The Effect of a Reduced Statutory Workweek on Familial Long-Term Care in Korea

Abstract

Objectives: We examine how statutory workweeks affect workers’ provision of long-term care for their non-coresident elderly parents. Method: The Korean government reduced its statutory workweek from 44 to 40 hr, gradually from larger to smaller establishments, between 2004 and 2011. Using multiple regressions, we assess how the reduction affected visits, financial transfers, and in-kind transfers to parents. Annual longitudinal data come from the 2005 to 2013 waves of the Korea Labor and Income Panel Study. Results: The reduction caused an increase in the frequency of visits and in-kind transfers among male workers, with no significant impact on their financial transfers. Among female workers, we found no impact on any outcomes. Discussion: We interpret the findings within the context of developed Asian countries with long work hours and Confucian traditions, and suggest regulating workweeks as a policy tool to encourage familial long-term care in the rapidly aging societies.

Keywords

statutory work hours elder care time support financial transfers in-kind support Korea

Introduction

Although the population is aging in many parts of the world, the pace of this demographic change is much faster in rapidly developed Asian countries, including South Korea (Korea hereafter), Singapore, Taiwan, and Hong Kong. While it took, or is expected to take, 115 years for France, 73 years for Australia, and 69 years for the United States for the proportion of people aged 65 or older to double from 7% to 14% (Kinsella & Phillips, 2005), the same transition is estimated to occur within 19 years for Korea, 17 years for Singapore, 25 years for Taiwan, and 32 years for Hong Kong (Lee, Mason, & Park, 2011). In the Asian societies, adult children have played an important role in providing care for their elderly parents. In particular, due to Confucian ideologies such as the patriarchal family system and filial piety, the eldest son and his wife, who have often lived together with parents, have tended to bear most elder care burden (Lin et al., 2003; Shi, 2009). However, there has been a steep decline in the traditional living arrangement in recent decades (Yeung & Cheung, 2015). For example, in Korea, the percentage of people aged older than 65 years who lived alone skyrocketed from 4.8% in 1989 to 21.5% in 2010 (Cheong et al., 2012).

With the decline in coresidence with adult children, care provided by non-coresident children has become critical. This is particularly true in the region because utilization of nursing homes remains low, in part, due to a cultural context in which institutionalizing elderly parents is to some extent considered as children’s neglect of their filial duties (Glass et al., 2010; Kao & Stuifbergen, 1999). Along these lines, we define familial long-term care broadly as support for non-coresident elderly parents of various types, including time support, financial transfers, and in-kind transfers. Thus, elder care in the current study goes beyond its narrow definition, which tends to refer to the care of elderly people with functional limitations. In particular, time support involves time spent on care as well as noncare activities, regardless of whether the parents have any functional limitations.

All these major types of familial care are known to make substantial contributions to parents’ well-being. Spending time with adult children improves the psychological well-being of elderly parents (e.g., Cicerelli, 1990; Do & Malhotra, 2012; Lawton, Silverstein, & Bengtson, 1994), and helps children identify and meet parents’ needs (e.g., Hogan, Eggebeen, & Clogg, 1993; Lye, 1996; Rossi & Rossi, 1990). As for financial transfers, for example, seven out of 10 Koreans aged 65 and older received financial transfers from their non-coresident children, and the financial support consisted of about one fourth of their income (E. H.-W. Kim & Cook, 2011). In urban China, 24.6% and 34.1% of people aged 50 or more years received financial transfers and in-kind transfers, respectively. For both types of care, about 80% of the recipients reported that adult children were the main provider, and about 90% said the amount of the transfers was enough (Sun, 2002).

In Western countries, regulating working times through statutory workweeks has been suggested as a policy tool to enable workers to fulfill more family duties, including child care and elder care (Gornick & Heron, 2006; Maume, 2015). A statutory workweek may affect familial long-term care by influencing workers’ time constraints (e.g., Goux, Maurin, & Petrongolo, 2014; Kawaguchi, Lee, & Hamermesh, 2013), physical strains (Bannai & Tamakoshi, 2014; Caruso, 2006), and psychological strains (Artazcoz, Cortès, Escribà-Agüir, Cascant, & Villegas, 2009; Takeuchi, Rahman, Ishiguro, & Nomura, 2014; Virtanen et al., 2011). If a statutory workweek impacts wage rates (Estevão & Sa, 2006; H. R. Kim & Lee, 2012), it might influence types of care workers provide (Becker, 1965). Despite these possibilities, evidence is lacking and mixed with regard to how statutory workweeks affect workers’ provision of long-term care for elderly parents. It remains a particularly interesting empirical question in the developed Asian countries, which tend to have the longest work hours among developed nations in the world (Feenstra, Inklaar, & Timmer, 2015; Organisation for Economic Co-operation and Development [OECD], 2016).

To fill the void in the literature, we examine how a reduction in the statutory workweek from 44 to 40 hr in Korea affected workers’ long-term care for their non-coresident elderly parents. We analyze the impacts on visits, financial transfers, and in-kind transfers, using the 2005 to 2013 waves of the Korea Labor and Income Panel Study (KLIPS).

A Theoretical Framework and Empirical Evidence

How Statutory Workweeks May Affect Familial Long-Term Care

This section conceptualizes a theoretical framework to understand how statutory workweeks may have an impact on familial long-term care. We look into three plausible mechanisms: (a) time constraints, (b) physical and psychological constraints, and (c) wage rates, and relate these mechanisms to the impact on major care outcomes. Relevant theory and empirical evidence are reviewed.

