May 24, 2007 | EPI Briefing Paper #189
The Work-Family Balance
An Analysis of European, Japanese, and U.S. Work-Time Policies
by Janet C. Gornick, Alexandra Heron, and Ross Eisenbrey
During the past 25 years, the workforce participation of women has increased substantially throughout the industrialized countries that make up the Organization for Economic Cooperation and Development (OECD). In the United States, for example, the employment rate of women increased from 47.5% in 1979 to 56.0% in 2004. Currently, in most OECD countries, the majority of couples with children have both parents in the workforce; single mothers' employment rates generally exceed those of married mothers. As a result, parents throughout these countries are struggling to balance the demands of employment with the needs of their families. Policy makers have been forced to tackle a range of problems: insufficient parental time spent with their children; excessive stress on working parents; gender imbalances in the workplace and in caregiving; and financial burdens imposed by the high cost of quality child care.
There is tremendous variation across countries in the ways in which social and labor market policies have been adjusted to accommodate the influx of mothers, especially with respect to the regulation of work hours. But one thing is clear: the United States has taken a very different path than many other high-income countries, with very different results.
Policy reforms outside the United States aimed at reducing work time appear to have had an effect. Average work hours in almost every European nation have fallen dramatically since 1979 (Mishel et al. 2006, Table 8.7). Even in Japan, known throughout the world for its long work hours, average work hours declined by over 300 hours a year. By contrast, the United States has not implemented or even seriously debated policies designed to reduce work time. Instead, most work-family advocates have focused on the need for child care, paid family leave, and programs that permit flexibility in determining which, rather than how many hours workers will spend on the job. Average annual work hours in the United States have changed very little since 1979 and now exceed even Japan's.
In this briefing paper, we analyze the regulation of work time in 10 countries, including the most common measures taken to achieve three over-arching goals: first, to reduce the full-time work week to less than 40 hours; second, to guarantee workers an adequate number of paid days, annually, away from the workplace; and third, to raise the quality and availability of part-time work.1
For U.S. workers, achieving the first two goals would, in effect, put a ceiling on annual hours worked, even among those with strong labor market ties. That would help to standardize the definition of full-time/full-year employment at a level that allows working parents to secure adequate time at home. Achieving the third goal would enable parents to choose part-time work, for short or long periods, without a disproportionate loss of compensation.
The 10 countries we analyze span the leading economies of the world—eight European Union (EU) countries, Japan, and the United States. We focus on these countries because there is extensive variation among them in both work time policies and in work-time outcomes—such as annual and weekly hours actually worked, and the availability and quality of part-time work.
Work-time policies in Europe, Japan, and the United States
Work-time policies, and efforts to reform them, operate in diverse institutional frameworks. The institutional backdrops of our comparison are summarized in Table 1. As indicated, in six of these EU countries, work time is typically governed by a combination of labor law and collective agreements, while in France and Italy, labor law traditionally dominates.
Coverage rates of collective bargaining are 60% to 80% in Germany, Italy, Luxembourg, and the Netherlands and 90% or higher in Belgium, France, and Sweden. The United Kingdom is set apart by the limited reach of collective bargaining; the coverage rate is about 30%, less than half that of most of the continental countries. In Europe, diversity in policy-setting mechanisms is supported at the supranational level. The EU Directives relating to working conditions allow member countries to implement required practices through legislation, formalized agreements among the social partners (groups representing employers and workers), or some combination of the two.
By contrast, in Japan and the United States, only about one worker in seven is covered by a collective agreement. Not surprisingly, work-time measures in these two countries are largely determined by labor law. And, clearly, individual agreements between employers and employees are also important for many workers—especially in the United Kingdom, Japan, and the United States, where collective coverage is no where near the norm.
