Men and Women's Different Dreams on the Future of the Gendered Division of Paid Work and Household Work after COVID-19 in South Korea.

Men's long hours of paid work and minimal commitment to household work, combined with the comparably low-level of women's labor force participation, characterize the gendered division of work and family in South Korea. Can the changes in work and family arrangements brought on by the COVID-19 pandemic alter the persevering gendered division of paid work and household work in Korea? Along with school closures and the increased number of employees working from home during COVID-19, do Korean men and women anticipate more equal sharing of paid work and household work? We collected data from 1000 Korean adults during the period of July 3-6, 2020, and asked their predictions about various dimensions of social changes, including the gendered division, after COVID-19. Although a substantial share of both men and women anticipate a reduction in the gendered division of paid work and household work after COVID-19, Korean women are not as optimistic as their male counterparts about this potential reduction. In particular, younger women are most skeptical about the prospect that paid work and household work will be less divided by gender beyond the pandemic.

Introduction

In 2018, an average worker in South Korea worked for 1,999 hours, higher than the average number of hours worked for each OECD country besides Mexico (2,149 hours) (OECD, 2020a). Considering that only half of Korean women aged 15 or older participate in the labor force, long hours of work reveal a substantial gendered division of paid work and household work (including family care). According to OECD statistics, Korean men aged 15 to 64 spend only 49 minutes per day in unpaid work (household and care work), the second lowest amount of time next to Japanese men (40 minutes) (OECD, 2020b). In comparison, men in Denmark and the United States spend 186 and 145 minutes per day, respectively, in unpaid work. The persistent gendered division of work and family in Korea is especially striking, given that young Korean women as well as men are one of the most educated populations in the world (OECD, 2018; Park, 2007). Despite rapid expansion of higher education among both women and men, a large share of Korean women still leave the labor force after marriage and childbirth, resulting in the continued M-shaped pattern of age and labor force participation.

Can the changes in work and family arrangements brought on by the COVID-19 pandemic alter the persevering gendered division of paid work and household work? Beginning in mid-February 2020, as the number of COVID-19 cases in Korea soared, more workers, especially in the public sector, started to work from home. The number of public-sector employees working from home has increased 22 times in the first quarter of 2020 compared to the first quarter of 2019 (Hanykung, 2020). Although systematic research from this time period is not yet readily available, many news articles report the increase in employees working from home since COVID-19, and allude to the possibility that remote work will expand in the Korean labor market even after COVID-19 (Kim, 2020; Shin, 2020). Along with more people working from home since the COVID-19 outbreak, household work and family care responsibilities may have increased as well. In particular, married couples with young children have faced the challenge of parenting and homeschooling in addition to working from home, as most K-12 schools in Korea were closed until the end of May.

Over the last few months, researchers from a wide range of countries have conducted studies, and have assessed how the gendered division of paid work and household work has changed or persisted since COVID-19. Some studies in the United States and United Kingdom have shown that fathers spent more time in household work and/or childcare during the pandemic, although mothers still spend more time doing this unpaid work than fathers (Andrew et al., 2020; Carlson, Petts, & Pepin, 2020). Another study expects that “the [COVID-19] crisis is likely to generate a large, if temporary, upward sift in men’s participation in childcare, with a sizable fraction of married men taking the main responsibility, in most cases for the first time” (Alon, Doepke, Olmstead-Rumsey, & Tertilt, 2020: p.22). From a recent study showing that the level of generalized social trust in Korea has actually increased in the midst of COVID-19 (Kye & Hwang, 2020), hinted is a possibility of diverse changes in Korean society after COVID-19 even in other dimensions such as gender inequality.

However, other scholars are more skeptical about reduction of gender inequality in work and family after COVID-19. Women were often more likely to lose their job than men during COVID-19 (Kristal & Yaish, 2020), which likely has increased gender disparities in household work time. Moreover, men’s increase in household work and childcare, particularly due to increased hours of working from home, may be limited especially compared to women’s increased household work and childcare, exacerbating gender inequality (Farré, Fawaz, González, & Graves, 2020; Lyttelton, Zang, & Musick, 2020).