First, given 24 hr in a day, long work hours constrain workers’ time in the family sphere (Bianchi & Milkie, 2010; Greenhaus & Beutell, 1985; Jacobs & Gerson, 2001). Empirical evidence has shown that reduced statutory workweeks in several countries did reduce workers’ actual working time, thereby relieving workers of the time constraints on nonwork activities. For example, in Korea, due to the 4-hr reduction in the statutory workweek, which is assessed in the current study also, actual workweeks declined by 43 minutes when male and female workers were analyzed together (H. R. Kim & Lee, 2012) and by 93 minutes when only male workers were analyzed (E. H.-W. Kim, Lee, & Do, 2018). The Japanese and French governments also shortened statutory workweeks, and employees in these countries did work fewer hours as a result (see Estevão & Sa, 2006; Goux et al., 2014; Saffer & Lamiraud, 2012, for France; and Kawaguchi et al., 2013, for Japan).

Among various types of familial long-term care, a decrease in actual work hours owing to reduced statutory workweeks may have the most immediate implications for workers’ time support. Relevant evidence is lacking and mixed. Saffer and Lamiraud (2012) failed to find any impact of the reduced statutory workweek in France on the number of workers’ visits to relatives, measured as a single item, including visits to parents, children, and other relatives. In contrast, Hsu (2002) showed that people who took advantage of the Two-day Weekend System in Taiwan improved their relationships with family members.

There is a relatively larger body of literature on the relation between adult children’s care provision and their actual, rather than statutory, workweeks. However, available evidence needs to be interpreted with caution due to endogeneity in the actual working time associated with reverse causality and omitted variable bias. Studies which pay particular attention to circumventing such problems are all from Western countries, and they tend to find that employment status or work hours have a weak effect at best on workers’ care provision (Carmichael, Charles, & Hulme, 2010; Stern, 1995; van Putten, Vlasblom, Dykstra, & Schippers, 2010).

Several studies, all of which controlled for household economic resources, are notable, as they investigated the relationship of employment not only with time support but also with financial transfers at the same time. The studies show mixed results. In the United States, while Boaz, Hu, and Ye (1999) found that at least one spouse being employed was negatively associated with time support but not with financial transfers, Chesley and Poppie (2009) found that work hours had a positive relation with financial transfers but no relation with time support. In an analysis of 10 countries in Europe, Bonsang (2007) showed that, compared with unemployed children, employed ones were more likely to provide financial support but spent fewer hours helping parents.

Second, reduced statutory workweeks may increase the familial provision of long-term care by relieving workers’ physical and psychological constraints. Strains produced by long work hours, which may include both physical strains (such as fatigue) and psychological strains (such as tension, anxiety, and depression), may make it difficult for workers to provide long-term care (Bartolomé & Evans, 1979; Greenhaus & Beutell, 1985; Pleck, Staines, & Lang, 1980). Reduced statutory workweeks may lead workers to provide more time support and more in-kind transfers if their physical and psychological constraints are a major barrier to providing the care. In contrast, financial transfers that workers can wire might not be affected as much, since the transfers are probably bounded by workers’ economics resources.

There is considerable empirical support that working times produce physical and psychological strains. Studies have found that long work hours lead to chronic tiredness, general fatigue, and fatigue-related health outcomes, such as diabetes mellitus, metabolic syndrome, and sleep conditions (Bannai & Tamakoshi, 2014; Caruso, 2006). Long hours of work are shown to lower overall mental health and raise the risk of depressive and anxiety symptoms (Artazcoz et al., 2009; Virtanen et al., 2011). Takeuchi et al. (2014) document that, among men, work hours are significantly correlated even with age-adjusted rates of suicide. As for evidence specific to statutory workweeks, Taiwanese people reported that the Two-Day Weekend System relieved workers’ fatigue and psychological stress (Hsu, 2002). Decreased working time due to reduced workweek policies led workers to exercise more often in Korea (Ahn, 2016) and in France (Berniell, 2012), and spend more time on personal maintenance, which included sleeping and eating, in Japan (Kawaguchi et al., 2013).

Third, statutory workweeks may affect familial elder care through wage rates. Since overtime work beyond a statutory workweek is supposed to be paid at higher premium, if a reduction in a statutory workweek increases overtime hours, then hourly wage rates tend to go up. Evidence has shown that reduced statutory workweeks did increase wage rates in Korea (H. R. Kim & Lee, 2012) and in France (Estevão & Sa, 2006). Changes in hourly wage rates, subsequently, may have implications for the types of elder care adult children provide: Intuitively, higher wage rates mean higher opportunity costs of nonworking time, thereby leading workers to spend more hours on work and fewer hours on nonwork activities. Accordingly, workers are predicted to provide less time-intensive care (such as time support) and more money-intensive care (such as financial transfers; Becker, 1965). In-kind transfers might increase if the major barrier to their provision is economic resources, while the transfers might decrease if the barrier is the time and effort needed to purchase and bring gifts.

Although the evidence is not entirely consistent (e.g., Zhan & Montgomery, 2003), findings from previous empirical studies are largely consistent with these predictions. Higher wage rates are shown to reduce time support (e.g., Chesley & Poppie, 2009; Couch, Daly, & Wolf, 1999; Saffer, 2008; Sarkisian & Gerstel, 2004), and increase financial transfers (e.g., Couch et al., 1999; Sloan, Zhang, & Wang, 2002; Zissimopoulos, 2001). In-kind transfers are also shown to be positively associated with wage rates (e.g., Shuey & Hardy, 2003; Sun, 2002).

Context of the Current Study

Work-Related Cultural Context in Korea and Other Developed Asian Countries

Among developed countries in the world, those in Asia tend to have the longest work hours. An average Korean worker spent 2,124 hr on work in 2014, which was the second longest among OECD countries after 2,228 hr in Mexico (OECD, 2016). Among non-OECD countries in Asia, average annual working times in 2011 were 2,287 hr in Singapore, 2,343 hr in Hong Kong, and 2,144 hr in Taiwan (Feenstra et al., 2015). In comparison, an average German worker had the shortest working time of 1,366 hr, and the OECD average was 1,763 hr.