One of the most powerful mechanisms for shaping work time is the establishment of a normal (or standard) full-time work week (see Table 2). Normal weekly hours generally refers to the threshold above which overtime becomes payable. Some EU countries establish normal weekly hours through legislation and collective agreement, while others regulate maximum hours (generally set at an average of 48) but leave the setting of normal hours exclusively to the bargaining table.2 In the continental EU countries included here, the normal full-time work week, for at least a substantial majority of workers, is set by collective agreements below 40 hours—35 in France and between 37 and 39 in the other countries. In the United Kingdom, an outlier among EU countries, there is no statutory normal work week3 and, while collective agreements, on average, set the week at about 37 hours, only a third of the UK labor force is covered. Both Japan and the United States set normal hours, via legislation, at 40 hours, above the standard typical in most EU countries—and a full five hours per week above the French standard. In the United States, any effects associated with the comparatively long standard week are compounded by the limited reach of the Fair Labor Standards Act (FLSA). The FLSA excludes many workers, including managers, professionals, and those over specified earnings limits, from its requirement that overtime is paid after 40 hours of weekly work; approximately 27% of full-time workers in the United States are exempt. Certain state overtime laws, most notably California's, provide broader and more protective coverage than the FLSA.
Vacation and holiday leave
In addition to setting weekly hours, countries effectively set the normal number of days worked per year—meaning that work-time policies define the meaning of not just full-time work, but full-year work as well. The full year is defined, in practice, by the establishment of paid vacation and holiday entitlements (see Table 3). As with normal hours, vacation entitlements are embedded in diverse institutional frameworks. In these European countries, a statutory minimum exists and collective agreements typically raise that minimum for many covered workers (Table 3, columns 1 and 2).
In Europe, some homogeneity is imposed by the EU Working Time Directive, which requires "that every worker is entitled to paid annual leave of at least four weeks in accordance with the conditions for entitlement to, and granting of, such leave laid down by national legislation and/or practice" (European Communities 1993). In practice, workers in these EU countries are typically entitled to between 25 and 33 days per year of paid vacation (or about five or six weeks)—meaning that full-year work corresponds to approximately 46 to 47 weeks a year.
Again, workers in Japan and, even more so, in the United States work under substantially different rules. Japanese workers are entitled by law to 10 vacation days after six months of continuous service, increasing with length of service to a maximum of 20 days—thus workers with long tenures are entitled to vacation time nearing European levels. While no data are available on average collective agreements, Japanese workers are entitled, in practice, to about 18 days of paid vacation each year. In the United States, national legislation is silent with respect to vacation days, and collective bargaining reaches only a small share of workers. It is difficult to determine average vacation entitlements in the United States. According to the National Compensation Survey, U.S. employees with 10 years of service with an employer average 16 days of paid vacation per year, and employees with less tenure are entitled to much less (Mishel et al. 2006).
The number of days that workers are permitted—and expected—to be away from work is also shaped by the establishment of public holidays. Public holiday laws vary widely across countries and, in some cases, employers can limit workers' rights to take off holidays and/or to be paid for them. For example, EU citizens generally have a statutory right to public holidays. However, in some member states—including France, Sweden and especially the United King-dom—some employers may require employees to work on these days or to take them as part of annual holiday entitlements (Mercer 2003). In the United States, the federal government designates 10 public holidays and, in addition, some employers observe state and local holidays. However, many employers reserve the right to schedule employees to work on holidays, although employers are required by law to allow workers to observe religious holidays consistent with their beliefs and practices.
While holidays also increase workers' time outside of work, sometimes substantially, they are generally less advantageous than vacation days, as workers typically have no control over when they can take them. When vacation and holiday entitlements are summed (see Table 3, column 3), we see that workers in the EU countries are granted from 28 annual days off (in the Netherlands and the UK) to as many as 36 days (in France). Japanese workers' entitlement, at 25 days, is well above the U.S. outcome (10 days), due to both the Japanese vacation statute and the larger number of public holidays.