Potential implications of work and family changes due to COVID-19 for the gendered division of paid work and household work are particularly significant in the Korean context, since Korea has long maintained large gender disparities in household work. In this study, we address this issue by assessing the extent to which Korean men and women agree with the statement that after COVID-19 the gendered division of paid work and household work will be reduced in Korean society. Instead of directly measuring time spent by men and women on paid work and family, we examine gender differences in the anticipation whether gender disparities in paid work and household work will be reduced after COVID-19. Although the actual change in men’s contribution to household work is important, we argue that men and women’s perception of whether such changes will lessen overall gender disparities is critical as a litmus with which to gauge the full implication of the COVID-19 pandemic for family and gender in contemporary Korea. Even if Korean men have done more household work since COVID-19 than they did in the past, men’s increased commitment to household work may not be significant if women do not perceive it as enough to produce substantial change in terms of gender inequality.

Popular media has portrayed the diverse stories of husbands in Korea during COVID-19, and has highlighted husbands’ increased participation in household work (including childcare), particularly for those whose wives are essential workers (Woo, 2020). However, it is difficult to assess how much men participate in household work in reality. It is also important to recognize that the degree of men’s increased contribution to household work may be perceived differently by men and women. Fathers often tend to overestimate their contribution to household work relative to the amount of work they actually do or relative to what mothers perceive (e.g., Miller, 2020). Therefore, compared to women, men may be more optimistic on potential reduction in the gendered division of paid work and household work after COVID-19.

Data and methods

In order to assess how men and women differ in their anticipation of societal changes, after COVID-19, including the gendered division of work and family, we included questions in an existing online survey conducted by a survey company, Hankook Research. The company maintains a panel of more than 500,000 registered individuals, and the final sample for our survey was obtained by randomly choosing out of those registered individuals stratified by age, gender, and region. Therefore, although it is not a probability sample, our sample represents the adult Korean population (see Kristal & Yaish, 2020 for a similar kind of online survey in Israeli). Among 1,336 individuals who participated in the survey with our question items, 1,000 completed it between July 3 and 6, 2020. It is important to note that the COVID-19 pandemic peaked in Korea from the end of February to March 2020. In early July 2020, when our survey was conducted, the number of newly infected cases was around 30.1

Our survey questions asked respondents to what extent they anticipate societal changes after COVID-19 across the following 8 dimensions: 1) the Korean economy will get worse; 2) income inequality will rise; 3) generational gaps (between older and younger generations) will widen; 4) the gendered division of paid work and household work will decrease; 5) ideological conflicts between conservatives and progressives will intensify; 6) the likelihood of personal data breach will increase; 7) native Koreans’ hostility toward immigrants and foreign residents will deepen and 8) collective consciousness will become more important. Respondents indicated the extent to which they agreed to each statement on a 5-point scale – 1) strongly agree, 2) somewhat agree, 3) neither agree nor disagree, 4) somewhat disagree, and 5) strongly disagree.

The survey also collected a variety of information on respondents’ demographic and socioeconomic characteristics. Age is a continuous variable from 18 to 88 years old. We distinguish respondents who have a two-year or four-year college degree or higher from those who do not using a binary variable for education. Marital status is a binary variable separating respondents who are currently married from all others. Household income was originally measured on the basis of 11 ordered categories, but we treat it as a continuous measure in our multivariate analysis. We also take into account whether respondents are working for pay or not. Respondents marked their political ideology on the scale of 0 (very progressive) to 10 (very conservative). We classify respondents into progressive (0-4), central (5), and conservative (6–10). 47 respondents who answered ‘don’t know’ are classified as central. Note that the survey did not ask for the amount of time that respondents spent in paid work and household work.

Based on simple mean tests for men and women, we first describe gender differences in the extent to which respondents agree with the statement that the gendered division of paid work and household work will decrease after COVID-19. In order to see how men and women differ in their anticipation of societal changes in other dimensions, we present mean differences in each of remaining 7 measures. Then, we move to a multivariate analysis by applying an ordered logistic regression to predict odds of a higher category (i.e., toward more disagreement) than a lower category (i.e., toward more agreement) (on the 5-point scale) in their anticipation of reduction in work-family gendered division after COVID-19, after controlling for demographic and socioeconomic variables.