Unlike the situation in most other OECD countries, employed women work almost as long as employed men in Korea: Among the working age population, in 2011, 85% of the latter and 72% of the former worked 40 hr or more a week (OECD, 2013). At the same time, Korean women’s participation in the labor force falls much below that of their male counterparts: The female labor force participation rate in 2010 was 54.5%, with one of the highest gender gaps, of 22.6 percentage points (%p; OECD, 2012). This contrast is associated with the heavy burden of elder care and child care, put mostly on women’s shoulders regardless of their employment status, which makes employed women double-burdened with work and family (E. H.-W. Kim & Cheung, 2015).

A study by Wharton and Blair-Loy (2006) well reflects the conflict between long work hours and a family-oriented elder care culture in the rapidly developed Asian countries. In their comparison of professionals in the United States, England, and Hong Kong offices of the same company, the authors showed that workers in Hong Kong were most concerned about long work hours, and attributed it to a greater filial obligation that Hong Kongese workers felt to their elderly parents.

A Reduction in the Statutory Workweek in Korea

Between 2004 and 2011, the Korean government lowered its statutory workweek from 44 to 40 hr for establishments with full-time wage workers to improve the quality of life of workers, boost leisure and leisure-related industries, and create new job opportunities (Korean Ministry of Labor, 2007). The law was gradually applied, spreading from larger to smaller establishments. The change began in July 2004 for establishments which had at least 1,000 employees and was completed in July 2011 for establishments which had five to 20 employees (see columns 1 and 2 in Table 1 for the detailed schedule of the policy change). As for exceptions, the reduction began in July 2004 for establishments in the public and financial sectors regardless of the number of employees. Establishments with fewer than five employees were exempt from workweek policies.

Table 1.

When Workers Became Subject to the 40-Hr Statutory Workweek and the Distribution of the Analysis Sample in 2005 by the Number of Employees.

When workers became subject to the 40-hr statutory workweek	Number of employees at establishments	Distribution of workers in the analysis sample in 2005 (%)
July 1, 2004	1,000+ employees	25.1
July 1, 2005	300-999 employees	9.8
July 1, 2006	100-299 employees	10.1
July 1, 2007	50-99 employees	9.5
July 1, 2008	20-49 employees	12.8
July 1, 2011	5-19 employees	19.3
Not applicable	1-4 employees	13.2
Total (Number of observations)	—	100.0 (1,006)

Source. Authors’ own calculation using the 2005 KLIPS.

Note. The analysis sample is restricted to those who were aged 18 to 64 years and were full-time wage workers (working 35 hr or more per week) in 2005, and had at least one parent who was alive and non-coresident at each wave. The “1,000+ employees” category additionally includes establishments that are categorized as being in the public or financial sector. KLIPS = Korea Labor and Income Panel Study.

The 40-hr workweek policy may affect not only hours worked but also days worked, since the majority of the workers tended to become free from the half-day work of 4 hr on Saturdays. Hourly wage rates might be affected also. The maximum overtime of 12 hr a week and the overtime premium of 150% of regular payment remained the same after the policy change (Korean Ministry of Labor, 2007), while overtime hours went up. As noted earlier, the reduction in actual workweeks as a result of the 40-hr workweek policy turned out to be much less than the legislated 4 hr (E. H.-W. Kim et al., 2018; H. R. Kim & Lee, 2012).

Method

Data and Analysis Sample

We use data from the KLIPS, which was collected by the Korea Labor Institute in 1998. The KLIPS is an annual longitudinal survey of nationally representative Korean households and their members (Korea Labor Institute, 2015). The KLIPS surveys long-term care of various types, including visits, financial transfers, and in-kind transfers workers provide to their elderly parents. The data set enables researchers to determine whether a worker was subject to the reduced 40-hr workweek policy in a given year, since it contains detailed information on the number of employees at the establishments to which workers belonged, the sectors of the establishments, and the exact dates of the interviews conducted each year.

We construct our analytic sample as follows. We restrict the sample to full-time wage workers who were 18 to 64 years old in 2005. In this year, the KLIPS started asking if respondents’ parents were alive. Hence, we take 2005 as the baseline year and analyze 9 years of data until 2013. According to the International Labour Organisation (ILO) criterion, we define full-time workers as those who work at least 35 hr per week (van Bastelaer, Lemaître, & Marianna, 1997). To examine the impact on care for non-coresident parents, the sample consists of workers who had at least one non-coresident parent at each wave. In other words, we exclude respondents who coresided with parents or had no living parents. Although there is a possibility that the 40-hr statutory workweek may have had an impact on adult children’s coresidence status with their elderly parents, our analysis confirmed no impact on the status (results not shown but available upon request).

The 2005 wave included 2,283 full-time wage workers who were aged 18 to 64 years, and seven out of 10 of the workers (70.1%, 1,600 persons) had a parent who did not live with them, highlighting the importance of care provided by non-coresident working children. We dropped those who did not participate in any of the later eight waves, after which the baseline wave was composed of 1,456 persons, with an attrition rate of 9.0%. Appending the eight waves to the baseline wave produced pooled data with 6,254 person-years. Finally, after excluding 516 person-years (8.3%) that had missing values on variables used in the analyses, the final sample included 5,738 person-years (or 1,006 persons in 2005). Table 1 shows the distribution of workers by the years in which they became subject to the 40-hr statutory workweek.

Key Independent Variable

The key independent variable is a binary variable, which indicates whether or not a worker was subject to the reduced 40-hr statutory workweek. Specifically, the variable is equal to 1 if a worker was subject to the 40-hr statutory workweek, and 0 if he or she was subject to the 44-hr statutory workweek, or was not subject to a statutory workweek (that is, the respondent was either a full-time wage worker at establishments with fewer than five employees in any given year or no longer a full-time wage worker in a year after 2005). We determine the value of the variable in each year based on the number of employees at the establishment, the sector categorization of the establishment, and the exact date of the interview, all reported in the KLIPS, and the schedule of the 40-hr workweek policy, shown in Table 1.