A third set of work-time measures complement those that directly influence work hours and days: policies that aim to raise the quality of part-time work and those that grant various rights to work part time. Measures intended to raise the quality of part-time work include, first, requiring pay and benefits parity between part-time and full-time workers, and, second, enabling workers to shift from full-time to part-time work without being forced to change jobs. Measures advocating the right to work part time are also intended to raise the availability of part-time work for full-time workers who wish to reduce their hours and, depending on the law, to encourage new labor market entrants who might other-wise refrain from employment. Measures that raise the availability of shorter-hour employment could, in turn, affect labor force participation rates. Many women—mothers especially—choose non-employment because no high-quality reduced-hour work is available (see Clarkberg and Moen (2001)).
Policies aimed at improving part-time work are widespread throughout Europe. A crucial force behind these measures is the 1997 EU Directive on Part-Time Work, whose official purpose was "to eliminate discrimination against part-time workers and to improve the quality of part-time work" (Europa 2004). All eight of the EU countries in this study have implemented the Directive via some mix of legislation and collective agreements. The Directive requires that member states enact measures prohibiting employers from treating part-time workers less favorably than "comparable full-time workers," unless they demonstrate that this is objectively justifiable. The national measures address various combinations of pay equity, social security and occupational benefits, training and promotion opportunities, and bargaining rights. In contrast, although Japan enacted a law aimed at the effective utilization of part-time workers' skills, Japanese law provides no pay and benefit parity protection. With the exception of coverage under the national minimum wage law, U.S. labor law is entirely silent on part-time workers' remuneration.
The Part-Time Directive also urged, but did not require, member states to eliminate obstacles that limit opportunities for part-time work and instructed employers to "give consideration" to workers who request transfers between part-time and full-time work as their personal and family needs change (Europa 2004). Long before the Part-Time Directive, Sweden had already set the gold standard on the right to part-time work. Since 1978, Swedish parents have had the right to work six hours a day (at pro rata pay) until their children turn eight. After the Directive, other European countries added new protections. Germany now grants the right to work part-time to employees in enterprises with more than 15 workers; the Netherlands enacted a similar right in enterprises of 10 of more workers. Belgium grants employees the right to work 80% time for five years. In most cases, employers have a safety valve; they can refuse a change on business grounds, but those grounds are often subject to official review. A recent UK law grants employees in enterprises of any size the legal right to request flexible work time—including part-time work—in order to care for a child under age six or a disabled child under age 18. The employer has seven different grounds on which to refuse an application and must give reasons for such a refusal. Italy and Luxembourg join Japan and the United States in granting workers no particular legal rights to seek part-time work.
Does policy matter? International variation in hours worked
How many hours per year do workers actually work in these 10 countries? Figure A presents an OECD estimate of annual hours worked in 2004 in our comparison countries, except for Luxembourg.
As the figure indicates, there is substantial variation within Europe. Annual hours are 1,530 or fewer in Belgium, France, Germany, and the Netherlands, just under 1,600 in Sweden and Italy, and substantially higher in the United Kingdom. Workers in Japan and the United States work the longest hours, logging about 1,800 hours annually, about one-third more than the average Dutch worker and more than 10% more than Swedish and Italian workers.
Undoubtedly, work hours vary sharply across these countries. Is this variation demonstrably shaped by the policy variation presented in the previous section? Clearly, there is an association between the institutional frameworks and the policies, on the one hand, and actual hours worked on the other. That association is most evident when we contrast the continental European countries as a group with the United Kingdom, Japan, and the United States. In the continental European countries, collective bargaining coverage is substantially higher than in the other three countries and policies go further in limiting weekly hours, capping annual days, and enabling and protecting part-time workers. Not surprisingly, then, workers in these countries work fewer hours than their British counterparts and even fewer than their Japanese and American counterparts.
Yet, an association does not establish causality, and it is possible that other factors matter more than these policies. Perhaps American and Japanese workers, including parents, simply want to work more hours than do most Europeans.4 It is also possible that other structural factors are more important than these direct measures. Bell and Freeman (2001), for example, attribute Americans' relatively long hours to higher levels of wage dispersion such that an extra hour worked has a higher return in the United States than elsewhere. Prescott (2004) argues similarly that lower taxation rates in the United States motivate comparatively longer hours, as returns to additional hours are greater.
Research on work time indicates that the policies reported in the previous section in fact do matter—especially the regulation of normal and maximum hours. Several empirical studies assess the effects of normal-hour thresholds, and they all find evidence that lowering overtime-pay thresholds reduces actual work time among employees (see OECD (1998) and Gornick and Meyers (2003) for reviews). A number of studies have estimated the magnitude of the effect of reducing regulated standard hours on actual hours worked. Estimates of the magnitude of the effect range from about 75% to nearly 100% of the change in standard work hours. Researchers have reported the effect on actual hours to be about 77% in the United Kingdom; 85% to 100% in Germany; and close to 100% in France (see Gornick and Meyers (2003) for a review of this research).
Although maximum hours have received less attention in empirical research, they too seem to have a strong effect on actual hours worked. Grubb and Wells (1993), for example, assessed the effects of restrictions on overtime hours. They found that, across Europe, maximum limits on annual overtime hours—which ranged from under 100 to over 500 hours per year—tended to reduce the frequency of overtime work.
In addition, the limited evidence that exists also indicates that the generosity of vacation entitlements has a strong effect on the days per year actually worked—with the possible exception of Japan, where average vacation take-up is substantially less than that allotted (Carley 2003). Like European workers, American workers take up their rights at relatively high levels; one recent study found that about 70% of employed Americans take all of their allocated vacation days (Expedia 2004).
The effects of legislation on part-time work rates and/or part-time workers' remuneration are not well known, in part because the EU Part-Time Directive and the national measures that followed were implemented only recently; several outcome evaluations are underway. There are correlational findings that link regulation to the availability or quality of part-time work; for example, more protective regulations are seen in countries with larger part-time labor markets and smaller pay penalties. One recent study, for example, finds that part-time/full-time wage differentials in Germany, the United Kingdom, and especially in Sweden are substantially smaller than those reported in the United States, where part-time workers' compensation is not protected by law (Bardasi and Gornick 2002). However, virtually no research persuasively establishes a causal link.
Lessons for U.S. policy
The costs and benefits of the several kinds of work hours regulation described above and the shorter work time they have created for the average worker have not been fully evaluated. On the one hand, increased time away from work obviously makes it easier to manage the various tasks of parenting. A recent OECD study using data from a 19-country survey verifies that work-family conflict is higher among those with longer work hours. And fully 58% of French parents report that the 35-hour law has made family care easier for them (Fagnani and Letablier 2004).
On the other hand, many of the recipients of shorter hours have gained them at the cost of more non-standard work scheduling and in many cases, of reduced control and predictability. EU law explicitly permits "annualized hours" schemes that calculate work hours up to the 48-hour weekly maximum over a period of four months; collective negotiation can lengthen the reference period by up to 12 months. This can permit an employer to schedule a worker for very long hours for weeks at a time, as long as they are balanced out with much shorter weeks at other times. Work-family conflict is higher when daily hours vary, work days per week vary, starting or finishing times vary, if schedules change with little or no notice, or if workers have little control over their work hours. Not surprisingly, the French parents with fluctuating or nonstandard schedules are much less satisfied with their statutory 35-hour work week than those with steady, standard hours.
Thus, one lesson from Europe regarding U.S. proposals to amend the FLSA to create compensatory time and "flex-time" schedules is clear. If employers get new scheduling flexibility to get around the standard 40-hour work week, protective mechanisms for workers—such as enforceable minimum notification periods and/or time-bank arrangements that divide control over scheduling between employers and workers—will be crucial. Otherwise, the advantages for parents of shorter work weeks and more time off may be more than offset by the disadvantages of problematic scheduling practices.