Descriptive: Gender Differences in Anticipation of Reduction in the Gendered Division of Paid Work and Household Work After COVID-19

The first set of bar plots in Fig. 1 shows mean values (on the 5-point scale) for men and women, respectively, in the extent to which they agree that the gendered division of paid work and household work will decrease in Korea after COVID-19. The mean value of 2.83 for women is significantly greater than the mean value of 2.66 for men (p < .01). In other words, compared to men, women are more likely to disagree with the statement that the gendered division of labor will decrease (1 = strongly agree, 2 = somewhat agree, 3 = neither agree nor disagree, 4 = somewhat disagree, 5 = strongly disagree). Korean women are more skeptical than men about potential reduction in the work-family gender divide after COVID-19. It is interesting to see that the mean values on this gendered division variable are greatest for both men and women among mean values for all 8 outcome measures presented in Fig. 1. Although Korean women are more pessimistic than men about the possibility that the gender divide will lessen after COVID-19, both men and women also anticipate less change in the gendered division of work and family than any other changes of society.

Fig. 1

How Do You Agree to the Statement? (1 = Strongly Agree, 2 = Somewhat Agree, 3 = Neither Agree Nor Disagree, 4 = Somewhat Disagree, 5 = Strongly Disagree). Note: ** p < .01.

Another notable pattern in Fig. 1 is that gender differences in anticipation are generally not significant in most social changes besides the gendered division of labor. Korean women are very similar with men in their expectation of potential impacts of COVID-19 on most dimensions including the economy, income inequality, generational gaps, personal data breach, attitudes toward foreigners, and collective consciousness. Besides the gendered division of labor statement, the only other statement for which gender difference is significant is whether ideological conflicts between conservatives and progressives will deepen.

Results of Ordered Logistic Regression

Table 1 presents the results of ordered logistic regression on the extent to which respondents disagree with the statement that the gendered division of paid work and household will decrease after COVID-19. Model 1 does not include an interaction term between gender and age, while Model 2 does include this interaction term. The coefficients presented in the table refer to log odds. The coefficient of female (vs. male), 0.44 in Model 1 means that the odds of choosing a response toward more disagreement (out of five categories from strongly agree to strongly disagree) on the statement that the gendered division of paid work and household work will decrease are 3.6 times (exp 0.44 = 3.56) larger for women than men, holding all other variables constant. In other words, the finding of Fig. 1 that women are significantly more skeptical than men remains robust even after controlling for gender differences in demographic and socioeconomic characteristics.

Table 1

Ordered Logistic Regression on the Extent to Which Respondents Disagree with the Statement that the Gendered Division of Paid Work and Household Work Will Decrease after COVID-19 (1 = strongly agree, 2 = somewhat agree, 3 = neither agree not disagree, 4 = somewhat disagree, 5 = strongly disagree).

		Model 1	Model 2
Women (ref. Men)		0.438***	1.269**
		(0.123)	(0.417)
Age (range: 18-88)		−0.018***	−0.008
		(0.005)	(0.006)
Women × Age			−0.018*
			(0.008)
Education attainment (ref. high school or less)
	College	−0.040	−0.058
		(0.126)	(0.127)
Household income		−0.005	−0.002
		(0.026)	(0.027)
Political ideology (ref. Central)
	Progressive	0.370**	0.354**
		(0.133)	(0.133)
	Conservative	0.086	0.083
		(0.154)	(0.154)
Marital status (ref. not married)
	Married	−0.113	−0.140
		(0.143)	(0.146)
Working status (ref. not working)
	working	0.122	0.083
		(0.130)	(0.130)
Cut-points
	1	−2.918***	−2.522***
		(0.290)	(0.353)
	2	−0.827**	−0.432
		(0.265)	(0.330)
	3	0.887***	1.289***
		(0.265)	(0.332)
	4	2.456***	2.869***
		(0.303)	(0.368)
Observations		1,000
Log pseudolikelihood		−1366.9	−1364.3

p < 0.05.

p < 0.01.

p < 0.001.

