Flexible schedules and shift work: Replacing the '9-to-5' workday?

Traditionally, much of the American labor force has worked in a structured environment, with the work schedule following a set pattern-what many people have termed the "9-to-5" workday. Recent studies show that employers are beginning to recognize that many workers prefer schedules that allow greater flexibility in choosing the times they begin and end their workday. Consequently, increasing numbers and proportions of full-time workers in the United States are able to opt for flexible work hours, allowing workers to vary the actual times they arrive and leave the work place. For some workers, however, the nature of their jobs requires that they work a schedule other than a regular day shift, what may be termed an "alternative shift."' Examples of such alternative shift workers are police officers, emergency room physicians, and assembly-line workers at a factory.

In contrast to the increasing proportion of workers with flexible work schedules, the incidence of shift work has not changed since the mid-1980s. If not for the sizable job gains in service occupations, the overall proportion of workers on shift work would have edged down in recent years.

Recent data on flexible work hours and shift work are from information collected in the May 1997 supplement to the Current Population Survey (CPS).2 This article uses that supplement to examine both the incidence and trends in flexible work hours and alternative shift work and, also, the relationship between the jobs in which people work and the prevalence of these digressions from the more traditional "9-to-5" workday. Flexible work schedules

In 1997, more than 25 million workers, or 27.6 percent of all full-time wage and salary workers varied their work hours to some degree. Note that flexible schedule arrangements for many workers are probably informal, as indicated by data from the Bureau of Labor Statistics Employee Benefits Survey (EBS), in which employers provide information about employee access to various types of work-related benefits. The latest Ess data, from 1994-97, show that less than 6 percent of employees have formal flexible work schedule arrangements 3 CPS data show that the proportion of workers on flexible work schedules-either formal or informal-has more than doubled since 1985, when such data were first collected.' The increase in flexible work schedules since then has been widespread across demographic groups. The following tabulation shows the percent of workers, by age and race and Hispanic origin, who work flexible schedules:

Although there has been relatively little difference in the proportions of men and women with flexible schedules during the 1985-97 period, whites have been more likely than blacks or Hispanics to have flexible work schedules. (See table 1.)

Occupations. To some degree, these differences reflect the varying occupational distributions of each of the worker groups. Generally, jobs with higher frequencies of flexible hours are those in which work can be conducted efficiently, regardless of the workers' start and end times. For instance, flexible work hours are most common among workers in executive, administrative, and managerial occupations, and for those in sales occupations-42.4 percent and 41.0 percent, respectively. (See table 2.) The incidence of flexible work hours is lower for groups of workers in occupations in which the nature of the work dictates that it begin and end at set times, for example, nurses, teachers, police, firefighters, and certain manufacturing operations.

As stated, the unique occupational distributions of the various demographic groups affect the overall proportion of workers on flexible work schedules within these respective groups. For example, as can be seen above, flexible work hours are considerably more prevalent among whites than either blacks or Hispanics. At first glance, this is not surprising because whites are most likely to be in managerial and professional specialty occupations, in which flexible hours are most common. Furthermore, blacks and Hispanics are highly represented in the category of operators, fabricators, and laborers. Because of the nature of the work, historically, this category is one that fails to lend itself to the practice of flexible schedules.

Because flexible schedules appear to be closely associated with particular occupations, it is worth investigating whether the recent increases in the proportion of workers with flexible work schedules reflect an increase in employment in occupations with high occurrences of flexible work schedules or an increase in the availability of flexible work hours across occupations. A shift-share analysis was applied to determine the portion of the increase that was due to changes in occupational employment and the portion that was due simply to an increased incidence of flexible work hours. Less than 3 percentage points of the total increase were a result of shifts in occupational employment. This suggests, therefore, that the majority of the increase was spurred by the increased incidence of flexible work schedules within occupations; indeed, this phenomenon occurred in nearly every occupational category.