Policy recommendations for the regulation of work time
In practice, work-time regulations are distinct from family leave and child care policies in that they typically apply to all workers, rather than being selectively designed for parents. There are some well-known exceptions—such as the Swedish Child Care Leave Act, which grants parents of young children an absolute right to work part time, and the British Right to Request Flexible Working law, which guarantees parents the right to request a schedule change. But, in most cases, work-time measures regulate and protect all employees without regard to family structure—partly for political reasons, partly for ease of implementation. Thus, we envision a package of work-time measures that increase parents' options for high-quality reduced-hour work, but that are not limited to parents.
First, work-time measures should limit weekly employment hours, setting normal full-time weekly hours in the range of 35 to 39 hours per week—as is standard in several European countries today. Limiting the standard full-time week to below 40 hours would grant parents more time for children on a daily basis. Limiting men's time in the labor market, in particular, would raise the likelihood of more gender-equitable time allocations between partners. Implementing reductions economy-wide would increase parents' opportunities to seek employment that is "full time" but at less than 40 hours, across a broad range of firms, occupations, and industries. Overtime regulations should both offer compensation for those who work longer hours and protect workers against compulsory overtime at excessively long hours.
Second, policies for paid time off should assure workers a substantial number of paid days off each year. Public measures should grant workers at least one month of paid time off annually; in practice, that means that the normal work year would be defined as 48 weeks of work per year. The right to paid time off of at least one month per year would alleviate some of the burden of arranging child care coverage during summer school breaks and would grant parents needed periods of uninterrupted family time.
Third, part-time workers should have the right to pay and benefit parity—in comparison to full-time workers per-forming similar work in the same enterprise. Such protections, aimed at preventing discrimination against part-time workers, could be modeled after the provisions required across Europe in the wake of the 1997 EU Directive on Part-Time Work. Improving the quality of part-time work would increase economic security for part-time workers and their families, and provide incentives for more men to participate in part-time employment.
Fourth, all workers should have the right to formally request a shift to reduced-hour or flexibly-scheduled work, subject to employer agreement. As they do in several European countries, employers would have the right to refuse "on business grounds," but their refusals would be subject to government review. To accommodate the needs of small employers, these general rights to work-hour changes could be restricted to workers in enterprises of a minimum size, but the minimum should be set relatively low.
In conclusion, policies that reduce total employment hours, and raise the availability and quality of part-time work, are crucial components of work-family reconciliation policy packages in many countries. The positive experiences in several EU member countries provide models for policies that, if implemented in the United States, would greatly benefit most employees, especially working parents.
— Janet C. Gornick is professor of political science and sociology at the Graduate Center of the City University of New York (CUNY), and professor of political science at Baruch College, in the United States. She is also director of the Luxembourg Income Study, a cross-national research institute and data archive, located in Luxembourg. She is co-au-
thor of Families That Work: Policies for Reconciling Parenthood and Employment (Russell Sage Foundation 2003), a comparative study of work-family reconciliation policies in 12 countries.
— Alexandra Heron has worked on labor law issues for trade unions and in government in Britain and Australia. She is currently based in France and works on international comparisons of discrimination law and work time. Her work has included consultancies for the Organization for Economic Cooperation and Development on labor market issues, and for the United Kingdom Equal Opportunities Commission on pregnancy discrimination.
— Ross Eisenbrey is the vice president and policy director of the Economic Policy Institute.
This Briefing Paper is adapted from an earlier article, "Working Time Regulation as Work-Family Reconciliation Policy: Comparing Europe, Japan, and the United States," published in 2006 in the Journal of Comparative Policy Analysis.