Outcome Variables

We examine three outcome variables on long-term care. First, as a measure of time support, we analyze how often workers visited parents in a year. The regression analyses take the log of the frequency because its distribution is skewed. The second outcome variable is financial transfers to parents. We analyze whether the amount was positive with a binary variable and, if so, how much the amount was for workers who had a positive amount. Since the distribution of the amount is skewed, we take log of the amount. Because the KLIPS surveys financial transfers during the past 12 months, we analyze financial transfers from wave t + 1, in relation to the independent variable from wave t. Thus, for financial transfers, we use waves up to 2014, not 2013. The third outcome variable is in-kind transfers, ranging from trivial daily goods to luxury goods. We analyze a binary variable which is equal to 1 if a worker made any in-kind transfers in a given year. The KLIPS started to ask about in-kind transfers in 2006, so we analyze the transfers over the 8 years, excluding 2005.

Research Design

To estimate the effect of the reduced workweek policy on workers’ provision of long-term care, we conduct multivariate regression analysis. We use ordinary least squares (OLS) regressions for two continuous dependent variables, that is, the logged frequency of visits and the logged amount of financial transfers, and use logit regressions for two binary dependent variables, that is, whether workers made any financial transfers and any in-kind transfers, respectively. The equation below shows the estimation model when the dependent variable is the logged frequency of visits. For all the regressions, we pool individual worker’s observations over waves and cluster standard errors within an individual. In the equation, i indexes each worker and t indexes each year. The key independent variable is 40-hr statutory workweek, which equals 1 if worker i’s statutory workweek in year t is 40 hr, and 0 otherwise. Then the slope coefficient, β₁, captures the effect of the 4-hr reduction in the statutory workweek.

\begin{array}{l} \log {(the number of visits a year)}_{i, t} = β_{0} + β_{1} 40 - hr statutory {workweek}_{i, t} \\ + β_{2} Y e a r F E s_{t} + β_{3} I n d i v i d u a l F E s_{i} + β_{4} X_{i, t} + ε_{i, t} . \end{array}

We include various covariates, denoted by a vector X. Above all, when workers became subject to the 40-hr statutory workweek depended on the number of employees and the sector. To capture heterogeneity among workers in terms of the characteristics, we control for the number of employees in groups (1-4 employees, 5-19, 20-99, 100-999, 1,000 or more) and whether the establishment was in the public or the finance sector. We control for additional work-related characteristics, such as whether a worker was a manual worker and whether a worker was a shift worker.

As common determinants of elder care provision (see, for example, Bonsang, 2007, and Davey, Janke, & Savla, 2005), we include respondents’ marital status and educational attainment, and parental education and parental survival (categorized as “father only,” “mother only,” or “both”). For married respondents, additional controls include their spouse’s characteristics: educational attainment and parental survival and living arrangements (categorized as “no parent alive,” “parent alive and coresident,” and “parent alive and non-coresident,” which is further subdivided into “father only,” “mother only,” and “both”).

To capture year-specific trends in the dependent variables, in all models we include year fixed effects (FEs), denoted by Year FEs, a vector consisting of dummy variables for each year. To eliminate bias related to unobservable individual characteristics that do not vary over time, we add individual FEs, denoted by Individual FEs, a vector consisting of dummy variables for each individual.

While we present summary statistics of the following time-invariant variables for a descriptive purpose, they are dropped from the FE regressions: the respondents’ sex, age in 2005, sibling composition in 2005 (whether a respondent had one or more elder brothers, elder sisters, younger brothers, or younger sisters), and spouse’s age (in the form of a spouse’s age subtracted from the respondent’s age). Although we first analyze male and female workers together, we also conduct subgroup analysis to see whether effects vary by sex.

Results

Descriptive Results

Table 2 shows summary statistics for the variables in the current study. In the pooled sample over years, the mean and median values of the frequency of visits were 33.9 times and 12.0 times a year, respectively, with a standard deviation (SD) of 64.15. About three quarters (74.7%) of observations had a positive amount of financial transfers. Among observations with the positive transfers, the mean amount was 1,205,594 Korean Won (with 1,000 Korean Won approximating US$1) and the median value was 500,000 Korean Won with a SD of 1,989,288 Korean Won. As for in-kind transfers, 35.4% of observations recorded a positive amount. A total of 56.5% of the observations were subject to the reduced statutory workweek.

Table 2.

Descriptive Statistics of Key Variables (N = 5,738).

Dependent variables	Mean (SD)
Number of visits a year	33.89 (64.15)
Any financial transfers	0.747
Amount of financial transfers if amount > 0 (in 10,000 Korean Won ≈ US$10)	120.56 (198.93)
Any in-kind transfers	0.354
Independent variables	Mean (SD)
Whether a worker was subject to the 40-hr statutory workweek	0.565
Male	0.729
Age in 2005: 18~29 / 30~39 / 40~49 / 50~64	0.117 / 0.537 / 0.291 / 0.055
Education: <High school/ high school / ⩾college	0.103 / 0.414 / 0.483
Number of employees at establishments:
1000+ / 300-999 / 100-299 /	0.234 / 0.092 / 0.101 /
50~99 / 20~49 / 5~19 / 1~4	0.080 / 0.155 / 0.154 / 0.095
Public sector	0.002
Financial sector	0.057
Manual worker	0.065
Shift worker	0.115
Employment status:
Full-time wage worker / part-time or unemployed / others	0.911 / 0.009 / 0.080
Parental survival: Both / father only / mother only	0.600 / 0.041 / 0.359
Father’s education:
No formal education / <high school /high school / ⩾college	0.135 / 0.558 / 0.225 / 0.082
Mother’s education:
No formal education / <high school /high school / ⩾college	0.282 / 0.599 / 0.109 / 0.010
Any older brother in 2005	0.526
Any younger brother in 2005	0.518
Any older sister in 2005	0.485
Any younger sister in 2005	0.518
Marital status: Single / married / other	0.043 / 0.911 / 0.046
Among married people only
Own age—spouse’s age: 0~3 / 4+ / <0	0.404 / 0.258 / 0.229
Independent variables	Mean (SD)
Spouse’s parental survival and living arrangement:
No parent alive / parent alive and coresident coresident /	0.091 / 0.028 /
parent alive and non-coresident (both / father only / mother only)	0.772 (0.486 / 0.028 / 0.258)
Spouse’s education: <High school / high school / ⩾college	0.106 / 0.364 / 0.421
Number of children: 0 / 1 / 2 / 3+	0.150 / 0.265 / 0.522 / 0.063
Year:
2005 / 2006 / 2007 /	0.175 / 0.144 / 0.123 /
2008 / 2009 / 2010 /	0.102 / 0.093 / 0.091 /
2011 / 2012 / 2013	0.091 / 0.090 / 0.091

