Work hours,work schedules,and subjective well-being in Korea

Abstract

Spanish

French

Korea is well known for long work hours. This study examined the effect of not only the total amount of working time but also work schedules on individuals’ subjective well-being in Korea. Drawing on the 2014 Korean Time Use Survey data, the authors selected currently employed people aged 19–65 and examined the effects of total work hours as well as work schedules on feeling rushed, feeling tired, and life satisfaction. Based on the ordered logistic regression analysis, long work hours were negatively related to subjective well-being regardless of work schedules. After adding an interaction between standard and nonstandard work schedules, the study found that while work hours during both standard and nonstandard schedules are associated with declines in psychological status, the negative effect of working time during nonstandard work schedules on subjective well-being is stronger as the work hours during standard schedules decrease.

Keywords

Korea nonstandard schedules standard schedules subjective well-being work hours

Introduction

Recently, a growing number of studies have focused on individuals’ subjective well-being (SWB), including the relationship between work hours and SWB. Prior studies have consistently found that employment itself is positively related to SWB, because it provides a basis for social participation and engagement as well as a source of income (Robertson and Cooper, 2011; Rudolf, 2014; Valente and Berry, 2016). Focusing on the effect of work hours on SWB, most previous studies have suggested that long work hours have an adverse effect on SWB (Craig and Brown, 2017; Kleiner and Pavalko, 2010).

Due to an expansion of the service sector and the emergence of the 24/7 economy, nonstandard work hours (e.g., working in the evening, night shifts, or working on weekends) have become common in recent years. Based on the assumption that working during nonstandard hours renders it difficult for workers to balance work and family responsibilities, empirical studies of nonstandard work hours have focused mostly on their effect on family time including childcare and marital stability (Craig and Brown, 2017; Gracia and Kalmijn, 2016; Kalil et al., 2010; Mills and Täht, 2010; Presser, 2000; Strazdins et al., 2006). While the effects of nonstandard work schedules on childcare time vary by gender and country, nonstandard work hours have consistently shown negative effects on overall family time, marital stability, and health.

Although previous studies have highlighted the importance of work-related factors in understanding individuals’ SWB, they had some limitations. First, the extant research on the effect of work hours on SWB has been conducted mostly in Western countries while the effect was not thoroughly examined among Asians. Second, prior studies have considered primarily the total number of hours worked without paying much attention to when these hours are worked. Considering people tend to prefer working during specific hours of the day, in accordance with the collective rhythm of society (Minnen et al., 2016), working from 9 a.m. to 5 p.m. may have different effects on SWB compared to working from 3 p.m. to 11 p. m., even if the total working hours per day are the same (e.g., 8 hours per day). Third, previous studies on work schedules examined mostly their effects on the marital relationship, family time or childcare time, and health. Few explored a direct effect of work schedules on individuals’ SWB.

To compensate for these limitations and to expand scholarly understanding of the relationship between work-related factors and individuals’ SWB, we examined the relationship between work hours, work schedules, and SWB in Korea. A thorough analysis using a large-scale nationally representative database (2014 Korean Time Use Survey data) will shed light on how Koreans spend their days, how long and when they work, and whether work hours and work schedules are related to various measures of SWB. In addition, recognizing not only total work hours but also work schedules (i.e., when to work) may be important for individuals’ SWB, we attempted to tease out the independent effects of work hours and work schedules as well as their joint effect. The current study offers insights into Korean people’s working lives, including the prevalence of long work hours in Korea and the effects of working nonstandard hours on SWB.

Literature review and research hypotheses

Work-related factors and individual subjective well-being

A significant body of research has found that long work hours limit the opportunity for the recuperative activities necessary to maintain health and well-being and may have deleterious effects on individuals’ fatigue, sleep deprivation, time pressure, and subjective well-being (Craig and Brown, 2017; Harrington, 2001; Kleiner and Pavalko, 2010). Extended work hours also decrease time with children and spouses and increase work–family conflict, which can negatively affect individual and family well-being (Kalil et al., 2010; Rudolf, 2014; Wharton and Blair-Loy, 2006). Although there are some inconsistencies in the findings about the relationship between long work hours and SWB, such as U-shaped associations or insignificant relationships (Luttmer, 2005; Okulicz-Kozaryn, 2011; Valente and Berry, 2016), working long hours is generally considered to have negative effects on individuals’ SWB.

Work schedules have only recently emerged as a research topic due to increases in the number of people working nonstandard hours, such as early in the morning, at night, or on the weekends. Daniel Hamermesh (1995) highlighted the importance of analyzing what people are doing at specific points in the day instead of focusing on the total amount of time. In general, those who are less educated, have fewer skills, and are members of minority groups tend to work at night, which indicates that people with high levels of human capital do not prefer a nonstandard work schedule (Glorieux et al., 2008; Hamermesh, 1995; Minnen et al., 2016).

Previous studies of work schedules have consistently found a negative relationship between nonstandard work schedules and time spent with spouses (Kalil et al., 2010; Mills and Täht, 2010; Presser, 2000). In addition, working on the weekends usually reduces both family time and shared leisure time (Craig and Brown, 2014) and some research findings have suggested a negative association between parental nonstandard work schedules and children’s emotional and behavioral difficulties, particularly among preschool-aged children (Starzdins et al., 2006).

As noted above, very few scholars have examined the separate and joint effects of work hours and work schedules on SWB, and among the small group of extant studies, few have used a direct measure of SWB. One recent study based on Australian data (Craig and Brown, 2017) addressed the relationship between work hours, nonstandard work schedules, and time pressure as a measure of SWB. The results indicated that while working long hours increased time pressure, working during nonstandard hours was not associated with time pressure. Although this study provided valuable novel information about work hours, work schedules, and SWB, it was limited in that it only tested for the independent effects of work hours and work schedules and used time pressure as an indirect measure of SWB.

Theoretical background: Relationships between work-related factors and SWB

Working a nonstandard schedule and working for extended hours frequently have detrimental effects on individuals’ mental and physical health and family life (Fenwick and Tausig, 2001; Harrington, 2001; Totterdell, 2004). The theoretical perspectives that account for the relationships between these work-related factors and SWB focus on the two main areas: (1) biological aspects and (2) social aspects.

