater demand for time, but as the child ages, there is a gradual shift toward the desire for more income when the children are teenagers.

    Full-time workers have a progressively higher likelihood of being overemployed the longer are their usual hours. Working from forty-one to forty-nine hours raises the probability of overemployment considerably, relative to those working thirty-five to thirty-nine hours per week (the reference group). Working fifty or more hours per week raises the probability of overemployment still further. Part-time workers (thirty-four or fewer hours) are less apt to be overemployed. Being a union member neither increases nor decreases the chances of overemployment. Being self-employed, perhaps surprisingly, heightens the probability of willingness to trade income for more time. Workers who have some ability to vary their daily and ending times of work have a somewhat reduced likelihood of overemployment if they have informal arrangements as such (Golden 2003), but, also, surprisingly, have a heightened probability of being overemployed if they have formal flextime scheduling. This result suggests that formal flexibility of scheduling (or doing work at home) exacerbates rather than curbs interference of work with family or personal time.

     Many of the major occupational classifications enhancing overemployment fall in the managerial and professional groups, including engineers, health diagnosticians, natural scientists, math/computer scientists, health assessment and treatment professionals, private-sector managers and administrators, lawyers/judges and management-related occupations. Two categories of technician jobs and other administrative support areas have significantly higher levels of overemployment. Several blue collar occupations, notably construction trades and laborers, fabricator/assembler, and service jobs such as personal, health, or food service occupations, are less likely to experience overemployment.

     Some industries heighten the likelihood of overemployment, especially public utilities and hospitals, as well as communications and transportation and, to a weaker extent, wholesale trade and medical services other than hospitals. These results are not surprising given the incidence of mandatory overtime work (involving high profile labor disputes) in these sectors. On the other hand, being employed in social services, construction, agriculture, private household, justice/public order, or stone/glass manufacturing significantly reduces the likelihood of overemployment. Either hours in these sectors are more responsive to workers' preferences for shorter hours, or workers in these sectors are less apt to prefer shorter than actual hours.

     Overemployment occurs among a non-trivial proportion of the employed, especially among certain jobs, sectors, and workers. In the United States, the only institutional restraint on hours of work are the FLSA overtime regulations which apply to those workers who are not exempt due to having primarily managerial, administrative, or professional duties (Appelbaum et al. 2001). If a genuine goal of legislative or regulatory reform is to curb the incidence, duration, and effects of overemployment, the empirical results suggest that this would be accomplished by more precisely targeting the types of workers and jobs more prone to overemployment. Currently pending FLSA "reforms" such as compensatory (comp) time legislation are thus mistargeted. Not surprisingly, by over a three to one ratio, workers paid hourly and covered by the FLSA would prefer pay over time-off as compensation for overtime work (Friedman and Casner-Lotto 2003). Rather, FLSA coverage ought be extended to employees in white collar, salaried occupations, particularly in industries where overemployment is most concentrated, such as telecommunications, hospitals, and transportation, and to workers at vulnerable points of their life cycle when non-market work time becomes most valuable, such as when a worker has a child younger than three years old.

Acknowledgements

     I would like to acknowledge Tesfayi Gebresel and Valen Costello for valuable research assistance and Eileen Appelbaum and Mark Montgomery for comments on segments of this research.

References

Altman, Morris. 2001. "Preferences and Labor Supply: Casting Some Light into the Black
Box of Income-Leisure Choice." Journal of Socio-Economics, Vol. 30, pp. 199-219.

Altman, Morris, and Lonnie Golden. 2004. "Alternative Economic Approaches to Analyzing Hours of Work Regulation and Reform." In Economics: Alternative Economic Approaches to Legal and Regulatory Issues, ed. Margaret Oppenheimer and Nicholas Mercuro. Armonk, N.Y.: M. E. Sharpe.

Altonji, Joseph, and Jennifer Oldham. 2003. "Vacation Laws and Annual Work Hours." Economic Perspectives, Vol. 28, no. 3, pp. 19-29.

Appelbaum, Eileen, Thomas Bailey, Peter Berg, and Arne Kalleberg. 2001. Shared Work--Valued Care: New Norms for Organizing Market Work. Washington, D.C.: Economic Policy Institute.

Bell, L. 2000. "The Incentive to Work Hard." In Working Time: International Trends, Theory and Policy Perspectives, ed. L. Golden and D. Figart. New York: Routledge, pp. 106-26.

Belman, Dale, and Michael Belzer. 1998. "The Regulation of Labor Markets: Balancing the Benefits and Costs of Competition." In Government Regulation of the Employment Relationship, ed. Bruce Kaufman. Champaign, Ill.: Industrial Relations Research Association, pp. 178-219.

