Working hours: insights from economic theory
At the microeconomic level, the number of hours an individual chooses to work depends on trade-offs between income and leisure.Wages are the most important determinant of hours worked. As wages increase, individuals have an incentive to work more hours, as this leads to a larger increase in their earnings, while leisure time becomes relatively more costly as taking more time off means forgoing larger earnings (substitution effect). At the same time, higher wages improve the income of individuals, who can afford to work less while keeping the same level of income (income effect). Depending on which one of these two effects dominates, hours worked may increase or decrease. Empirical studies have found that the individual labour supply is hump-shaped: for lower wages, the substitution effect predominates, while for higher wages, the income effect comes gradually to the forefront (see for example Blundell, 1992; Sharif, 2000; Tavor et al., 2022). According to empirical studies, there can be gender differences: for men, a 10 % increase in wages decreases the annual hours worked by around 1 %, which means that the income effect is larger than the substitution effect (Borjas, 2020). For women, on the other hand, a 10 % increase in wages increases the annual hours worked by around 2 %, meaning that in their case the substitution effect dominates, and the reaction of working time to wages is stronger than for men. A part of this difference can be due to the lower labour market participation of women overall and the tendency for women to work fewer hours on average . Gender differences have been on the decline with the increasing labour force participation of women and the declining gender gap in working time.
In addition to this simple theoretical framework, the decision on hours worked is influenced by a broad range of other factors.A rise in non-wage income, or in the income of other household members, can lead to a decline in hours worked through the income effect . Individuals with strong preferences for leisure would choose to work fewer hours even at lower wage rates, while others may have strong preferences for income and longer hours. Preferences for leisure are influenced by cultural and social factors, such as age, gender and income. The choice of the number of hours worked by individuals also involves decisions about childcare and household production, which are determined by the family situation. The choice of working time also has a dynamic element: people can decide to work more when their wages are higher and defer leisure to periods with lower wages (e.g., to old age). Labour market regulations and public policies (such as minimum wage laws, taxation policies, employment protection legislation, policies promoting female labour force participation and family policies) can significantly influence the choices of individuals.
Constraints on the choice of hours can lead to working time mismatches for workers.This can manifest as both overemployment and underemployment. In the case of underemployment, workers would be willing to work more at their given wage. For example, they can be involuntarily in part-timework and unable to find jobs with longer or full-time hours . In the case of overemployment, workers would prefer to work less for their given hourly wage. For example, employers may expect workers to work longer than their preferred hours, as it may be cheaper for them to impose overtime than to hire new workers. Mismatches between actual and preferred hours of work can result from rigidities in the labour market (contracts only offered for certain types of hours, for example for full-time work and for specific types of part-time hours). Mismatches can also result from asymmetry in the bargaining power of employers and employees, or the monopsony power of employers (Manning, 2003; Cahuc and Zylberberg, 2008; Hermann, 2014; Visser, 2013). Section 3.4.3. will look in more detail at the patterns of overemployment and underemployment in the EU.
In the long run, technological developments and the substitution between labour and capital can influence working time developments.In general, new technologies may both complement and replace labour, with uneven impacts between skills groups, sectors and occupations. Increasing labour scarcity, due to the decline in the working age population or to labour and skills shortages, can increase the substitution of labour by machinery and automation, which would boost productivity (Acemoglu and Restrepo, 2017a, 2017b, 2019; Zeira, 1998). However, this shift could have unexpected and hard-to-predict effects on employment: on the one hand, technological improvements in production processes increase efficiency, reduce costs and increase productivity so that more can be produced with the same labour input; on the other hand, technological innovations in consumer goods provide novel benefits, create new needs and change consumption patterns (Bessen, 2018; Vercherand, 2014) (). Technological change may be accompanied by innovations that lead to expansions in consumer demand, thus contributing to employment. Demand could also expand for labour-intensive services. Overall, the net effect of technological change on employment, hours worked, and the wages of different groups of workers remains uncertain. Even if the overall employment effect will be positive, rapid technological progress is likely to lead to significant restructuring in the labour market.
Working time policies can aim to tackle both over- and underemployment.The main goals of working time policies have historically been to promote occupational health and safety and to cater to workers’ preferences for more leisure time. Thus, they have been focused on addressing overemployment and its negative consequences. Yet they have been also addressing underemployment, which is an issue in the current context of labour shortages and shrinking working-age population, in particular with the regulation of part-time work, aiming to ensure its equal treatment in terms of employment rights, social protection and working conditions. Thus, more broadly, working time policies can narrow the gap between actual and preferred hours. The rigidities in the labour market that prevent voluntary reductions in working hours could be addressed by statutory limits on working hours and by overtime regulations. Working time policies can support the underemployed by broadening the options available to individuals and firms to adjust working hours without administrative barriers, while enforcing equal treatment between part-time and full-time employees, and effectively addressing the potential social risks of this flexibilization (Eurofound, 2020). Namely, they should prevent the increase in precarious jobs with limited access to social protection and employment rights. Moreover, policies to address labour shortages could reduce the perceived overemployment of workers in shortage occupations by promoting an increased participation of workers with perceived underemployment. This could be promoted by facilitating transitions in the labour market, with coherent policy packages, . with an emphasis on training tools. Box 3.1 provides an overview of different approaches to the reorganisation and reduction of working time.
