Printer Friendly

Work-time reduction in the U.S. and Western Europe.

Available data indicate that while the United States once pioneered in providing reduced working time for workers, achieving a 40-hour workweek well in advance of most other industrial nations, Western Europe has now caught up and passed the United States in this respect.

A number of reasons-none completely satisfactorycan be offered for the different work-time patterns in the postwar period. It has often been said that the European taste for leisure is greater than the American. The longer hours of work by Europeans before the war appeared to contradict that stereotype, but this was misleading because Europeans also had much lower incomes than Americans during this period, and high income has been found to be positively related to demand for leisure. In the intervening years, hourly earnings have risen rapidly in Europe, largely eliminating the income gap. A strong preference for leisure would be consistent with the Europeans' taking a large part of their earnings gain in the form of increased leisure.

Some support for the view that American workers are not ready to trade income for reduced hours was provided in a recent survey of employed Americans by the Bureau of Labor Statistics, which found that only 8 percent would be willing to decrease their hours of work if the change were accompanied by a proportionate reduction in earnings.

A second plausible explanation emphasizes institutional differences. High marginal tax rate policies in Europe may tend to discourage Labor supply there. Other welfare state policies may also have this effect, if less directly. For example, it has been argued that work-time reduction in the United States has been slowed because Americans have instead chosen to increase outlays for education and for pensions, so that they can both enter the Labor force later and leave it earlier.' A possible explanation of this is that American workers feel that their ability to provide for either their own retirement or the education of their children is more significantly dependent on their own earnings. While similar changes in Labor force participation rates have also occurred in West ern Europe, employees there have not been deterred from also seeking reductions in annual work time. The relatively greater role of the state in Europe in subsidizing education, retirement income, and health care somewhat reduces the pressures on the individual to work for pay.

Another difference is the greater influence of society (in the form of strong, politically oriented trade unions as well as state legislation) in Europe in directly determining work schedules. For example, many European countries, unlike the United States, have laws that provide for minimum vacation time or which set a maximum level for overtime work.

Finally, reduction in hours has, since at least the late 1970's, been regarded by Europeans as a way of sharing scarce employment opportunities, and this social concern has placed work-time reduction at the top of the bargaining agenda in a number of industries in Western European countries. Some observers place a major emphasis on this factor. But whatever the reason, it is clear that in one important respect-reduced working time-the United States no longer leads the industrialized world. U.S. working time

In this article, two sources of hours of work data are used. Hours paid data are collected from the BLS survey of establishments, while data on hours worked are obtained from the Current Population Survey (CPS, the household survey provided by the Bureau of the Census for the BLS). Establishment survey data indicate that hours paid on the average job have continued to decline in the postwar years in the United States. (See table 1.) True, the average workweek in the traditional core has shown very little reduction. The manufacturing workweek has been virtually unchanged, and only small changes have occurred in mining, contract construction, transportation, and public utilities. However, workweeks have been reduced substantially in the service sector, broadly defined. Work time in the wholesale and retail trade industries fell by about 9 hours, from more than 40 to fewer than 32 hours a week. More moderate but still substantial declines are also seen in the other service sector industries: finance, insurance, and real estate, and other services. Moreover, because employment in this service sector grew relative to the industrial core, the lower level of hours in the other services industries also contributed to a decline in the average of all hours.

The difference in hours between the industrial core and the service sector appears to be largely attributable to differences in the use of part-time employees. The establishment data do not distinguish between part- and full-time employment, but other data sources indicate that when those persons working fewer than 35 hours a week are excluded, there is little or no difference between average hours in the two sectors.

However, these employer-based establishment data may both underestimate the number of hours worked by the average employee and exaggerate the decline over time in his or her hours.' This can be seen by a comparison of establishment survey data with CPS data. The CPS indicates that about 6 percent of employees work two jobs. Hence, data on hours worked at the average job underestimate the number of hours worked by the average employee.