Working at night or early in the morning can disturb biological rhythms, so-called circadian rhythms. Many biological functions, such as body temperature, hormone production, and blood pressure, operate on a 24-hour cycle that is sensitive to sunlight and darkness (Harrington, 2001; Totterdell, 2004). Working extended hours or a nonstandard schedule can cause a mismatch between the circadian system and work hours, which may harm mental and physical health. Specifically, long work hours and nonstandard schedules often increase fatigue and work-related stress, and interfere with sleep quality or regular sleep times, which in turn can reduce motivation and efficiency or lead to depression (Kleiner and Pavalko, 2010).

Further, the disruptive effects of working long hours and/or a nonstandard schedule extend to family and social life (Harrington, 2001; Kleiner and Pavalko, 2010). Individuals must meet the competing demands of work, family, and other social activities. Because both excessive work hours and nonstandard work schedules keep people from maintaining family routines and family time, the level of work–family conflict increases. Those working under these conditions may find that they are out of sync with family and friends in regard to the time available for social interaction and participation, which may lead to social isolation, anxiety, and depression (Fenwick and Tausig, 2001). Given that life satisfaction encompasses multiple dimensions, the disruptive effects of long working hours and nonstandard work schedules on individuals’ biological rhythms and coordination with family or social life may generate negative effects on SWB.

Due to the lack of prior studies assessing a possible interaction effect of work hours and work schedules on SWB, there are few theoretical perspectives that seek to explain this relationship. However, some scholars have noted that people’s perceptions of long working hours differ by country (Kim et al., 2018; Okulicz-Kozaryn, 2011; Valente and Berry, 2016), and thus cultural characteristics that affect the perception of long working hours may explain the interaction between work hours and work schedules. For example, in countries where long working hours are prevalent and people accept this situation without much resistance, working a nonstandard schedule may have a stronger negative effect on SWB than hours worked during a standard schedule.

The context of work in Korea

Over the past decade, Korea has continuously ranked in the top three nations in terms of long work hours (OECD, 2018). Since 2008, the average annual work hours per worker in Korea have been more than 2000 hours every year, which is much higher than the average of OECD countries (1700–1800 hours). In response to social concerns about the potential adverse effects of long work hours on individuals’ health and SWB, as well as work–family balance in general, the Korean government has implemented policies to reduce work hours. A 5-day work week policy was introduced in 2004. The policy was first implemented in large corporations and then expanded to most sectors of the economy (except small-scale service jobs) over the next four years (Rudolf, 2014). To further reduce work hours, the government introduced a limit on weekly work hours (i.e., 52 hours per week) in July 2018.

Along with long work hours, a work ethic that places a high value on hard work, diligence, and dedication to the company or employer remains prominent in Korea. In Korea’s collective culture, frequently having dinner with colleagues after work is considered important and good for group solidarity (Kim et al., 2018). However, the pursuit of individual well-being and work–family balance based on an individualistic attitude has gradually become more important, particularly among younger generations, hence Korea seems to be undergoing a transition in both work hours and work culture.

Given that self-employed people and people with few resources are more likely to work nonstandard schedules (Glorieux et al., 2008; Hamermesh, 1995; Minnen et al., 2016), high rates of self-employed people and temporary employment in Korea may be associated with working nonstandard schedules or excessive hours. In Korea, the self-employment rate, including unpaid family workers, is about 25% and the temporary employment rate is about 21.2%; both measures place Korea seventh highest among OECD countries (OECD, 2018). These proportions are much higher than the average proportions for the 28 EU countries: 15% (self-employment rate) and 14.2% (temporary employment rate).

Research hypotheses

Based on a review of the extant literature, we propose the following hypotheses. First, working extended hours has negative effects on individuals’ SWB. As individuals’ total work hours increase, their feelings of tiredness and time pressure increase and overall life satisfaction decreases. Second, working a nonstandard schedule is negatively related to SWB. Increases in work hours and a nonstandard schedule raise the perceived level of fatigue and time pressure, which in turn decrease overall life satisfaction. Third, although both extended hours of work and working a nonstandard schedule have a negative effect on SWB, the influence of working a nonstandard schedule on SWB may vary with work hours during standard schedules because of perceptions about long work hours.

Assessing these hypotheses requires information about how long and when employed people in Korea work. Therefore, we gathered the following information: (1) how long employed people in Korea work per day; (2) what proportions of Koreans work standard and nonstandard schedules; (3) whether socioeconomic characteristics differ by work hours and work schedules.

Data and sample

The data for this study were drawn from the 2014 Korean Time Use Survey (KTUS hereafter), which is the most recent one. KTUS data have many strengths. First, the data include diaries in which the respondents reported the beginning and ending time of every activity. This allowed us to calculate the total amount of time for a specific activity per day as well as when it started and ended. Second, unlike time use survey data for some countries, like the USA, which do not include information about household members, in KTUS, all household members who are older than 9 years answered the survey. Therefore, we could capture household composition and analyze the data at the couple’s level. Third, in the 2014 KTUS, SWB-related questions (e.g., feelings of tiredness, perceived time pressure, and overall life satisfaction) were included. Hence, we could see the relationship between respondents’ time spent in various activities in daily life and SWB.

A nationally representative sample of about 12,000 households comprising roughly 27,000 respondents was surveyed, with all household members over 9 years of age, as noted, completing sociodemographic questionnaires, three items measuring SWB, and time diaries for two consecutive days at 10-minute intervals. Because separate analyses using the information from the first and second days produced substantively similar results, we report the results from the day 1 diary.

Among total respondents, we limited our sample to 13,413 currently employed people aged 19–65. After removing those who were not working on the survey day (3380) due to a day off or sick leave and those who reported a working time of 0 minutes (40), the final study sample consisted of 9993 workers (4207 women and 5786 men).

On average, sample members were in their mid-forties (43.6) and one-fifth of the sample members (20.0%) were single, seven in 10 (72.2%) were married, just over 5% were divorced (Table 1), and the remaining (2.7%) were widowed. Most respondents assessed their health as 3 on a 1–5 scale. Roughly half ended their formal education after middle school (13.6%) or high school (36.5%) while about 30% attended a four-year college or graduate school and the rest had finished a two-year college course (19.6%).

Table 1.

Descriptive statistics for adult workers by gender (in % unless noted otherwise).

		Total	Women	Men
Age^a		43.6	43.0	44.0
Marital status	Single	20.0	22.0	18.5
	Married	72.2	65.7	76.8
	Divorced	5.2	6.9	3.9
	Widowed	2.7	5.4	0.8
Health^a	1 = very bad, 5 = very good	3.3	3.2	3.4
Education	Middle school	13.6	17.8	10.7
	High school	36.5	37.9	35.5
	Two-year college	19.6	19.6	19.7
	Four-year college or higher	30.2	24.8	34.2
Occupation	Managerial/professional	19.3	21.8	17.4
	Office	17.9	16.6	18.8
	Service	26.1	34.4	19.9
	Manual labor	31.5	22.3	38.3
	Agricultural	5.3	4.8	5.6
Employment status	Employer/self-employed	23.8	16.1	29.5
	Regular/permanent	51.7	45.8	55.9
	Temporary/daily	19.1	26.7	13.7
	Unpaid employee at family business	5.4	11.4	1.0
Individual monthly income^a (1–12 scale)	1 = 0, 2 = $500 and less, 3 = $500–1000,12 = $5000 and more	6.0	4.5	7.1
Number of preschool children	0	84.5	89.8	80.6
	1	10.4	7.2	12.7
	2 or more	5.1	3.0	6.7
Number of school children	0	68.1	68.7	67.6
	1	17.0	16.8	17.1
	2 or more	15.0	14.5	15.3
Rural		7.3	7.6	7.1
Weekend work		24.6	24.1	25.0
Life dissatisfaction^a	1 = very satisfied, 5 = very dissatisfied	2.9	2.9	2.8
Time pressure^a	1 = always relaxed, 4 = always rushed	3.1	3.1	3.0
Fatigue^a	1 = not at all, 4 = very much	3.2	3.3	3.2
N		9993	4207	5786

Means for these variables.

When we see an overall distribution of occupation, a little fewer than one in five sample members were managerial/professional (19.3%) or office (17.9%) workers, more than half were employed as manual labor (31.5%) or service (26.1%) workers, and just over 5% (5.3%) worked in agriculture. However, a few differences by gender were noted. For example, a higher proportion of women were in managerial or professional occupations (21.8%) compared to men (17.4%) because KTUS classifies nurses and teachers as professional. More women (34.4%) were in service sectors compared to men (19.9%) while more men (38.3%) were working in manual labor compared to women (22.3%).

Approximately half (51.7%) of the respondents were regular or permanent workers while just under one in four (23.8%) were employers or self-employed and a slightly smaller group (19.1%) were temporary or daily employees. The remaining 5.4% were unpaid workers at family-owned businesses, such as a cashier in a spouse’s restaurant or a salesperson at a parents’ clothing store. Gender differences in employment status were also observed. Men were more likely to be employers (29.5%) or regular/permanent workers (55.9%) than women (16.1% and 45.8% respectively). Women were more likely to be temporary or daily workers (26.7%) than men (13.7%). The most striking difference emerged between women and men unpaid workers at the family-owned business, at 11.4% and 1.0%, respectively.

The individual monthly income was measured on a 12-point scale, ranging from 1 (no income) to 12 ($5000 and more), with a $500 range of each scale point (i.e., 2 = $500 and less, 3 = $500–1000, 4 = $1000–1500, etc.). The average individual monthly income was 6.0 on the 12-point scale or between $2000 and $2500. One in 10 participants (10.4%) had one preschool child in the household, one in 20 (5.1%) lived with two or more preschool-aged children, 17.0% had one school-aged child at home, and 15.0% had two or more school-aged children at home. In the total sample, only about 7% resided in rural areas. Overall, 24.6% of men and women answered the survey on the weekends. People who worked on the weekends were mostly self-employed.

Previous research has shown that work hours and employment schedules affect many dimensions of subjective well-being including general satisfaction (Keizer et al., 2010), feeling rushed (Craig and Brown, 2017), being distressed (Kleiner and Pavalko, 2010), and feeling tired (Niu et al., 2011). To explore the psychological implications of hours worked during standard and nonstandard schedules more extensively, we examine all measures of subjective well-being in the 2014 KTUS: time pressure, fatigue, and life satisfaction. Information regarding subjective well-being was measured using the following three questions: ‘Do you usually feel rushed or relaxed?’ (range: 1 = always relaxed to 4 = always rushed); ‘How tired do you feel at the end of the day?’ (range: 1 = not at all to 4 = very much); and ‘Are you usually satisfied with your overall life?’ (range: 1 = very dissatisfied to 5 = very satisfied). To be consistent with response categories of the first two questions, we recoded the response categories for the last question as 1 = ‘very satisfied’ and 5 = ‘very dissatisfied,’ and call it ‘life dissatisfaction’ in Table 1.

Most workers in the sample reported feeling at least some time pressure and physical exhaustion. The average reported levels of feeling rushed and tired were 3.1 (3 = often feel rushed, 4 = always feel rushed) and 3.2 (3 = feeling tired a little, 4 = feeling tired very much) on a 1–4 scale, respectively. The average level of general dissatisfaction was 2.9 on a scale from 1 to 5.

Analysis

The proportion of workers engaging in paid work by time of day

Figure 1 shows the proportions of the participants who reported that they were working at each hour on the survey day. We excluded reports of commuting or searching for paid work.

Figure 1.

Percentages of adult workers engaging in paid work by hours of the day.

More than 20% of the participants had already arrived at their workplace at 8:00 a.m. These workers needed to wake up at dawn, considering time for getting ready and commuting. By 11:00 a.m., more than 60% of employees were working. The proportion of working respondents dropped to about 30% at lunchtime and then increased to 60% after lunch. About 30% of the respondents worked at lunchtime. Roughly one in five employees were working at 7:00 p.m. and roughly 15% were working at 9:00 pm. From midnight until dawn, just over 3% of respondents were working.

Figure 1 shows that men generally arrived at their workplaces earlier than women did, with 32.2% of men and 16.4% of women reporting that they were at work by 8:00 a.m. This gender difference emerged likely in part because the responsibility for getting children ready for school generally falls on mothers (Cha and Song, 2017) rather than fathers. Men also worked later than women, on average, but the gender gap in the proportion of those working at night was smaller compared to that in the proportion of those during the morning hours, with 9.9% of men and 8.7% of women working at 10:00 p.m. Coupled with the Korean cultural norm of women being expected to take responsibility for the bulk of childcare and housework (Oshio et al., 2013), a similar proportion of women and men working late at night can keep women from combining work and family responsibilities (Kim, 2018).

Working hours by work schedules

Korea is well known for long work hours and many Korean workers take long work hours for granted. From July 2018, Korean labor laws require employees to work no more than 40 hours per week, on average, and any extra work time must be based on reciprocal consent and cannot exceed 12 hours, for a total weekly maximum of 52 hours. Therefore, a standard work schedule would be from 9:00 a.m. to 6:00 p.m. (with a one-hour lunch). However, as shown in Figure 1, one in four employees in the sample were at work from 8:00 a.m. to 7:00 p.m. Hence, in this study, we defined standard work hours (or standard work schedules) as 8:00 a.m. to 7:00 p.m. and nonstandard hours (or nonstandard work schedules) as 7:00 p.m. until 8:00 a.m.

We first divided the workers in the sample into three categories: those working only during standard schedules, those working during both standard and nonstandard schedules, and those working only during nonstandard schedules. We called the first group ‘standard’ workers (43.8%). We further divided the second group into the following two categories: (1) ‘long’ workers (47%) working more hours during standard schedules than nonstandard schedules and (2) ‘late’ workers (9.1%) working more hours during nonstandard schedules than standard schedules. Because only 2.8% of the participants worked only during nonstandard hours, we combined this group with ‘late’ workers. Table 2 includes the descriptive statistics for the total sample of workers by work schedules.

Table 2.

Descriptive statistics for adult workers by work schedules (in % unless noted otherwise).

		Total	Standard workers	Long workers	Late workers	Tests^b
Gender	Woman	42.1	48.6	36.4	40.2	**
	Man	57.9	51.4	63.6	59.8
Age^a		43.6	42.7	44.6	42.3	**
Marital status	Single	20.0	20.3	18.4	26.9	**
	Married	72.2	73.0	73.2	62.7
	Divorced	5.2	4.3	5.6	6.9
	Widowed	2.7	2.4	2.9	3.5
Health^a	1 = very bad, 5 = very good	3.3	3.4	3.3	3.3	**
Education	Middle school	13.6	10.7	16.3	14.2	**
	High school	36.5	33.2	38.6	41.7
	Two-year college	19.6	21.0	18.3	19.8
	Four-year college or higher	30.2	35.1	26.8	24.4
Occupation	Managerial/professional	19.3	23.8	16.6	11.6	**
	Office	17.9	24.9	13.7	5.3
	Service	26.1	20.9	28.1	40.0
	Labor	31.5	26.5	34.9	38.3
	Agricultural	5.3	3.8	6.7	4.8
Employment status	Employer/self-employed	23.8	19.5	27.3	27.0	**
	Regular/permanent	51.7	56.3	49.8	39.4
	Temporary/daily	19.1	19.4	17.3	27.1
	Family	5.4	4.9	5.6	6.5
Individual monthly income^a	1 = 0, 2 = $500 and less, 3 = $500–1000,12 = $5000 and more	6.0	5.9	6.2	5.5	**
Number of preschool children	0	84.5	83.3	85.2	86.6	*
	1	10.4	11.2	9.9	9.1
	2 or more	5.1	5.6	4.9	4.3
Number of school children	0	68.1	65.2	69.9	72.6	**
	1	17.0	18.1	16.4	14.7
	2 or more	15.0	16.8	13.7	12.7
Rural		7.3	6.1	8.7	6.0	**
Weekend		24.6	21.8	25.7	33.0	**
Life dissatisfaction^a	1 = very satisfied, 5 = very dissatisfied	2.9	2.8	2.9	3.0	**
Time pressure^a	1 = always relaxed, 4 = always lacking	3.1	3.0	3.2	3.0	**
Fatigue^a	1 = not at all, 4 = very much	3.2	3.2	3.3	3.3	**
N		9993	4382	4700	911

p < .05; ** p < .01.

Means for these variables.

One-way and chi-square tests are done for linear and categorical variables respectively.

As we can see from the last row of Table 2, out of the total sample (9993 respondents), about half were long workers (4700) and slightly less than half were standard workers (4382). Roughly 10% of the respondents were late workers. Most employees in our sample worked either only in standard work schedules or in standard work schedules as well as extended hours. A large proportion of long workers reflect working circumstances in Korea, where long working hours are prevalent.

For standard workers, the gender proportion was similar, with a slightly higher proportion of men (51.4%) than women (48.6%). In contrast, a large gender disparity was noted for long workers, who comprised a much higher proportion of men (63.6%) than women (36.4%). Similarly, a large gender disparity was observed for late workers, with 59.8% of men and 40.2% of women. Overall, men were more likely to work long hours or in nonstandard schedules compared to women. The average age of the employees in our sample by work schedule was similar.

The distribution of marital status for standard and long workers was similar, with seven out of 10 respondents being married and about two out of 10 being single. Compared to standard and long workers, late workers were more likely to be single (26.9%) and less likely to be married (62.7%). A somewhat higher proportion of late workers were divorced and widowed compared to standard and long workers.

Health status was similar regardless of work schedules, showing an average score of 3.3 or 3.4 on a 1–5 scale. Concerning the educational distribution, standard workers attained the highest educational attainment, specifically: 35.1% had a four-year university diploma while only 26.8% of long workers and 24.4% of late workers had a four-year university degree. Compared to nonstandard workers, standard workers were more likely to be married and have higher educational attainment.

Unsurprisingly, our data showed that employment features vary by working schedules. Standard workers were more likely to have white-collar jobs than the other two groups, with 23.8% in managerial/professional occupations and 24.9% in office occupations. In contrast, higher proportions of service workers, 28.1% and 40.0% among long workers and late workers, and higher proportions of manual workers, 34.9% and 38.3% among long workers and late workers, respectively, were observed. It seems that standard workers are concentrated in white-collar jobs while long and late workers are concentrated in service or blue-collar jobs.

Regarding employment status, the percentage of permanent or regular workers was the highest regardless of work schedules, although it was slightly higher among standard workers (56.3%) followed by long workers (49.8%) and late workers (39.4%). In this analysis, an employer category included both an employer with employees and self-employed without any employees. Because the proportion of employers was higher among long (27.3%) and late (27.0%) workers than among standard workers (19.5%), we can assume that most employers were self-employed in service-related occupations. Both temporary workers and unpaid family workers showed the biggest proportions among late workers at 27.1% and 6.5%, respectively. In terms of the individual’s monthly income, long workers earn the most money, followed by standard and late workers, in that order. Considering employment status and monthly income, late workers seem to be less stable and earn less money than the other two groups.

Consistent with the results of the previous studies (Hamermesh, 1995; Minnen et al., 2016), standard workers are more likely than their counterparts to have preschool-age and school-age children. Those who have preschool or school-age children seem to work in standard schedules to accommodate childcare needs. In terms of residential areas, fewer than 10% of employees in our sample lived in rural areas. Most respondents answered the survey on weekdays. Those who worked only during standard hours felt less time pressure and less tired than either long or late workers. In a similar vein, standard workers felt less dissatisfied with their lives than the other groups. As expected, long workers tend to feel more time pressure than other two groups while both feeling tired and life dissatisfaction were rated in the mid-range.

Table 3 shows how much time each group of workers spent working at a job. On average, respondents worked about 7 hours per day (428.9 minutes). Although it is not shown in the table, the median time spent at work in the full sample was 440 minutes, slightly higher than the average. While seven out of 10 standard workers worked fewer minutes than the full-sample median, about 70% of long employees spent more than 440 minutes at work. In contrast, late workers had a wide range of daily work time, the bottom decile of this group working a little more than 2 hours while the top decile working more than 10 hours.

Table 3.

Daily minutes on paid work for adult workers by work schedules.

	Total	Standard workers	Long workers	Late workers
Averages of daily minutes on paid work	428.9	365.9	494.1	395.4
N	9993	4382	4700	911

While standard workers spent a little more than 6 hours at work each day, long and late workers allocated more than 8 hours and 6.5 hours to paid labor, respectively. Although it is not shown in the table again, out of an average of 8 hours of work per day, long workers spent 6.5 hours working standard hours and 1.5 hours working nonstandard hours either early morning or late evening. Late workers spent 1.5 hours at work in standard hours and 5 hours at work at night, on average. Typically, long workers reported for duty during standard hours and stayed at work until night while late workers arrived in the afternoon or later and worked until very late, sometimes until dawn.

The time use patterns of standard, long, and late workers

In this section, we examine how much time standard, long, and late workers spent on certain major daily activities (see Table 4). The results are consistent with many prior studies suggesting that an increase in work hours is associated with decreased time spent on family responsibilities, leisure activities, and sleep (Cha and Eun, 2014; Gracia and Kalmijn, 2016). Long workers spent the least time engaged in the four major daily activities, specifically, sleep, housework, care work for family members, and leisure.

Table 4.

Daily minutes on selected activities for adult workers by work schedules.

	Women					Men
	Standard workers	Long workers	Late workers	One-way tests	Bartlett’s test^a	Standard workers	Long workers	Late workers	One-way tests	Bartlett’s test^a
Sleep	450.9	434.9	429.6	**	AB	464.8	435.3	442.6	**	AB
Housework	120.5	86.2	125.0	**	AC	19.3	13.9	30.3	**	ABC
Care work	25.1	10.3	14.6	**	AB	12.3	6.7	12.1	**	AC
Leisure	159.8	133.0	180.0	**	ABC	208.0	160.6	224.1	**	ABC

p < .05; ** p < .01.

A, B, and C indicate statistically significant differences between standard and long workers, standard and late workers, and long and late workers respectively.

Among women, standard workers seemed to sleep the longest per day and late workers tended to devote long hours to leisure activity. Female long workers devoted about 30 minutes less to housework and spent 5–15 minutes less on care work for other family members compared to standard and late workers. They also spent 30 minutes and 50 minutes less on leisure activities than did standard and late workers, respectively. Similarly, among men, long workers tended to spend less time in the above-mentioned four activities compared to the other two groups. Male late workers spent the most time on housework and leisure. They devoted 10–15 minutes more on housework and spent 50–60 minutes more in leisure activities than standard and long workers.

Compared to standard and long workers, late workers were more involved in leisure activities and household tasks regardless of gender maybe because they were more likely to have free time during the morning and afternoon. Standard workers slept 20 and 30 minutes longer compared to long and late workers, respectively, regardless of gender. Finally, standard women workers contributed the most to family caring, which is consistent with the literature suggesting that women tend to scale back their work time to adjust to the expectation that they will care for family members (Craig and Mullan, 2013).

Multivariate associations between work hours/work schedules and subjective well-being

To examine the associations between work hours, work schedules, and subjective well-being, we estimated a set of ordered logistic regression models with time pressure, fatigue, and life dissatisfaction as the dependent variables. The models included a set of control variables as well as the following independent variables: total working minutes per day (Model 1), working minutes during standard hours and nonstandard hours separately (Model 2), and working minutes during standard hours and nonstandard hours as well as an interaction term between these two variables (Model 3). Because the distributions of time on standard and nonstandard schedules are highly right-skewed, we use their logarithmic values in the models.

The results of Model 1 show that for all three outcomes, working long hours was associated with a reduction in subjective well-being when controlling for a set of demographic and employment variables. In line with prior research (Craig and Brown, 2017; Kleiner and Pavalko, 2010) and our research hypothesis, working people in Korea are more likely to be dissatisfied, pressed for time, and worn out when they work longer hours. Spending an additional 10 minutes at work was associated with being more discontented with life (odds ratio = 1.358), feeling more pressed for time (odds ratio = 2.380), and experiencing a higher level of exhaustion (odds ratio = 2.153). Notably, given that spending only 10 additional minutes at work produces a modest decrease in subjective well-being, being at work for several additional hours is likely to be associated with a substantial decrease in these measures of well-being.

Next, we modeled the effects of time spent working during standard hours and nonstandard hours on the three indicators of subjective well-being. Model 2 shows that both working longer during standard hours and working longer during nonstandard hours were related to declines in subjective well-being, which is consistent with our research hypothesis. Spending 10 additional minutes at work during standard hours increased general dissatisfaction (odds ratio = 1.063), time pressure (odds ratio = 1.406), and physical exhaustion (odds ratio = 1.262). Similarly, working 10 more minutes during nonstandard hours was associated with experiencing an increase in life dissatisfaction (odds ratio = 1.096), feeling more pressed for time (odds ratio = 1.288), and greater physical exhaustion (odds ratio = 1.332).

In Model 3, we added an interaction term between standard work hours and nonstandard work hours. The interaction itself was negatively related to general dissatisfaction, time pressure, and fatigue, indicating that the associations between working during nonstandard hours and well-being vary with time spent working during standard hours. Figures 2, 3, and 4 illustrate the associations of working extra minutes during nonstandard schedules with the three indices of subjective well-being across three levels of working minutes during standard schedules (logs of working minutes at standard schedules are 2, 3, and 4). For example, in Figure 2, the y-axis indicates life dissatisfaction levels and the x-axis indicates daily minutes of nonstandard schedules. Standard 2, 3, and 4 indicate logarithmic values of 78.3, 200.8, and 549.9 minutes spent during standard schedules, selected to show how the associations between work hours during nonstandard schedules and subjective well-being behave as work hours during standard schedules increase.

Figure 2.

The association between minutes at nonstandard schedules and life dissatisfaction by minutes at standard schedules.

Figure 3.

The association between minutes at nonstandard schedules and time pressure by minutes at standard schedules.

Figure 4.

The association between minutes at nonstandard schedules and fatigue by minutes at standard schedules.

In Figure 2, the y-axis shows that both working minutes in standard and nonstandard schedules contribute to deterioration in life dissatisfaction, as confirmed in Model 2. For instance, workers who work all day long feel much more dissatisfied about life than those who work only at night. But while adult workers feel dissatisfied more as they work longer during nonstandard schedules, the negative associations between work time during nonstandard schedules and subjective well-being become weak as the work time during standard schedules increases. Clearly working at night or dawn is associated with an increase in the life dissatisfaction in either case, but the influence of extra minutes during nonstandard schedules becomes less as workers spend longer working in the morning or afternoon.

Both Figure 3 and Figure 4 show similar patterns. As working minutes in standard schedules increase, both time pressure and fatigue increased, as obviously shown in y-axis. Therefore, employees who work from morning till night feel more tired and more rushed than those who work from afternoon till night. However, the effect of working minutes in nonstandard hours was higher as working minutes in standard hours decrease, reducing slopes for time pressure and fatigue level.

These observed relationships between work hours, work schedules, and the three indicators of SWB may occur because some Korean workers recognize and even value the corporate culture in which working late into the night is a symbol of the utmost effort and organizational loyalty (Kim et al., 2018). Compared to employees working mainly during nonstandard schedules, those who arrive at work in the morning and stay until late at night can receive recognition from colleagues and the corporation for their hard-working attitudes. As a result, the negative consequences of additional hours worked on a nonstandard schedule may be weaker as work hours during standard schedules are longer. Prior research has found that ‘workaholism,’ a cultural tendency to prioritize work over other activities and to emphasize overindulgence in work, are deep-seated in some cultures (Andreassen et al., 2010) and thus long work hours are associated with increases in happiness or life satisfaction (Okulicz-Kozaryn, 2011; Valente and Berry, 2016). A similar set of relationships between work hours, work schedules, and psychological status may exist in Korea.

A review of the effects of control variables in Table 5 reveals several patterns. In general, women felt more pressed for time and more tired than men. Nevertheless, women were on average more satisfied with their lives than men. Age was negatively associated with feeling rushed, tired, and life satisfaction. Those who feel healthy also report feeling less rushed, less worn out, and less dissatisfied. People with higher educational attainment are less dissatisfied with their lives.

Table 5.

Ordered logit models of the three indices of subjective well-being for adult workers.

	Model 1			Model 2			Model 3
	Life dissatisfaction	Time pressure	Fatigue	Life dissatisfaction	Time Pressure	Fatigue	Life dissatisfaction	Time pressure	Fatigue
	Coefficients	Coefficients	Coefficients	Coefficients	Coefficients	Coefficients	Coefficients	Coefficients	Coefficients
	(SE)	(SE)	(SE)	(SE)	(SE)	(SE)	(SE)	(SE)	(SE)
Total daily minutes (log)	0.306***	0.867***	0.767***
	(0.048)	(0.049)	(0.052)
Daily minutes at standard schedules (log)				0.061*	0.339***	0.233***	0.288***	0.580***	0.423***
				(0.026)	(0.026)	(0.028)	(0.060)	(0.060)	(0.065)
Daily minutes at nonstandard schedules (log)				0.092***	0.253***	0.287***	0.374***	0.555***	0.524***
				(0.016)	(0.016)	(0.017)	(0.069)	(0.069)	(0.075)
Standard (log) × Nonstandard (log)							−0.084**	−0.089**	−0.070***
							(0.020)	(0.029)	(0.022)
Women	−0.258***	0.396***	0.381***	−0.252***	0.407***	0.411***	−0.250***	0.410***	0.412***
	(0.046)	(0.046)	(0.050)	(0.047)	(0.046)	(0.050)	(0.047)	(0.046)	(0.050)
Age	0.005*	−0.024***	−0.024***	0.005	−0.025***	−0.025***	0.006*	−0.025***	−0.025***
	(0.003)	(0.003)	(0.003)	(0.003)	(0.003)	(0.003)	(0.003)	(0.003)	(0.003)
Health	−0.727***	−0.381***	−0.679***	−0.729***	−0.385***	−0.679***	−0.726***	−0.382***	−0.677***
	(0.027)	(0.026)	(0.029)	(0.027)	(0.026)	(0.029)	(0.027)	(0.026)	(0.029)
Education	−0.151***	0.038	−0.045	−0.156***	0.026	−0.052*	−0.150***	0.033	−0.047
	(0.025)	(0.025)	(0.027)	(0.025)	(0.025)	(0.027)	(0.025)	(0.025)	(0.027)
Marital status (ref: Single)
Married	−0.324***	0.304***	0.369***	−0.321***	0.304***	0.383***	−0.316***	0.308***	−0.388***
	(0.075)	(0.074)	(0.079)	(0.075)	(0.074)	(0.080)	(0.075)	(0.074)	(0.080)
Divorced	0.213*	0.151	0.381***	0.216*	0.158	0.389***	0.214*	0.155	0.388***
	(0.108)	(0.107)	(0.115)	(0.108)	(0.107)	(0.116)	(0.108)	(0.107)	(0.116)
Widowed	−0.052	0.210	0.503***	−0.061	0.175	0.487***	−0.049	0.186	0.495***
	(0.141)	(0.143)	(0.152)	(0.141)	(0.143)	(0.152)	(0.141)	(0.143)	(0.152)
Occupation (ref: Agricultural)
Managerial/professional	−0.237	−0.145	−0.614***	−0.197	−0.072	−0.522**	−0.200	−0.076	−0.526**
	(0.156)	(0.155)	(0.168)	(0.156)	(0.115)	(0.168)	(0.156)	(0.155)	(0.168)
Office	−0.109	−0.170	−0.704***	−0.047	−0.047	−0.547***	−0.056	−0.060	−0.559***
	(0.156)	(0.155)	(0.169)	(0.156)	(0.156)	(0.169)	(0.157)	(0.156)	(0.169)
Service	0.016	−0.102	−0.518***	0.038	−0.053	−0.490**	0.031	−0.065	−0.499**
	(0.149)	(0.149)	(0.162)	(0.149)	(0.148)	(0.161)	(0.149)	(0.148)	(0.161)
Labor	0.119	−0.159	−0.493**	0.162	−0.066	−0.405*	0.145	−0.087	−0.422**
	(0.149)	(0.148)	(0.161)	(0.149)	(0.148)	(0.161)	(0.331)	(0.148)	(0.161)
Employment status (ref: Regular)
Employer	0.086	0.135*	0.165**	0.038	0.021	0.050	0.075	0.061	0.080
	(0.053)	(0.053)	(0.056)	(0.053)	(0.053)	(0.056)	(0.053)	(0.053)	(0.057)
Temporary	0.165**	−0.213***	−0.086	0.129*	−0.208***	−0.155*	0.153**	−0.256***	−0.136*
	(0.057)	(0.056)	(0.061)	(0.057)	(0.056)	(0.061)	(0.057)	(0.056)	(0.061)
Family	−0.116	0.119	0.108	−0.178	0.016	−0.051	−0.135	0.028	−0.017
	(0.103)	(0.103)	(0.110)	(0.102)	(0.103)	(0.111)	(0.103)	(0.103)	(0.111)
Individual income levels	−0.093***	0.032***	0.026**	−0.090***	0.038***	0.031**	−0.093***	0.035***	0.029**
	(0.009)	(0.009)	(0.010)	(0.009)	(0.009)	(0.010)	(0.009)	(0.009)	(0.010)
Preschool children	−0.032	0.298***	0.219**	−0.035	0.287***	0.231**	−0.030	0.295***	0.217***
	(0.044)	(0.045)	(0.047)	(0.044)	(0.045)	(0.047)	(0.044)	(0.045)	(0.047)
School children	0.056	0.096***	0.119***	0.056	0.096***	0.124***	0.059*	0.100***	0.126***
	(0.029)	(0.029)	(0.031)	(0.029)	(0.029)	(0.031)	(0.029)	(0.029)	(0.031)
Rural	−0.114	0.089	−0.317*	−0.117	0.068	−0.328*	−0.103	0.081	−0.317*
	(0.122)	(0.120)	(0.130)	(0.122)	(0.120)	(0.130)	(0.122)	(0.120)	(0.130)
Weekend	0.010	0.165***	0.135**	−0.020	0.110*	0.069	−0.001	0.129**	0.083
	(0.045)	(0.045)	(0.048)	(0.045)	(0.045)	(0.048)	(0.045)	(0.045)	(0.048)
R ²	0.058	0.043	0.062	0.058	0.043	0.066	0.058	0.044	0.066
N	9993

p < .05; ** p < .01; *** p < .001.

While married respondents felt more time pressure and physical fatigue than those who had never been married, they were more satisfied than the latter. Married people tend to have higher demands for family care or responsibilities compared to single people. Compared to those who work in agriculture, people who work in other sectors felt less tired, on average. Temporary workers tend to feel less time pressure but more dissatisfaction with their lives compared to permanent or regular workers.

Income level was positively associated with time pressure, fatigue, and life satisfaction. Having children consistently increased time pressure and fatigue, regardless of the age of the children. Those living in rural areas reported feeling less tired than those living in urban areas. Those who worked and answered the survey on the weekends tended to feel more time pressure and fatigue than their counterparts. These results are generally consistent across models, with little change in coefficients.

Discussion and conclusion

Work is an important part of daily life, not only for earning income but also for building social relationships and self-identity. However, working excessive hours or to a nonstandard schedule may have negative repercussions on an individual’s family life, relationships, health, and life satisfaction. Despite increasing social concern about the relationship between work and individual well-being, most previous studies have focused on the effect of total working hours without considering when those hours are worked. Further, most extant research has been conducted in Western countries while paying little attention to countries that are famous for long work hours, such as South Korea.

Based on the analysis of data from the 2014 Korean Time Use Survey, we found that most working people in Korea worked between 9:00 a.m. and 6:00 p.m. After 6:00 p.m., the proportion of people working decreased, but approximately 10–20% were still working at 10:00 p.m. In the late night and early morning hours, the proportion working remained under 5%. Depending on when and how long people work, we divided workers into three categories: standard workers (47%), long workers (43%), and late workers (10%).

As in other countries such as the United States and Belgium (Glorieux et al., 2008; Hamermesh, 1995; Minnen et al., 2016), both long and late workers were more likely to have less educational attainment and to work in the service sector or manual labor compared to standard workers. Particularly, late workers tended to be temporary workers with limited job stability. The average work time was 7 hours per day in the full sample. However, long workers spent an average of 8 hours, with some working longer than 10 hours per day. Late workers had a greater variation in daily working hours, from 2 to 10 hours.

The results of ordered logistic regression analysis supported our research hypotheses. Total time spent working was positively associated with time pressure, fatigue, and life dissatisfaction, controlling for socioeconomic characteristics. Both the amount of time worked during standard hours and during nonstandard hours increased time pressure and fatigue and decreased life satisfaction. However, adding an interaction term revealed that spending more time at work during nonstandard hours had a stronger positive effect on tiredness, feeling rushed, and life dissatisfaction as workers spend less time working standard hours. These results suggest that longer work hours have a detrimental effect on individual SWB but the effect is stronger for those working in nonstandard hours. As mentioned above, the strong work ethic prevalent in Korea may offset the negative effects of an increase in working time during standard hours.

This study has certain limitations. First, the small number of the respondents who worked only during nonstandard hours prevented us from analyzing long workers and late workers separately, even though working in nonstandard work schedules may affect these two groups differently. Second, we did not limit the sample to a specific age group or life cycle, although these subgroups may have different work experiences.

Despite these limitations, this study contributes to the scholarly literature on Korean people’s working lives and sheds light on how work hours and schedules affect time pressure, fatigue, and life dissatisfaction. The study also raises concerns and issues for scholars and policymakers. Given that both long and late workers tend to have lower educational attainment and less stable employment than standard workers, the negative effects of working a nonstandard schedule on SWB may increase socioeconomic discrepancies. If the recent trend toward a 24/7 economy continues, more and more people with low levels of human capital will be working long hours during nonstandard times. This pattern may lead to a polarization of work structure in which white-collar workers with high levels of human capital have the autonomy to choose work times and schedules but workers with low levels of human capital do not. Therefore, researchers should pay more attention to the effects of both total working hours and various types of work schedules on the well-being of individuals.

Footnotes

Acknowledgements

Yoo-Jean Song appreciates the support from Dong-A University.

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 received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Yun-Suk Lee

Author biographies

Yoo-Jean Song is a professor in the Department of Sociology at Dong-A University in Korea. Her research interests include fatherhood and childcare, division of labor at home, and how people spend their time in daily life and its association with life satisfaction. In particular, she has recently examined how husbands and wives in dual-earner couples in Korea orchestrate time allocations, using time diary studies.

Yun-Suk Lee is a professor in the Department of Urban Sociology at the University of Seoul in Korea. He is interested in diverse forms of families, family relationships, and time use patterns among family members. His recent papers appear, among others, in Journal of Comparative Family Studies, Asian Journal of Social Science, International Migration, and Journal of Women & Aging.

References

Andreassen

Hetland

Molde

Pallesen

(2010) ‘Workaholism’ and potential outcomes in well-being and health in a cross-occupational sample. Stress and Health 27: e208–e214.

Cha

Eun

(2014) Gender difference in sleep problems: Focused on time use in daily life of Korea. Social Indicators Research 119(3): 1447–1465.

Cha

Song

(2017) Time or money: The relationship between educational attainment, income contribution, and time with children among Korean fathers. Social Indicators Research 134(1): 195–218.

Craig

Brown

(2014). Weekend work and leisure time with family and friends: Who misses out? Journal of Marriage and Family 76(4): 710–727.

Craig

Brown

(2017) Feeling rushed: Gendered time quality, work hours, nonstandard work schedules, and spousal crossover. Journal of Marriage and Family 79(1): 225–242.

Craig

Mullan

(2013) Parental leisure time: A gender comparison in five countries. Social Politics 20(3): 329–357.

Fenwick

Tausig

(2001) Scheduling stress: Family and health outcomes of shift work and schedule control. American Behavioral Scientist 44(7): 1179–1198.

Glorieux

Inge

Minnen

(2008) The coming of the 24-hour economy? Time and Society 17(1): 63–83

Gracia

Kalmijn

(2016) Parents’ family time and work schedules: The split-shift schedule in Spain. Journal of Marriage and Family 78(2): 401–415.

10.

Hamermesh

(1995) Who works when? Evidence from the US and Germany. NBER working paper.

11.

Harrington

(2001) Health effects of shift work and extended hours of work. Occupational and Environmental Medicine 58: 68–72.

12.

Kalil

Ziol-Guest

Levin Epstein

(2010) Nonstandard work and marital instability: Evidence from the National Longitudinal Survey of Youth. Journal of Marriage and Family 72(5): 1289–1300.

13.

Keizer

Dykstra

Poortman

(2010) The transition to parenthood and well-being: The impact of partner status and work hour transitions. Journal of Family Psychology 24(4): 429–438.

14.

Kim

McLean

Park

(2018) The cultural context of long working hours: Workplace experiences in Korea. New Horizons in Adult Education and Human Resource Development 30(2): 36–51.

15.

Kim

(2018) Precarious employment among South Korean women: Insights from a comparison to France. Mathematical Population Studies 25(1): 41–61.

16.

Kleiner

Pavalko

(2010) Clocking in: The organization of work time and health in the United States. Social Forces 88(3): 1463–1486.

17.

Luttmer

(2005) Neighbors as negatives: Relative earnings and well-being. The Quarterly Journal of Economics 120(3): 963–1002.

18.

Mills

Täht

(2010) Nonstandard work schedules and partnership quality: Quantitative and qualitative findings. Journal of Marriage and Family 72(4): 860–875.

19.

Minnen

Glorieux

Tienoven

TPV

(2016) Who works when? Towards a typology of weekly work patterns in Belgium. Time and Society 25(3): 652–675.

20.

Niu

Chung

Chen

, et al. (2011) The effect of shift rotation on employee cortisol profile, sleep quality, fatigue, and attention level: A systematic review. Journal of Nursing Research 19(1): 68–80.

21.

OECD (2018) Available at: https://stats.oecd.org/index.aspx?DataSetCode=ANHRS

22.

Okulicz-Kozaryn

(2011) Geography of European life satisfaction. Social Indicators Research 101(3): 435–445.

23.

Oshio

Nozaki

Kobayashi

(2013) Division of household labor and marital satisfaction in China, Japan, and Korea. Journal of Family and Economic Issues 34(2): 211–223.

24.

Presser

(2000) Nonstandard work schedules and marital instability. Journal of Marriage and Family 62(1): 93–110.

25.

Robertson

Cooper

(2011) Well-being: Productivity and Happiness at Work. Basingstoke: Palgrave Macmillan.

26.

Rudolf

(2014) Work shorter, be happier? Longitudinal evidence from the Korean five-day working policy. Journal of Happiness Studies 15(5): 1139–1163.

27.

Strazdins

Clements

Korda

, et al. (2006) Unsociable work? Nonstandard work schedules, family relationships, and children’s well-being. Journal of Marriage and Family 68(2): 394–410.

28.

Totterdell

(2004) Work schedule. In: Barling

Kelloway

Frone

(eds) Handbook of Work Stress. London: Sage, pp. 35–62.

29.

Valente

Berry

(2016) Working hours and life satisfaction: A cross-cultural comparison of Latin America and the United States. Journal of Happiness Studies 17(3): 1173–1204.

30.

Wharton

Blair-Loy

(2006) Long work hours and family life: A cross-national study of employees’ concerns. Journal of Family Issues 27(3): 415–436.