Source. Authors’ own calculations using the 2005-2013 KLIPS.

Note. See the notes at the end of Table 1 for the analysis sample. The sample size is 5,738 for all variables except for any in-kind transfers (n = 4,732), which KLIPS began surveying in 2006, and for the amount of financial transfers (n = 4,288), which is restricted to workers with a positive amount of financial transfers. KLIPS = Korea Labor and Income Panel Study.

Figure 1 presents trends in the key variables. In 2005, 30.6% of workers were under the 40-hr statutory workweek, but about three quarters (76.0%) were so in 2013. Over the years, workers tended to visit their parents less often but provided more financial transfers and in-kind transfers. The mean frequency of visiting parents a year was 37.0 times in 2005, and it decreased to 28.9 times in 2013. By contrast, the proportion of workers who provided a positive amount of financial transfers increased from 69.4% in 2005 to 84.0% in 2013; for in-kind transfers, there was an increase from 21.4% in 2006 to 54.4% in 2013.

Figure 1.

Trends in the proportion of workers who were subject to the 40-hr statutory workweek and in workers’ provision of long-term care for non-coresident parents in Korea, 2006-2013.

Results of the Regression Analyses

Table 3 reports the results of the multiple regression analyses. For all outcome variables, we report only the regression coefficients for the 40-hr statutory workweek, that is, the estimated effects of the 4-hr reduction in the statutory workweek, although the aforementioned other variables are also controlled for in the regressions (the full regression results are available upon request). We present the OLS regression coefficients for the continuous dependent variables and marginal effects (with all other variables assumed to be at their mean values) for the binary dependent variables.

Table 3.

Regression Estimates on the Effect of the Reduced Statutory Workweek on Familial Long-Term Care by Workers’ Sex and Age in Korea.

Dependent variables	All	Male	Female	Older	Younger
Dependent variables	(1)	(2)	(3)	(4)	(5)
Number of visits a year (in log)
OLS regression coefficients	0.100**	0.117**	0.036	0.103*	0.048
(SE)	(0.048)	(0.054)	(0.103)	(0.060)	(0.082)
Number of person-years	5,738	4,185	1,553	3,618	2,120
Any financial transfers
Marginal effects from logit regressions	0.031	0.003	0.097*	0.000	0.061
(SE)	(0.033)	(0.049)	(0.051)	(0.001)	(0.046)
Number of person-years	2,867	2,103	764	1,667	1,200
Amount of financial transfers if amount > 0 (in log)
OLS regression coefficients	0.038	0.025	0.106	0.003	0.088
(SE)	(0.047)	(0.053)	(0.107)	(0.059)	(0.079)
Number of person-years	4,288	3,368	920	2,641	1,647
Any in-kind transfers
Marginal effects from logit regressions	0.078**	0.082**	−0.005	0.109	0.013
(SE)	(0.035)	(0.038)	(0.044)	(0.066)	(0.055)
Number of person-years	3,088	2,400	688	1,869	1,219
Individual fixed effects	Yes	Yes	Yes	Yes	Yes
Year fixed effects	Yes	Yes	Yes	Yes	Yes

Source. Authors’ own calculations using the 2005-2013 KLIPS.

Note. See the notes at the end of Table 2. Robust standard errors clustered at the individual level are in parentheses. We estimate OLS regressions for the number of visits and the amount of financial transfers and logit regressions for whether workers made any financial transfers and any in-kind transfers. For the binary dependent variables, the table reports marginal effects when all other variables are assumed at their mean values. The logit regressions with individual fixed effects drop 2,871 person-years for any financial transfers and 1,644 person-years for any in-kind transfers, due to the absence of variation in outcome values. Additional time-variant independent variables in Table 2 are controlled for, but their regression coefficients are not shown. “Older workers” and “younger workers” denote workers aged 35 to 64 and 18 to 34 years in 2005, respectively. OLS = ordinary least squares; KLIPS = Korea Labor and Income Panel Study.

p < .1. **p < .05. ***p < .01.

The results show that the reduced statutory workweek increased the number of visits by 10.0% (β = 0.100, p < .05; Model 1, Table 3). The policy change had no statistically significant impact on whether workers made any financial transfers or on how much the amount of the transfers was. The 40-hr statutory workweek increased the predicted probability of providing in-kind transfers by 7.8%p (β = 0.078, p < .05).

Models 2 and 3 in Table 3 present the results of the subgroup analyses by sex. Among males, the policy had a significant positive impact on the frequency of visits (β = 0.117, p < .05) and the probability of making in-kind transfers (β = .082, p < .05). There was no significant effect on male workers’ financial transfers. For female workers, only weak evidence was found for a positive impact on their chances of providing financial transfers (β = .097, p < .10).

Discussion

We estimate how the reduction in the statutory workweek from 44 to 40 hr affected Korean workers’ provision of long-term care for their non-coresident elderly parents. The results show that the policy increased the frequency of workers’ visits to parents and the proportion of workers who provided in-kind transfers. These positive effects of the 40-hr statutory workweek are consistent with what Hsu (2002) found in Taiwan (i.e., a positive impact of the Two-Day Weekend System in Taiwan on relationships with family members), but contrary to what Saffer and Lamiraud (2012) showed in France (i.e., no impact of the reduction in the French workweek on the number of workers’ visits to relatives, including visits to parents). Accordingly, we demonstrate that cultural context matters to the effectiveness of workweek policies on family outcomes, as Wharton and Blair-Loy (2006) suggested.

The size of the effects appears significant not only statistically but also in a practical sense. According to the regression estimations, the 4-hr reduction in the statutory workweek is predicted to increase a median worker’s frequency of visits by 1.20 (i.e., 12.00 × 0.100) times a year and an average worker’s frequency by 3.39 (i.e., 33.89 × 0.100) times a year. As for in-kind transfers, the reduced statutory workweek increases an average worker’s predicted probability of providing transfers by 7.8%p. If actual workweeks had been fully reduced by 4 hr, then the size of these effects could have been even larger.

Although higher wage rates imply providing less time-intensive care and more money-intensive care (Becker, 1965), our study finds strong evidence of a positive impact on visits and fails to find any evidence regarding financial transfers. These findings can be explained if the reduced workweek policy relieved workers’ time constraints and physical and psychological strains substantially, but did not affect wage rates much. In fact, H. R. Kim and Lee (2012) showed that the 40-hr workweek policy increased hourly wage rates at a statistically significant level but the size of the effect was practically small.

In contrast, the 40-hr workweek policy did reduce workers’ actual working time (E. H.-W. Kim et al., 2018; H. R. Kim & Lee, 2012). Probably what is more important practically than the reduction in work hours is that the number of work days also went down. In keeping with this, E. H.-W. Kim et al. (2018) noted that the extra day on the weekend could have helped workers visit non-coresident parents who lived far away; conceivably, visiting them on one day had been difficult. Reduced physical and psychological strains from work may also have contributed to workers’ visits by reducing fatigue (Bannai & Tamakoshi, 2014; Caruso, 2006) and psychological stress (Artazcoz et al., 2009; Virtanen et al., 2011).

In cultures where adult children do not visit parents with empty hands, more frequent visits may be accompanied by increases in financial and in-kind transfers. If the major barriers to providing in-kind transfers are time and effort needed to purchase and bring the gifts while financial transfers are mainly bounded by economic resources, our finding on in-kind transfers makes sense, together with the positive impact on visits and absence of an effect on financial transfers.

Interestingly, the positive effects on visits and in-kind transfers are mainly driven by male workers. When we analyze the effects separately by sex, the effects on the two outcomes are statistically significant among males only. In Korea, where female labor participation is rather low, and employed women do not work much fewer hours compared with their male counterparts, full-time female workers in our sample might be a select group of women who are highly committed to their careers despite the double-burden of work and family. Their time and energy could be already too restricted for a few extra hours a week to motivate them to spend more time with parents or to make in-kind transfers during visits. This speculation is consistent with the increased probability of women’s making financial transfers, although the evidence is weak. In contrast, Korean men, who contribute much less to familial care relative to women, but who are under substantial normative pressure due to Confucian values, might prioritize visiting parents when they suddenly gain extra hours and bring in-kind gifts when they visit.

This study has several caveats. First, our estimation models fail to directly control for geographic proximity to parents, which is known as a common determinant of adult children’s provision of long-term care (Greenwell & Bengtson, 1997; Hank, 2007). However, omitting this variable does not cause bias unless it relates to when workers became subject to the 40-hr statutory workweek in a systemic way. Moreover, the individual FE models eliminated the bias if the proximity was time-invariant for a given worker. We found that at least workers’ coresidence status with parents did not change as a result of the policy.

Next, familial long-term care in the current study, measured with visits, financial transfers, and in-kind transfers, goes beyond the care of elderly people with functional limitations. To have a closer look at the impact of the 40-hr workweek policy on elder care under the narrower definition, we repeated the analysis in Model 1, Table 3 for two age-subgroups, one with workers who were aged 35 or above and the other with workers aged below 35 years. As shown in Models 4 and 5 in Table 3, the impact on visits and in-kind transfers tended to be found in the older subgroup only, implying that the policy benefited parents who were older and in poorer health with a greater need for long-term care. (The effect on in-kind transfers was statistically significant in the model without individual FEs, but not in the FE model probably due to the small sample size relative to the logit regression model with individual FEs.)

Third, for married people, we analyzed only a respondent’s statutory workweek, without considering how the policy affected the respondent’s spousal work status or working time. Several previous studies have shown that statutory workweeks may affect such characteristics of partners (e.g., Goux et al., 2014), which may have a subsequent impact on the long-term care of workers’ parents. However, in Korea, Kawaguchi et al. (2013) found little change in spousal time allocation owing to the 40-hr workweek policy.

In conclusion, the current study suggests that regulating the workweek of male workers can be an effective policy tool to encourage them to provide more care for their non-coresident elderly parents, just as Maume (2015) proposed. Our findings make sense in the context of developed Asian countries, which is characterized by long work hours and Confucian ideologies. We further speculate that the effects arose through the alleviation of male workers’ constraints in terms of time and physical and psychological strains. Further evidence from other countries is needed to see whether the findings from Korea are replicable in other contexts. In addition to the well-being of workers and their nuclear families, statutory workweek policies have a spillover impact on the well-being of extended families. Regulating long work hours could be a cost-effective approach to meeting the increasing long-term care needs in rapidly aging societies in Asia.

Footnotes

Acknowledgements

We thank Dr. David Canning for his constructive suggestions.

Author’s Note

Changjun Lee is also affiliated to Spatial Dynamics Lab, University College Dublin, Dublin, Ireland.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Erin Hye-Won Kim was supported by the Humanities and Social Sciences Research Fund from the National University of Singapore and Young Kyung Do was supported by a grant from the National Research Foundation of Korea (NRF-2014S1A3A2035458).