Race. In order to estimate how much of the difference in the rate of flexible work schedules between blacks and whites is accounted for by differences in occupations, a standardization was performed. This process showed that if blacks had the same occupational distribution as whites (at the most detailed level of occupational classification), then the rate of black workers on flexible work schedules would have been 20.5 percent, instead of 20.1 percent; the difference between the rates for whites and blacks would have been 7.9 percentage points instead of 8.6 percentage points. A similar analysis was performed in which the white rates of flexible work by occupation were applied to the black occupational distribution. Results show that, in each job category, if blacks were as likely as whites to be able to vary hours, then the overall black rate would rise to 24.4 percent, or 4.3 percentage points higher. This would have reduced the overall difference between blacks and whites to 4.3 percentage points. While even at the detailed level there may be differences in jobs held by blacks and whites, these findings suggest that factors other than occupational employment contribute to the disparity in access to flexible schedules.

Industry. To a lesser degree, the prevalence of flexible work schedules also varied by industry. These schedules were more common among private sector employees than among those in the public sector (28.8 percent versus 21.7 percent) in 1997. In the public sector, Federal government employees (34.5 percent) were more likely than their counterparts in State government (29.4 percent) or local government ( 13.1 percent) to have a flexible schedule. The rate for local government workers reflects the fact that local governments provide services that are often rigidly scheduled. More than half of those employed in local governments work in the field of education, in which the nature of the work for most employees prohibits flexibility (only 7.6 percent of workers in education, the largest component of local government employment, could vary work hours). Within private industry, the proportion of workers with flexible schedules was higher in serviceproducing industries (31.7 percent) than in goods-producing industries (23.3 percent), reflecting the more rigid work hours in manufacturing, construction, and mining.

Shift work

Although most workers report usually working between the hours of 6 a.m. and 6 p.m., more than 15 million, or 16.8 percent of all full-time wage and salary workers, worked alternative shifts. The most prevalent alternative shifts were the evening shift (accounting for 4.6 percent of all full-time wage and salary workers), for which work hours typically fall between 2 p.m. and midnight, and irregular shifts (3.9 percent) for which employers schedule shifts to fit the needs of the business for a particular time. Other shifts worked included night shifts (3.5 percent) for which work hours fall between 9 p.m. and 8 a.m., and rotating shifts (2.9 percent) that change periodically from days to evenings or nights. (See table 3.)

As with flexible work schedules, the nature of the work is a major determinant of whether the worker is scheduled on an alternative shift. Hence, shift work is highly prevalent within certain occupations and industries and almost entirely absent from others. Alternative shifts were most common among occupations that provide services that are needed at all hourssuch as protective service (55.1 percent) and food service (42.0 percent)and among those employed as operators, fabricators and laborers (27.0 percent). (See table 4.) In contrast, teachers, construction workers, and executives and administrators were among the least likely to work an alternative shift.

Similarly, the incidence of shift work was much greater among industries providing services used at all hours of the day as opposed to "9-to-5" industries. For instance, about 47.2 percent of the total labor force employed in eating and drinking places worked an alternative shift, as did 35.9 percent in transportation, and 25.8 percent in hospitals. Conversely, shift work was much less common in industries such as finance, insurance, real estate, construction, and agriculture-industries in which most work is done during the daytime.

Some goods-producing industries operate on extended production schedules and therefore had high proportions of workers on alternative shifts. In many of these industries, it is more costly to shut down the production process at the end of the day and restart the next morning than it is to simply operate on extended, and in some cases, around-the-clock production cycles.' Among industries with a high frequency of shift work were paper products (33.3 percent), automobiles (31.3 percent), and mining (24.8 percent).

Shift work occurred less frequently in the public sector than in the private sector, and varied little across Federal, State, and local governments. Within local government, however, the incidence of shift work varies widely by function. Nearly half of the local government employees in justice, public order, and safety functions worked alternative shifts; but only 4.5 percent of those employed in educational services worked an alternative shift.

The CPS supplement included a question intended to derive workers' main reason for working an alternative shift; the results support the notion that the occurrence of shift work is highly correlated with particular industries and occupations 5 More than half of all full-time employees who worked an alternative shift did so because it was the "nature of the job." It is also apparent that very few of these workers chose to work one of these shifts for the purpose of obtaining greater monetary compensation or to alleviate nonwork conflicts. Only 6.1 percent of all alternative shiftworkers reported working a shift for better pay. About 4.1 percent worked an alternative shift for better childcare arrangements; and only a small fraction did so for an easier commute (0.7 percent) or because it allowed time for school (2.9 percent). Roughly 13.0 percent reported that they were on one of these shifts specifically because alternative shifts were mandated by their employer to meet transportation demand, management, or pollution abatement program requirements. A small percentage of shiftworkers (5.7 percent) worked an alternative shift because they were unable to find another job. (See table 5.)

As is the case with differences in flexible work schedules among workers, a portion of the differences among demographic groups in the incidence of shift work can be traced to the occupational distributions of the groups. As indicated in table 2 for example, men were more likely than women to work on an alternative shift: 19.1 percent versus 13.7 percent, respectively; a difference of 5.4 percentage points. A standardization analysis shows that if women had the same occupational distribution as men, then the overall proportion of women on alternative shifts would be 16.3 percent, reducing the difference between men and women to 2.8 percentage points. If the rates of alternative shift work by occupation for men are applied to the occupational distribution of women, then the difference in shift work rates falls to I.5 percentage points. Thus, shift work is more common among men for two reasons: first, men are more likely then women to choose occupations in which shift work is common; and, on the same job, men are typically more likely than women to work an alternative shift.

Among other major groups, workers who had never been married were employed on one of these shifts more often than married workers (21.0 percent versus 14.8 percent, respectively), and a greater proportion of blacks (20.9 percent) worked alternative shifts than either whites ( 16.1 percent) or Hispanics (16.0 percent). Another shift-share analysis shows that only a small proportion of the disparity in alternative shift work between blacks and whites can be explained by different occupational groupings; on the same jobs, it is usually the case that more blacks than whites work an alternative shift. In addition, the incidence of alternative shift work varied to some degree by age: nearly one-third of employed teenagers worked an alternative shift. This is not surprising as daytime school commitments prevent many teenagers from working during normal business hours. The prevalence of shift work declines with age to a low of 14.6 percent for workers aged 45 to 54 years. (See table 3.)

In general, the proportion of workers on alternative shifts has changed very little for all of the major demographic groups over the last 12 years. The following tabulation shows the percent working alternative shifts, 1985-97:

Part-time workers. Alternative shift work was much more common among workers who usually worked part time than among full-time workers. Of the 20.3 million part-time wage and salary workers, roughly 7.3 million, or 36.0 percent, usually worked an alternative shift on their primary job. The majority of these workers usually worked an evening shift or an irregularly scheduled shift. In many cases, part-timers are students, parents, or persons with other daytime commitments that conflict with a regular "9-to-5" schedule.b Another explanation for the high rates of shift work among part-timers is that a sizable proportion of businesses maintain operating hours that extend past the traditional 8-hour day; part-time workers are needed to fill this gap. While the proportion of full-time wage and salary workers who worked alternative shifts was unchanged between May 1991 and May 1997, the proportion of part-timers on alternative shifts fell from 45.6 percent to 36.0 percent over the period.

THE "9-To-5" WORKDAY does not appear to be in jeopardy of fading from its prominence in U.S. workplaces; yet the data do suggest that the rigidity of those hours continues to relax. In May 1997, about one-fourth of all full-time wage and salary workers could vary the times they began or ended work, nearly double the proportion in May 1985. In contrast, the proportions working alternative shifts-something other than a regular daytime shift-have not increased over the period.

Clearly, the prevalence of both flexible work schedules and alternative shifts is linked to the nature of the work involved in a particular job or industry. However, this explains only a portion of the variation in the frequency of these types of work schedules across demographic groups. Even within the most detailed occupational groupings, sizable differences remain, in both the rates of alternative shift work and flexible work hours among the various demographic groups, differences that the available data do not completely explain. 0