Moreover, the household survey data indicate that measured reductions in the average workweek have reflected large-scale changes in the composition of the work force; as a result, the average workweek does not provide a good measure of changes in the hours of the groups that compose the work force. The household data provide rich detail on the age, sex, student status, and other demographic characteristics of the work force and show wide diversity among groups: Employed adult men work about 7 hours more a week than do employed women, and employed students on average work about 15 fewer hours than does the average woman. 7 The household data also show how changes in the composition of the Labor force have influenced movements over time in the statistical average of hours. Nonstudent men dominated the Labor force 40 years ago, but now only account for about onehalf The declining proportion of nonstudent men in the work force has tended to reduce the statistical average of working time, quite apart from any changes that may have occurred in the schedules of any of the groups that compose the Labor force. In fact, as the following tabulation shows, when the hours of work of nonstudent men are isolated, one finds virtually no downward movement in the past four decades: their weekly hours were 42.7 in 1948 and 42.8 in 1986:

Such changes in the composition of the Labor force make it more difficult to assess the implications of reductions in the statistically average workweek for changes in employee work time. Interpretation was simpler in the century ending in 1950, when the Labor force was more homogeneous.

Then, reduction in hours did mean significant increases in time off for the average employee. But when the statistical average of hours is reduced today because of the entry into part-time employment of women and students who may have other types of responsibilities, the effects are more ambiguous. The series on working hours of nonstudent men presented here probably come closest to a measure of changes over time in individual work times because this group has a minimum of other types of responsibilities. And this series has remained stationary.

Breaks, vacations, and holidays

It can also be argued that all workweek data tend to overestimate work. In the first place, scheduled work time includes break time, and it has been estimated that the average American worker takes 45 minutes a day in coffee and other breaks. Some fragmentary evidence suggests that these breaks may have increased over time."

In the second place, even the CPS data on weekly hours worked do not properly reflect changes in vacation and holiday time which have grown sharply over the past 40 years. Adjustment of weekly hours data for changes in vacations and holidays provides a more meaningful measure of working time.

Unfortunately, vacation and holiday times are extremely difficult to measure in the United States. The BLS does publish data on vacations and holidays in its nationwide sample of employee benefits in medium and large firms. Data from the benefits survey indicate that high seniority workers receive generous vacation provisions after 20 years of service, the typical employee obtains about 4 weeks of paid vacation." However, the average employee receives somewhat less paid vacation, because he or she typically has much less seniority -the average worker, as of the time of the BLS survey, had been on his or her job for about 4 years.' 3 The average firm in the survey gives less than 9 days of vacation after 1year, and 2 weeks after 3 years of service. However, many workers are employed in small firms (which were not included in the survey) and these often offer less vacation and holiday time.

Another measurement problem ariser, because vacation time provided is not the same as vacation time taken. In the benefits survey, on the one Sixteen percent of the plan participants were allowed to cash in unused vacation time." 14 On the other hand, some employees may take unpaid vacation time, which is not included in the survey.

Estimates of vacation and holiday time actually taken can be obtained from the CPS. Each month the survey asks about vacation and holiday time taken during the week of the 12th and an annual average of these estimates can be constructed from these data." Unfortunately, this estimate is also imperfect, because the survey week contains no major holidays. As a result, the survey measure of holiday time is a gross underestimate and is not used here. The vacation time estimate is useful, but requires adjustment: in an early study by the BLS, Peter Henle showed that the avoidance of holiday weeks results in an undercount of vacation time because workers often take their vacations in conjunction with major holidays and suggested that this requires that the CPS vacation data be increased by a correction factor of 20 percent.

According to these CPS data, the average nonagricultural employee enjoyed 9.5 days of vacation time in 1985. "When the Henle correction factor of 20 percent is applied, estimated vacation time rises to 11.4 days.

Estimates of holidays as well as vacations are needed to adjust weekly hours for annual time off. A measure of paid holidays is provided by the BLS survey of medium and large firms: an average of 10.1 holidays per year. This is probably too high an estimate because the survey includes "Extended holiday plans, such as the Christmas-New Year's Day period provided in the auto industry"" and some of this extended holiday time may also be counted in the vacation time estimate of the CPS. Hence, it should not be added here as an adjustment to it. Moreover, the exclusion of small firms very likely imposes an upward bias on this measure of holidays. Noting that 35 percent of these medium and large firms offer 9 fewer days off as holidays, a figure of 9 days might be indicated as a more reasonable estimate of annual holiday time. This yields an estimate of about 20 days per year of time taken in vacations and holidays.

This calculation of vacation and holiday time can be confirmed by an alternative method. Here one begins with the BLS hours at work surveyof establishments (which does include small as well as medium and large firms). This survey offers data on hours paid for, but not worked. From this estimate, one substracts an estimate (from another BLS survey) of paid absences for illness, injury, and miscellaneous reasons 20 to obtain an estimate of paid time for vacations and holidays. Finally, an estimate of total vacation time can be obtained by using the ratio of total to paid vacation time provided by the CPS data mentioned above. This alternative method also yields an estimate of 20 days of vacations and holidays.