ORCID iDs

Erin Hye-Won Kim

Changjun Lee

References

Ahn

(2016). Reduction of working time: Does it lead to a healthy lifestyle? Health Economics, 25, 969-983. doi:10.1002/hec.3198

Artazcoz

Cortès

Escribà-Agüir

Cascant

Villegas

(2009). Understanding the relationship of long working hours with health status and health-related behaviours. Journal of Epidemiology & Community Health, 63, 521-527.

Bannai

Tamakoshi

(2014). The association between long working hours and health: A systematic review of epidemiological evidence. Scandinavian Journal of Work, Environment & Health, 40, 5-18.

Bartolomé

Evans

P. A. L.

(1979). Professional lives versus private lives-shifting patterns of managerial commitment. Organizational Dynamics, 7(4), 3-29.

Becker

G. S.

(1965). A theory of the allocation of time. The Economic Journal, 75, 493-517.

Berniell

M. I.

(2012). The effects of working hours on health status and health behaviors (CEMFI-UIMP Discussion Paper). Madrid, Spain: Center for Monetary and Financial Studies.

Bianchi

S. M.

Milkie

M. A.

(2010). Work and family research in the first decade of the 21st century. Journal of Marriage and Family, 72, 705-725.

Boaz

R. F.

(1999). The transfer of resources from middle-aged children to functionally limited elderly parents: Providing time, giving money, sharing space. The Gerontologist, 39, 648-657.

Bonsang

(2007). How do middle-aged children allocate time and money transfers to their older parents in Europe? Empirica, 34, 171-188.

10.

Carmichael

Charles

Hulme

(2010). Who will care? Employment participation and willingness to supply informal care. Journal of Health Economics, 29, 182-190.

11.

Caruso

C. C.

(2006). Possible broad impacts of long work hours. Industrial Health, 44, 531-536.

12.

Cheong

K. H.

Nam

Jung

Lee

Lee Yoon

Kim

J. S.

. . . Jin

M. J.

(2012). Policy implications of changes in family structure: Focused on the increase of single person households in Korea (Research Paper 2012-47-25). Seoul: Korea Institute for Health and Social Affairs.

13.

Chesley

Poppie

(2009). Assisting parents and in-laws: Gender, type of assistance, and couples’ employment. Journal of Marriage and Family, 71, 247-262.

14.

Cicerelli

V. G.

(1990). Family support in relation to health problems of the elderly. In Brubaker

T. H.

(Ed.), Family relationships in later life (pp. 212-228). Newbury Park, CA: Sage.

15.

Couch

K. A.

Daly

M. C.

Wolf

D. A.

(1999). Time? Money? Both? The allocation of resources to older parents. Demography, 36, 219-232.

16.

Davey

Janke

Savla

(2005). Antecedents of intergenerational support: Families in context and families as context. In Silverstein

Giarrusso

Bengtson

V. L.

(Eds.), Annual review of gerontology and geriatrics: Intergenerational relations across time and place (Vol. 24, pp. 29-54). New York, NY: Springer.

17.

Y. K.

Malhotra

(2012). The effect of coresidence with an adult child on depressive symptoms among older widowed women in South Korea: An instrumental variables estimation. The Journal of Gerontology, Series B: Psychological Sciences & Social Sciences, 67, 384-391.

18.

Estevão

(2006). Are the French happy with the 35-hour workweek? (IMF Working Paper 06/251). Washington, DC: International Monetary Fund.

19.

Feenstra

R. C.

Inklaar

Timmer

M. P.

(2015). The next generation of the Penn World Table. The American Economic Review, 105, 3150-3182.

20.

Glass

A. P.

Chen

L. K.

Hwang

Ono

Nahapetyan

(2010). A cross-cultural comparison of hospice development in Japan, South Korea, and Taiwan. Journal of Cross-Cultural Gerontology, 25(1), 1-19.

21.

Gornick

J. C.

Heron

(2006). The regulation of working time as work-family reconciliation policy: Comparing Europe, Japan, and the United States. Journal of Comparative Policy Analysis, 8, 149-166.

22.

Goux

Maurin

Petrongolo

(2014). Worktime regulations and spousal labor supply. The American Economic Review, 104, 252-276.

23.

Greenhaus

J. H.

Beutell

N. J.

(1985). Sources of conflict between work and family roles. Academy of Management Review, 10, 76-88.

24.

Greenwell

Bengtson

(1997). Geographic distance and contact between middle-aged children and their parents: The effects of social class over 20 years. The Journals of Gerontology, Series B: Psychological Sciences & Social Sciences, 52, S13-S26. doi:10.1093/geronb/52B.1.S13

25.

Hank

(2007). Proximity and contacts between older parents and their children: A European comparison. Journal of Marriage and Family, 69, 157-173. doi:10.1111/j.1741-3737.2006.00351.x

26.

Hogan

D. P.

Eggebeen

D. J.

Clogg

C. C.

(1993). The structure of intergenerational exchanges in American families. American Journal of Sociology, 98, 1428-1458.

27.

Hsu

M.-R.

(2002). The impacts of two-day weekend system on living culture along with efficient countermeasures-Taiwan Experience. Journal of Korean Home Economics Association (English Edition), 3, 137-150.

28.

Jacobs

J. A.

Gerson

(2001). Overworked individuals or overworked families? Explaining trends in work, leisure, and family time. Work and Occupations, 28, 40-63.

29.

Kao

H. F.

Stuifbergen

A. K.

(1999). Family experiences related to the decision to institutionalize an elderly member in Taiwan: An exploratory study. Social Science & Medicine, 49, 1115-1123.

30.

Kawaguchi

Lee

Hamermesh

D. S.

(2013). A gift of time. Labour Economics, 24, 205-216. doi:10.1016/j.labeco.2013.09.003

31.

Kim

E. H.-W.

Cheung

A. K.-L.

(2015). Women’s attitudes toward family formation and life stage transitions: A longitudinal study in Korea. Journal of Marriage and Family, 77, 1074-1090.

32.

Kim

E. H.-W.

Cook

P. J.

(2011). The continuing importance of children in relieving elder poverty: Evidence from Korea. Ageing & Society, 31, 953-976.

33.

Kim

E. H.-W.

Lee

Y. K.

(2018). The effect of adult children’s working hours on visits to elderly parents: A natural experiment in Korea. Population Research and Policy Review. doi:10.1007/s11113-018-9486-0

34.

Kim

H. R.

Lee

J. M.

(2012). The impacts of the 40 hour work week standad on actual working hours, wages and employment. Korean Journal of Labor Economics, 35(3), 83-100.

35.

Kinsella

K. G.

Phillips

D. R.

(2005). Global aging: The challenge of success (Vol. 60, No. 1). Washington, DC: Population Reference Bureau.

36.

Korea Labor Institute. (2015). User guider for KLIPS. Seoul: Author.

37.

Korean Ministry of Labor. (2007). Manuals for the introduction of 40-hour workweek. Seoul: Author.

38.

Lawton

Silverstein

Bengtson

(1994). Affection, social contact, and geographic distance between adult children and their parents. Journal of Marriage and Family, 56, 57-68.

39.

Lee

S. H.

Mason

Park

(2011). Why does population aging matter so much for Asia? Population aging, economic security and economic growth in Asia (ERIA Discussion Paper Series). Retrieved from https://ideas.repec.org/p/era/wpaper/dp-2011-04.html

40.

Lin

I. F.

Goldman

Weinstein

Lin

Y. H.

Gorrindo

Seeman

(2003). Gender differences in adult children’s support of their parents in Taiwan. Journal of Marriage and Family, 65, 184-200.

41.

Lye

D. N.

(1996). Adult child-parent relationships. Annual Review of Sociology, 22, 79-102.

42.

Maume

D. J.

(2015). Can men make time for family? Paid work, care work, work-family reconciliation policies, and gender equality. Social Currents, 3, 43-63.

43.

Organisation for Economic Co-operation and Development. (2012). Closing the gender gap: Act now. Paris, France: Author.

44.

Organisation for Economic Co-operation and Development. (2013). OECD labour force statistics, version July 2013. Paris, France: Author.

45.

Organisation for Economic Co-operation and Development. (2016). Hours worked (indicator). Paris, France: Author. doi:10.1787/47be1c78-en

46.

Pleck

J. H.

Staines

G. L.

Lang

(1980). Conflicts between work and family life. Monthly Labor Review, 103(3), 29-32.

47.

Rossi

(1990). Of human bonding: Parent-child relations across the lifecourse. New York, NY: Aldine de Gruyter.

48.

Saffer

(2008). The demand for social interaction. The Journal of Socio-Economics, 37, 1047-1060.

49.

Saffer

Lamiraud

(2012). The effect of hours of work on social interaction. Review of Economics of the Household, 10, 237-258.

50.

Sarkisian

Gerstel

(2004). Explaining the gender gap in help to parents: The importance of employment. Journal of Marriage and Family, 66, 431-451.

51.

Shi

(2009). “Little quilted vests to warm parents’ hearts”: Redefining the gendered practice of filial piety in rural North-eastern China. The China Quarterly, 198, 348-363.

52.

Shuey

Hardy

M. A.

(2003). Assistance to aging parents and parents-in-law: Does lineage affect family allocation decisions? Journal of Marriage and Family, 65, 418-431.

53.

Sloan

F. A.

Zhang

H. H.

Wang

(2002). Upstream intergenerational transfers. Southern Economic Journal, 69, 363-380.

54.

Stern

(1995). Estimating family long-term care decisions in the presence of endogenous child characteristics. Journal of Human Resources, 30, 551-580.

55.

Sun

(2002). Old age support in contemporary urban China from both parents’ and children’s perspectives. Research on Aging, 24, 337-359.

56.

Takeuchi

Rahman

Ishiguro

Nomura

(2014). Long working hours and pregnancy complications: Women physicians survey in Japan. BMC Pregnancy and Childbirth, 14, Article 245. doi:10.1186/1471-2393-14-245

57.

van Bastelaer

Lemaître

Marianna

. (1997). The definition of part-time work for the purpose of international comparisons (OECD Labour Market and Social Policy Occasional Papers, 22). Paris, France: Organisation for Economic Co-operation and Development.

58.

van Putten

A. E.

Vlasblom

J. D.

Dykstra

P. A.

Schippers

J. J

. (2010). The absence of conflict between paid-work hours and the provision of instrumental support to elderly parents among middle-aged women and men. Ageing & Society, 30, 923-948.

59.

Virtanen

Ferrie

J. E.

Singh-Manoux

Shipley

M. J.

Stansfeld

S. A.

Marmot

M. G.

. . . Kivimäki

(2011). Long working hours and symptoms of anxiety and depression: A 5-year follow-up of the Whitehall II study. Psychological Medicine, 41, 2485-2494.

60.

Wharton

A. S.

Blair-Loy

(2006). Long work hours and family life: A cross-national study of employees’ concerns. Journal of Family Issues, 27, 415-436.

61.

Yeung

W.-J. J.

Cheung

K.-L.

(2015). Living alone: One-person households in Asia. Demographic Research, 32, 1099-1112.

62.

Zhan

H. J.

Montgomery

R. J.

(2003). Gender and elder care in China: The influence of filial piety and structural constraints. Gender & Society, 17, 209-229.

63.

Zissimopoulos

(2001). Resource transfers to the elderly: Do adult children substitute financial transfers for time transfers? (Labor and Population Program Working Paper Series, 01-05). Santa Monica, CA: RAND.