Bluestone, B., and S. Rose. 1998. "Macroeconomics of Work Time." Review of Social Economy, Vol. 56, no. 4, pp. 425-41.

Brett, Jeanne, and Linda Stroh. 2003. "Working 61 Plus Hours a Week: Why Do Managers Do It?" Journal of Applied Psychology, Vol. 88, no. 1 (February), pp. 67-78.

Clarkberg, Marin, and Phyllis Moen. 2001. "Understanding the Time-Squeeze: Married Couples' Preferred and Actual Work-Hour Strategies." American Behavioral Scientist, Vol. 44, pp. 1115-36.

Contensou, François, and R. Vranceanu. 2000. Working Time: Theory and Policy Implications. Cheltenham, UK: Edward Elgar.

Dunn, L. F. 1996. "Loss Aversion and Adaptation in the Labor Market: Empirical Indifference Functions and Labor Supply." Review of Economics and Statistics, Vol. 78, no. 3, pp. 441-50.

Eastman, W. 1998. "Working for Position: Men, Women and Managerial Work Hours." Industrial Relations, Vol. 37, pp. 51-66.

Feather, Peter, and Douglass Shaw. 2000. "The Demand for Leisure Time in the Presence of Constrained Work Hours." Economic Inquiry, Vol. 38, no. 4, pp. 651-62.

Friedman, Will, and J. Casner-Lotto. 2003. Time is of the Essence: New Scheduling Options for Unionized Employees. Work in America Institute and Labor Project for Working Families.

Golden, Lonnie. 2003. "The Flexibility Gap: Employee Access to Flexibility in Work Schedules and Location." Report to the Alfred P. Sloan Foundation.

Gunderson, M., and K. Weiemair. 1988. "Labor Market Rigidities: Economic Analysis of Alternative Work Schedules including Overtime Restrictions." In Management Under Differing Labour Market and Employment Systems, G. Dlugo, W. Doron, and K. Weiermair. Berlin: Walter de Gruyter, pp. 153-63.

Hamermesh, Daniel, and Stephen Trejo. 2000. "The Demand for Hours of Labor: Direct Evidence from California." Review of Economics and Statistics, Vol. 82, no. 1 (February), pp. 38-47.
Heldrich Center for Workforce Development. 1999. "Who Will Let the Good Times Roll?: A National Survey on Jobs, the Economy, and Race for President." Work Trends Survey, Vol. 1, p.16.

Jacobs, J., and K. Gerson. 2001. "Who Are the Overworked Americans?" In Working Time: International Trends, Theory, and Policy Perspectives, ed. Lonnie Golden and D. Figart. New York: Routledge, pp. 89-105.

Kaufman, Bruce. 1999. "Expanding the Behavioral Foundations of Labor Economics." Industrial and Labor Relations Review, Vol. 52, pp. 361-92.

Landers, R., J. Rebitzer, and L. Taylor. 1996. "Rat Race Redux: Adverse Selection in the Determination of Work Hours in Law Firms." American Economic Review, Vol. 86, June, pp. 3229-48.

Lang, Kevin, and Shulamit Kahn. 2001. "Hours Constraints: Theory, Evidence and Policy Implications." In Working Time in a Comparative Perspective, Volume 1, ed. G. Wong and G. Picot. Kalamazoo, Mich.: Upjohn Institute for Employment Research.

Rebitzer, J., and L. Taylor. 1995. "Do Labor Markets Provide Enough Short-Hour Jobs? An Analysis of Work Hours and Work Incentives." Economic Inquiry, Vol. 33, April, pp. 257-73.

Reynolds, Jeremy. 2003. "You Can't Always Get the Hours You Want: Mismatches between Actual and Preferred Work Hours in the United States." Social Forces, Vol. 81, no. 4, pp. 1171-99.

Rothschild, K. 1982. "A Note on Some of Economic and Welfare Aspects of Working Time Regulations." Australian Economic Papers, Vol. 21, pp. 214-18.

Schor, J. 1995. "Trading Income for Leisure Time, Is There Public Support for Escaping Work-and-Spend?" In The North the South and the Environment: Ecological Constraints and the Global Economy, ed. V. Bhaskar and Andrew Glyn. United Nations University Press.

------. 2001. "The Triple Imperative: Global Ecology, Poverty and Worktime Reduction." Berkeley Journal of Sociology, Vol. 45, pp. 2-17.

Shank, S. 1986. "Preferred Hours of Work and Corresponding Earnings." Monthly Labor Review, Vol. 109, pp. 40-44.