Twenty days of vacation and holiday time clearly represent a major improvement over conditions during the early postwar period. The CPS series on full weeks of vacation taken, available since 1947, shows an increase of 77 percent over a 39-year period. The following tabulation illustrates an index of full-week vacation time of nonagricultural wage and salary employees (1947= l00):

A similar series on annual holidays is lacking, but if the amount of holidays rose at the same rate as vacation days, then one would say that total vacation and holiday time increased from about 11 days a year in 1947 to about 20 days a year in 1986.

While this indicates a rapid rate of growth, it does not mean that increases in vacations and holidays have been sufficient to offset the effects of the near-leveling off in weekly hours. When translated into hours per week (assuming an 8-hour day and a 52-week year), the average vacation time would rise from 1.8 hours per week in 1947 to 3.3 hours in 1986. A change of 1.5 hours -less than 4 percent of a 40-hour week -over a period of four decades does not represent a very impressive gain. This can be seen in the following tabulation, in which the data for weekly hours of work for nonstudent men (text tabulation on page 42) are adjusted for vacations and holidays (text tabulation on page 43). Very little net change is observed from 1950 to 1986:

Working times in Western Europe

A comparison of work time changes in Western European nations with those in the United States shows major differences. Work times in Europe continued to decline at a fairly steady rate throughout the postwar period. Forty years ago, the typical European worker had a 6-day workweek of close to 48 hours. He now has a scheduled workweek of 40 hours or, in an increasing number of industries, of less than 40 hours. He also enjoys an annual vacation of more than 1 month. These gains are reflected in the available statistics on work time. The following tabulation gives percentage change in hours of work in manufacturing for nine European countries (weekly hours, adjusted for increases in annual vacations and holidays). Substantial reductions-ranging from 14 to 30 percentare observed in each of these nine European countries. Comparable data on annual hours in manufacturing in the United States are not available, but data on average weekly hours paid actually show an increase during this period. And, as was noted earlier, downward adjustment for increases in vacations and holidays in the United States would yield only a modest decline in the hours estimate.

West Germany provides more complete data on working times than do most countries. The available data show a sharp decline in the hours paid to industrial wage workers in Germany from the early 1950's, from 49 to 40.5 hours for men and from 48 to 40.2 hours for all workers. This reflected a workweek reduction from six 8-hour days to five 8-hour days for most workers.

More detailed data are available since 1960 on all German employees (nonindustrial as well as industrial). 2 6 Between 1960 and 1970, the normal workweek (that is, the agreed upon, or standard, workweek for full-time employees) fell from 5.5 days to 5 days. Further reductions in the 1980's have pushed the standard workweek below the 40-hour level for many Germans. Forty-three percent of German workers on full-time schedules now have a 38.5-hour workweek and 2 percent have less than that. 21

The actual workweek was also reduced by decreases in average weekly overtime -from 3.6 hours in 1970 to 1.8 hours in 1985. Increases in vacations and holidays have also played a major role in reducing working time in Germany. By 1960, these stood at more than 25 days per year, but they have since risen further to 43 days per year (about 30 days of scheduled vacation time and 13 annual holidays).

In fact, the major difference between working time in the United States and Germany today is the longer vacation and holiday benefits enjoyed by German workers. In Germany, not only are generous benefits provided for more seniority workers than would be the case in the United States, but they are offered to new employees as well. As a result, the average German employee now has an annual schedule that is close to being the equivalent of an 8-hour, 4-day week. Indeed, if the employee's 10 sickness days per year are added to 43 vacation and holiday days, we obtain a total of 53 days per year, or approximately one day a week. Given a standard 5 -day workweek, this additional time off each year yields the equivalent of a 4-day workweek.
COPYRIGHT 1988 U.S. Bureau of Labor Statistics
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1988 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Owen, John D.
Publication:Monthly Labor Review
Date:Dec 1, 1988
Previous Article:BLS to produce monthly indexes of export and import prices.
Next Article:Westinghouse's settlement deviates from pattern.

Related Articles
A global market for US food products.
How our cars are cleaning up their act; DRIVETIME.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters