hiring domestic help © the author(s) 2015 in hong kong ... · domestic helpers increased from...

Journal of Family Issues 1 –27

© The Author(s) 2015Reprints and permissions:

sagepub.com/journalsPermissions.nav DOI: 10.1177/0192513X14565700

jfi.sagepub.com

Article

Hiring Domestic Help in Hong Kong: The Role of Gender Attitude and Wives’ Income

Adam Ka-Lok Cheung1 and Lake Lui2

AbstractThe associating factors of hiring domestic help have not been thoroughly studied in a non-Western context. Using household survey data (N = 974), this article investigates the interactive role of gender attitude and women’s income on the decision to hire domestic help in Hong Kong. Some previous studies fall short in finding a significant association between respondents’ gender attitudes and the hiring of domestic help, while wives’ income is a consistent factor in the hiring of domestic help across a number of studies. In this study, we found that husbands’ traditional gender attitudes and wives’ high income sharply increase the likelihood of hiring domestic help. However, their associations with the hiring of domestic help are conditional on each other. In addition to women’s socioeconomic status, ideational factors should be taken into account in projecting local demand for domestic help and in understanding the increasing trend of domestic outsourcing.

Keywordsgender and family, household labor, work and family, domestic help, domestic outsourcing, Hong Kong

1Hong Kong Institute of Education, New Territories, Hong Kong2University of Washington, Seattle, WA, USA

Corresponding Author:Adam Ka-Lok Cheung, Department of Social Sciences, Hong Kong Institute of Education, D3-1/F-22A, 10 Lo Ping Road, Tai Po, New Territories, Hong Kong. Email: [email protected]

565700 JFIXXX10.1177/0192513X14565700Journal of Family IssuesCheung and Luiresearch-article2015

at Hong Kong Institute of Education on January 7, 2015jfi.sagepub.comDownloaded from

mailto:[email protected]

http://jfi.sagepub.com/

2 Journal of Family Issues

Introduction

Working women in postindustrial societies are increasingly caught in work–family conflict. Many of them are struggling to satisfy the demands of two “competing devotions”—family devotion and work devotion—that are both time-intensive and emotionally demanding (Blair-Loy, 2005). In response to this double burden on women, husbands have steadily increased their partici-pation in housework, although they have never reached parity with women. Certain research findings suggest that the increase is only marginal (Baxter, 2002; Bianchi, Milkie, Sayer, & Robinson, 2000; Gershuny, 2000; Hook, 2006; Windebank, 2001). Although some studies show that the total time women spend doing housework has decreased, others suggest that there are “unquantified” household responsibilities that intensify the domestic work-load (DeVault, 1991; Doucet, 2001; Lee, 2002; Lui, 2013).

Past studies show that women have three main options when seeking to resolve work–family conflict (Gershuny, Bittman, & Brice, 2005). First, some women “exit” by either leaving the marriage or altering their employ-ment status. They may leave their jobs or reduce their working hours to accommodate their children and husbands’ schedules. Second, some women “suffer” through what Hochschild (1989) calls “supermoming,” continuing to bear the tension between work and family (Crompton & Lyonette, 2006). Third, other women choose to directly “voice out,” confronting and negotiat-ing with their husbands to address unfair household arrangements (Gershuny et al., 2005; Kluwer, Heesink, & Van de Vliert, 1996). In general, however, many women who use these strategies remain discontented because of their sacrifices (Blair-Loy, 2005), increased conflict with their unmotivated hus-bands, or both (Hochschild, 1989, 1997).

Outsourcing household tasks seems to be a feasible strategy to reduce both work–family conflict and marital conflict (Hochschild, 1989, 1997). A num-ber of recent studies have suggested that households in many postindustrial societies are increasingly outsourcing domestic tasks traditionally carried out by wives, such as meal preparation, housekeeping, and child care (Baxter, Hewitt, & Western, 2009; Bittman, Matheson, & Meagher, 1999; Oropesa, 1993; Soberon-Ferrer & Dardis, 1991).1 As a result, demand for domestic helpers is escalating. In the United Kingdom, household spending on domes-tic workers increased from £1.1 billion in 1987 to £4.3 billion in 1997 (Anderson, 2000). In Hong Kong, the number of full-time, live-in foreign domestic helpers increased from 21,517 in 1982 to 237,104 in 2002 (Task Force on Population Policy, 2003). At least one fifth of Hong Kong house-holds earning HK$40,000 or more a month hired a domestic helper in 2001 (Hong Kong Census and Statistics Department, 2001).