Box 3.1: Working time reorganisation and reduction
Approaches to reorganising or reducing working time typically take one of the following forms .
- Compression of working time without reducing it. The work week is shortened by extending daily working hours on the days worked .
- Shortening the hours worked during the work week. An option of this approach can be to shorten the work week without reducing wages, but with the same output requirements for workers, thus requiring an increase in workers’ hourly productivity .
- Tightening overtime regulations and increasing overtime pay. This policy aims to reduce (very) long work hours.
- Legislative downward liberalisation. This is the elimination of administrative rigidities that constrain businesses and workers from choosing shorter (daily or weekly) working times .
- Legislative mandatory reduction of working time. This involves a reduction of the maximum number of (daily/weekly) working hours for regular full-time work.
Some of these approaches can be implemented as government policies. Moreover, working time reduction can be implemented through collective bargaining or the voluntary actions of economic actors, within the constraints set out in labour market regulations.
Some observers advocate for policies to reduce working time, based on a range of arguments.They argue that productivity improvements due to technological change can allow for working time reduction and that shorter hours can improve labour productivity per hour. They call for investigating policies that can contribute to ensuring that technological progress and artificial intelligence have fair distributional effects on working time. Moreover, proponents of working time reduction argue that a shorter working week can increase employment and reduce unemployment. Further arguments in favour of working time reduction have been to promote individual and social well-being, work-life balance and gender equality, to reduce the mismatch between actual and preferred hours worked and increase the attractiveness of companies (see Section 3.5.).
Yet, the economic and social impacts of working time reduction are uncertain.Economic theory suggests that the employment effects of working time reduction are not so clear-cut, as they depend on multiple factors (see Box 3.2). In the current context of labour shortages, working time flexibility or reorganisation has also been proposed as a tool to attract and retain employees. At the same time, labour shortages across the EU signal a need to increase the overall labour supply, which may be at odds with working time reductions. Demographic developments also carry important economic and social risks, since a shrinking working-age population due to population ageing puts pressure on labour markets and welfare states, increases the old-age dependency ratio and raises the per-capita burden of public debt.
Box 3.2: Employment impacts of working time reduction
Reductions in working time can have both positive and negative impacts on employment, with an ambiguous net effect. From a theoretical point of view, the employment impact of working time reduction depends on the economic model used.
- In standard competitive models of the labour market, the potential positive impact of working time reduction on employment is explained by the work-sharing argument. However, this is counteracted by a potential decrease in labour demand, driven by increasing wages and declines in productivity, via the following channels: (i) hourly wages may increase to ensure that the income of workers does not decline, (ii) due to the fixed costs of employment (such as the costs of hiring, training and employment taxes), wages per unit of time can effectively increase, (iii) labour productivity may fall if non-productive time remains constant while hours are reduced and (iv) capital utilisation may decrease if not only working hours, but also operating hours are reduced (Boeri and Van Ours, 2008; Calmforms and Hoel, 1988). Employment can also decline if labour is replaced by capital in parallel with working time reduction. In general, in the long run, the impact of working time reduction will depend on its impact on productivity and labour costs (Cahuc and Zylberberg, 2007). In this model, a mandatory reduction in hours with full wage compensation is not expected to increase employment . To increase employment, hourly or weekly wages must be decreased. If the productivity increases due to the reduction in hours are small, large drops in wages are required to maintain employment.
- In models with imperfect competition on the labour market and monopsony power of employers, small reductions in working hours (if they do not go below the competitive levels) induce an increase in employment, while larger reductions reduce employment (Cahuc and Zylberberg, 2007; Manning, 2001; Marimon and Zilibotti, 2000). In these models, without a ceiling on hours, employers tend to offer employment at lower wages and with longer hours than in the fully competitive situation. The regulation of working hours needs to be accompanied by a minimum wage to provide workers with the competitive outcome. However, a common mandatory ceiling on working hours does not take into account the diversity of people and cannot systematically improve employment and welfare .
Working time reduction is often advocated for as a way to bring about an increase of (headcount) employment. However, this idea is based on a false assumption. The expectation behind this argument is that if employed people work less, the remaining labour can be redistributed to unemployed people, improving the welfare of everyone concerned. This has been the idea behind ‘work sharing’, advocated for by trade unions in the 1980s and in the 1990s in Europe. However, this argument is subject to the ‘lump of labour’ fallacy, as it does not consider the broader impacts of working time reduction on wages, productivity and labour demand. .
Moreover, for some companies, working time reductions may reduce their ability to effectively organise their services or production.