In addition to gender, age and political ideology are significantly associated with respondents’ anticipation that the gendered division of paid work and household work will be reduced. Older respondents are less likely to choose a response toward more disagreement (i.e., they are more likely than younger respondents to agree that the gendered division of paid work and household work will decrease after COVID-19). Stated differently, younger Koreans are more pessimistic about the future of a more equitable gendered division of paid work and household work after COVID-19. Respondents who consider themselves as progressives are more likely to disagree with potential reduction in the gendered division after COVID-19 than their central counterparts. Interestingly, none of other socioeconomic and demographic characteristics is significantly associated with anticipation of reduction in the gendered division.

Model 2 reveals that gender differences in anticipation of reduction in the gendered division depend on age, given that the interaction term between female and age is statistically significant. The negative coefficient of the interaction term indicates that women’s higher odds of choosing a response toward more disagreement become less evident along with age. In other words, relative to men, women increasingly become more likely to agree with the statement that the gendered division of paid work and household work will decrease after COVID-19 as they get older. To facilitate interpretation of the interaction term, Fig. 2 presents how the predicted probability of choosing each response category changes along with age for men and women, respectively. The figures are based on the coefficients in Model 2 in Table 1 with all other variables except for female and age fixed at their means. For both men and women, the predicted probabilities of strongly agreeing and somewhat agreeing that there will be a less gendered division increase along with age, while the predicted probabilities of neither agreeing nor disagreeing, somewhat disagreeing, and strongly disagreeing decrease along with age. In other words, older Koreans are more optimistic that the gendered division of paid work and household work will decrease after COVID-19 than their younger counterparts. However, it is notable that changes in those predicted probabilities across ages are more substantial among women than men. Therefore, the difference in their anticipation between older and younger Koreans is much more visible for women than men.

Fig. 2

Changes across Age in Predicted Probabilities of Each Response Category to the Statement that the Gendered Division of Paid Work and Household Work Will Decrease after COVID-19? (from Model 2 in Table 1).

The more substantial age variation among women than men is more clearly revealed in Fig. 3 that separately presents only predicted probabilities of responding “strongly agree” and or “somewhat agree” to the statement that the gendered division of paid work and household work will reduce after COVID-19. The predicted probabilities for both responses do not change much across age among men. But those predicted probabilities increase across age among women. In the results, after age 70, women are actually more likely to “strongly agree” or “somewhat agree” with the statement that there will be less of a gendered division after COVID-19.

Fig. 3

Changes across Age in Predicted Probabilities of “Strongly Agree” or “Somewhat Agree” with the Statement that the Gendered Division of Paid Work and Household Work Will Decrease after COVID-19? (from Model 2 in Table 1).

Conclusion

Almost half (48%) of Korean adult men in our sample strongly or somewhat agree that the gendered division of paid work and household work will be reduced after COVID-19. Along with the expanded practice of working from home after COVID-19, Korean men may feel that more men are now contributing to household work (including childcare) than before. Given Korean men’s comparably minimal commitment to household work so far, even a little change, if any, may be perceived as a significant change from the male perspective. However, Korean women are not as optimistic as their male counterparts about the potential reduction in the gendered division. It is true that still 37 percent of Korean women strongly or somewhat agree with the statement that the gendered division will decrease after COVID-19. However, apparently, the share of Korean women who are optimistic on the future of the gendered division of work and family is 11 percentage-points lower than the corresponding share of Korean men. Our analysis, furthermore, indicates that this relative reservation about the potential reduction in the gendered division after COVID-19 is particularly evident among younger, rather than older, Korean women. Our model shows that women aged 70 or older anticipate that the gendered division of paid work and household work will decrease after COVID-19 as much as or even more so than their male counterparts, while younger women are more skeptical about this prospect than their male peers. This pattern in Korea demonstrates another example with which to highlight heterogeneous impacts of COVID-19 on behavior and perceptions across different demographic groups.