Cheung and Lui 3

The factors that lead households to hire domestic help have not been thor-oughly studied, especially in non-Western contexts. Although past studies show that women’s high absolute income is a consistent factor in hiring domestic help, the role of gender attitude is controversial (e.g., Baxter et al., 2009; Bittman et al., 1999; Oropesa, 1993; Spitze, 1999; van der Lippe, Tijdens, & de Ruijter, 2004). Some studies have asserted that individuals with more traditional gender attitudes are less likely to outsource domestic work, as they may believe that it is a woman’s responsibility to do housework. However, other studies show that there is no significant association between respondents’ gender attitudes and the hiring of domestic help (Baxter et al., 2009; Oropesa, 1993). With couple-level data, van der Lippe, Frey, and Tsvetkova (2013) show a gender-specific effect of gender ideology on hiring domestic help whereby the likelihood of hiring help for tasks such as home maintenance and child care is associated with men’s gender ideologies but not women’s. Given these inconsistent results, our primary research goal is to examine whether and how a woman’s income and an individual’s gender attitude matter in the use of paid help. We hypothesize that there is an interac-tion effect between gender attitude and women’s income.

Studies of women’s autonomy have shown that with higher absolute income, wives can buy themselves out of housework (Gupta, 2006, 2007). However, Blumstein and Schwartz (1983) find that gender ideology is a moderator of income and household arrangements. This implies that a wife’s income does not necessarily translate into power in the household, including in the decision to hire help. A wife with high income might find herself less powerful when negotiating with her husband if he has a traditional gender attitude and is not willing to share housework (Atkinson & Boles, 1984; Tichenor, 1999, 2005). Outsourcing may thus be a more feasible strategy for wives to deal with the conflicts between the family and work domains. Yet some tradition-oriented women do not take pride in their identities as major earners and might equate housework with “women’s work.” Therefore, it is also plausible that high-income but tradition-oriented wives might accentuate their “femininity” by deemphasizing their earnings and devoting more time and attention to their families. In this sense, wives with high income and traditional gender attitudes are less likely to hire help than are high income, egalitarian wives.

Using household survey data from Hong Kong (N = 974), we focus on the interaction effect of individual-level gender attitudes and women’s income on hiring domestic help. Hong Kong is a good testing ground for studying the fac-tors associated with domestic outsourcing because its family devotion and work devotion schemas are both strong. Traditional gender ideology still pre-vails in this Chinese society despite the rising economic contribution of women and the increasing expectation that women will work outside the home.




Literature Review and the Present Study

Household Resources and Hiring Domestic Help

The resource hypothesis posits that greater resources, in terms of wealth, total household income, and the husband’s and wife’s income, allow families to have more funds for domestic services and thus more options to manage time (Baxter et al., 2009; Cohen, 1998; Oropesa, 1993; Spitze, 1999; van der Lippe et al., 2004; Windebank, 2001). Spending time on domestic work is not attractive because the opportunity cost of household members’ time in terms of potential earnings in the labor market is higher than the cost of hiring domestic help. Warren (2003) uses the British Household Panel Survey of 1995 to show that 13% of the middle-class, dual-earner couples surveyed reported using a third party to do household cleaning, whereas this was the case for only 3% of work-ing-class couples. Van der Lippe et al. (2004) find similar results in the Netherlands and assert that household resources are the most important factor in hiring domes-tic help for dual-earning couples. Similarly, Hanson and Ooms (1991) find that higher income families in the United States often employ domestic help.2

When we break down household income into individual earnings, the results show that the impact of income on outsourcing is not gender-neutral. Treas and de Ruijter (2008) suggest that women’s earned income is more important than men’s for determining spending on traditionally female tasks such as laundry, housekeeping, and preparing meals. Cohen (1998) finds that the wife’s income predicts twice the spending on housekeeping as the hus-band’s income. Oropesa (1993) provides a similar result, showing that wives use their income to “buy out” their time spent on housekeeping. Wives’ income, rather than husbands’, is also more important for purchasing child care (Brayfield & Hofferth, 1995). Consequently, women’s rising income is consistently considered in the literature to be one of the most important fac-tors in projecting the demand of the domestic service economy.

Cohen (1998) provides some caveats that should be considered when interpreting the importance of women’s income. Instead of a gender-neutral approach, he alludes to the relevance of gender ideology in the interpretation of income. Citing Gershuny and Robinson (1988), he argues that

some women may only seek paid employment if they are able to make compatible housework arrangement; also women who have fewer children, or who have more cooperative husbands, may be able to keep their jobs longer or advance further than others. (Cohen, 1998, p. 222)

This implies that the effect of wives’ income on hiring decisions could be confounded by gender ideologies that correlate with both “the wife’s income” and “hiring decisions” at the same time.