1. A note on terminology: We use the terms "the regulation of work time" and "work-time policies" to refer broadly to national and local legislation, public labor market regulations, and collective agreements that affect a large share of the workforce. We use the term "reduced-hour work" to refer to paid work at less than 40 hours per week. We use the term "part-time work" to mean work that is not considered, by national standards, to be "full time." Across our comparison countries, legal and statistical definitions of part-time work vary. In some, for example, the statistical definition of part-time work is less than 35 hours per week, in others, less than 30 hours per week; in the EU, the legal definition refers to someone whose normal hours of work are less than the normal hours of a comparable full-time worker. Finally, by the EU, we mean the EU-15 prior to the 2004 enlargement.
2. The meaning of maximum hours varies. In most cases, maximum hour policies mean that workers may not work above the set ceiling, while in others workers may not exceed the ceiling unless they opt to do so.
3. This is also the case in Germany, Luxembourg, and the Netherlands, but collective bargaining coverage is much greater in those countries.
4. Evans et al. (2001) report a revealing finding about cross-national variation in work-time preferences. An international survey, in 1994 asked workers if they would prefer a reduction in work hours or an increase in pay. Americans were less likely than Europeans to choose a reduction in work hours. But when no reference was made to the trade-off with earnings, a somewhat higher percentage of U.S. workers than European workers replied that they wished to reduce their hours of work. Americans' preferences for long hours appear to be bound up with concerns about economic security, at least more so than among many of their European counterparts.
Bardasi, E., and Gornick, J. C. 2002. Explaining Cross-National Variation in Part-Time/Full-Time Wage Differentials Among Women. Paper presented at the workshop on Comparative Political Economy of Inequality. Cornell University, Ithaca, New York (April 5–7).
Bell, L., and Freeman, R. B. 2001. The incentive for working hard: explaining hours worked differences in the U.S. and Germany. Labour Economics. Vol. 8, No. 2, pp. 181–202.
Carley, M. 2004. Working Time Developments, 2003. European Industrial Relations Observatory (EIRO). Downloaded on January 18, 2005 from http://www.eiro.eurofound.eu.int/2004/03/update/tn0403104u.html
Clarkberg, M., and Moen, P. 2001. Understanding the time squeeze: married couples preferred and actual work-hour strategies. American Behavioral Scientist. Vol. 44, pp. 1115–36.
Europa. 2004. The Part-Time Work Directive. Downloaded on January 19, 2005 from http://europa.eu.int/scadplus/leg/en/cha/ c10416.htm
European Communities. 1993. Council Directive 93/104/EC, of 23 November, concerning certain aspects of the organization of working time. Downloaded from http://europa.eu.int/smartapi/cgi/sga_doc?smartapi!celexplus!
Evans, J., Lippoldt, D. C., and Marianna, P. 2001. Labour Market and Social Policy: Trends in Working Hours in OECD Countries. Occasional Paper Number 45. Paris: Organization for Economic Cooperation and Development. pp. 17–59.
Fagnani, J., and Letablier, M.-T. 2004. Work and family life balance: the impact of the35-hour laws in France. Work, Employment, and Society. Vol. 10, No. 3, pp. 551–72.
Gornick, J. C., and Meyers, M. K.. 2003. Families That Work: Policies for Reconciling Parenthood and Employment. New York, N.Y.: Russell Sage Foundation.
Grubb, D., and Wells, W. 1993. Employment regulation and patterns of work in EC countries. OECD Economic Studies. Vol. 21 (Winter), pp. 7–58.
Mercer. 2003. Human resource consulting. Downloaded on January 15, 2006 from http://www.mercerhr.com/pressrelease/details. jhtml/dynamic
Mishel, L., Bernstein, J., and Allegretto, S. 2006. The State of Working America: 2006–2007. Washington, D.C.: Economic Policy Institute.
OECD. 1998. "Working hours: latest trends and policy initiatives," Chapter 5 in Employment Outlook. Paris: Organization for Economic Cooperation and Development. pp. 154–188.
Prescott, E. C. 2004. Why Do Americans Work So Much More Than Europeans? NBER Working Paper 10316. Cambridge, Mass: National Bureau of Economic Research.