Cheung and Lui 5

Gender Ideology, Household Division of Labor, and Hiring Help

“Gender ideology functions as a lens through which many social processes and events are viewed, interpreted, and acted upon, based on the belief about what the relationship between women and men should be” (Davis & Greenstein, 2009, p. 100). In the literature on the division of household labor, gender ideology informs “what sphere the person wants to identify with (home or work) and how much power in the marriage she or he wants to have (less, more, or the same amount)” (Hochschild, 1989, p. 15). Traditionally, housework and child care are viewed primarily as the respon-sibilities of women, whereas breadwinning is primarily the responsibility of men. Doing housework is also an integral part of “doing gender” for women—an avenue on which to display a woman’s love for her family and subordination to her husband (Berk, 1985). However, with the ongoing women’s revolution and changing world economy, there is an emerging egalitarian “code of honor and identity for men and women that fits the evolving circumstances” (Hochschild, 1989). Gender ideologies are becom-ing more variegated, with Hochschild (1989) delineating three main types—traditional, transitional, and egalitarian—that are measured based on marital power and the expectation of a production role.

Many studies find that men with egalitarian gender ideologies take a greater share of the household labor responsibilities and have greater paternal involvement in child care (Bianchi et al., 2000; Blair & Lichter, 1991; Coltrane & Ishii-Kuntz, 1992; Evertsson, 2014; Presser, 1994). Qualitative studies also show that egalitarian men define success by their relationships with their children more so than by their paid work (Coltrane, 1998; Hochschild, 1989). In contrast, studies on women’s gender ideologies sug-gest that traditional women “do gender” more often and are less likely than egalitarian women to perceive unfairness in the unequal division of labor, and thus are likely to do more housework (Davis & Greenstein, 2009; Evertsson, 2014).

As Hochschild (1989) suggests, gender ideals sometimes have to be negotiated within the framework of reality (given the constraints of time, energy, and money, for instance). Outsourcing domestic work, according to Hochschild (1989), is a gender strategy ideal for women who are “cautious,” “less disturbing,” and “compatible,” as it avoids the need to make men change. Groves and Lui (2012) assert that “hiring help” is sometimes a strat-egy for men to buy “the gift” of domestic help for their wives so that they can be released from housework. However, these studies are limited, as they do not tease out how the effects of men’s and women’s gender attitudes on the decision to employ assistance differ along with women’s various income




levels. In this study, we aspire to uncover how people with different ideolo-gies act in different realities based on economic conditions and levels of demand in the family.

There are two sides to the argument that explains the effect of gender ide-ology on hiring domestic help. Past studies tend to hypothesize that tradi-tional wives are less likely to hire help because they view housework and child care as an “essential” and “natural” aspect of a woman’s identity. A traditional husband, emphasizing this gender boundary, may also feel frus-trated if his wife does not take up the “wife’s responsibility.” In this context, hiring help is considered “inappropriate” and “undesirable” because it chal-lenges this gender expectation and does not uphold a woman’s identity as “a caring wife and a loving mother” (Baxter et al., 2009).

In contrast, there will be less demand for domestic help if the husband has egalitarian views and is willing to share the labor. None of the respondents with egalitarian ideologies in the Hochschild (1989) study had hired help. Instead, they mostly shared the household work with their spouses. Some egalitarian fathers, given the option of outsourcing, would rather take up household labor than hand it over to “a stranger” (Lui, 2013). Furthermore, as Ehrenreich and Hochschild (2003) note, some women with egalitarian gender ideologies may avoid employing other women (especially those from the Third World) to do less attractive domestic tasks, as they may feel that it is exploitative. It is possible that the idea of hiring domestic help is still associ-ated with the image of the exploited muijai3 from previous eras. Abusive practices of employers toward foreign domestic helpers are familiar to people in Hong Kong. As Constable (2007, p. 52) suggests,

Filipinas, like muijai, are often not paid their wages, fed leftovers, forced to work long hours until they become sick with exhaustion . . . the uses to which muijai were put also resemble the illegal work often required by the employers of foreign domestic workers. Contracts stipulate that the foreign domestic helpers must work only for the employer who signed the contract, but employers of foreign workers, like those of muijai, loan them out to friends, put them to work in markets or factories, and keep their earnings. (Jaschok, 1988, p. 103)

Lui (2013) also reports that some of the wives she interviewed still evoked the term shao nai nai (an appellation of “mistress” by muijai and other forms of domestic workers) to illustrate what an ideal husband should provide. We speculate that some egalitarian men and women may associate themselves with a starkly exploitative and outmoded relationship when employing domestic helpers. Hence, it is also plausible that egalitarian men and women in Hong Kong, feeling embarrassed and uncomfortable, might not hire for-eign domestic helpers (Groves & Lui, 2012).



Cheung and Lui 7

Empirical support for the effect of gender ideology on hiring domestic help is scarce and contradictory. For instance, from a sample of married women, Oropesa (1993) finds no relationship between respondents’ gender ideologies and the hiring of housekeepers. However, he finds that more tradi-tional women are less likely to have meals delivered home and explains that cooking is probably central to these women’s identity. They do not want to compromise this identity by buying meals, regardless of their fondness, or lack thereof, for cooking. In another study with individual-level data, Baxter et al. (2009) demonstrate that there is little evidence for the influence of men’s and women’s individual-level gender attitudes in predicting the use of paid domestic help. However, using couple-level data, van der Lippe et al. (2013) show that a man’s gender ideology has an impact on the likelihood of the outsourcing of maintenance and child care.

The Present Study: An Interaction Effect Hypothesis

In this study, we hypothesize that the influence of respondents’ gender atti-tudes and women’s income on hiring help are conditional on each other. Given the conflicting views in the literature, there are two possible explana-tions for the way in which gender ideology and high income among women interact to determine the likelihood of hiring help.

It is plausible that men and women with traditional gender ideologies may be more likely to hire help when they can afford it because housework is shared unequally. Negotiating with and/or shifting the household burden, interpreted as “womanly work,” to the husband is a nonsolution because it violates their attitudes about gender. Instead, a high-income wife could avoid directly con-fronting or negotiating with her traditional husband by employing a strategy of “buying herself out” of housework. An egalitarian husband, in contrast, is more willing to share the housework with a high-income wife, as he is likely to feel less threatened by her income (Lui, 2013). This obviates the need to hire help. Hence, the positive association between a wife’s income and hiring help is magnified if the respondents hold traditional gender attitudes. Moreover, as previously suggested, some egalitarian men and women in Hong Kong may feel embarrassed and uncomfortable about hiring help even if they can afford it because it reminds them of the practices of exploitation in the long history of muijai in Hong Kong and China, which violates their principle of egalitarian-ism. Hence, the positive association between a wife’s income and hiring help is likely to be weaker when the respondents hold egalitarian attitudes.

However, past studies also show that high-income wives with traditional beliefs “do gender” by associating the desire to be good wives with doing housework (Atkinson & Boles, 1984; Tichenor, 1999, 2005). These wives




downplay their economic contributions and emphasize the time and attention they devote to their families. Killewald and Gough (2010) show that the effect of wives’ income is conditional on the type of household tasks, as some tasks are more laden with gender meaning than others. Traditional husbands further maintain a gender boundary, not only by allowing their wives to carry this bur-den but also by applauding their wives’ efforts in the home (Atkinson & Boles, 1984). Based on this logic, respondents with traditional beliefs would be less likely to hire help despite the wife’s high income. In contrast, egalitarian women may not base their gender identity on doing domestic work, and egali-tarian men will not identify housework as solely a wife’s duty and will thus allow them to use paid help for housework. This line of reasoning suggests that men and women with egalitarian gender attitudes and ample resources will be more likely to hire domestic help than men and women who are bounded by gender expectations. In short, the literature implies that the effects of income and gender ideology on hiring help are conditional on one another, although this interaction effect hypothesis has not been examined thoroughly.

The level at which a respondent’s gender ideology interacts with the wife’s income on the decision to hire help may differ between male and female sam-ples. As Davis and Greenstein (2009) suggest, the husband’s gender ideology is a stronger determinant of housework divisions than the wife’s gender ideol-ogy. Using couple-level panel data from Sweden, Evertsson (2014) also finds a stronger effect of men’s rather than women’s gender ideologies on the man’s share of the housework. In parenting and child care, Bulanda (2004) finds that while egalitarian fathers are more involved than traditional fathers, the moth-er’s gender ideology cannot predict paternal involvement. In other words, even if the mother is egalitarian, she will be unable to negotiate greater pater-nal involvement if the father has traditional gender views. The study con-ducted by van der Lippe et al. (2013) also confirms the relative importance of men’s gender ideologies, as compared with women’s, in the outsourcing of some household tasks. We hypothesize that a man’s gender ideology should be more salient than a woman’s in hiring help when a wife has a high income.

Context for the Current Study: Hong Kong

Hong Kong has a long tradition of using domestic servants. Long before the establishment of colonial Hong Kong, wealthy Chinese families commonly relied on free and enslaved men and women to do domestic work. Through slavery, pawning, or indenture, muijai (young female indentured or bonded ser-vants) were commonly hired to enter households as domestic workers. Although slavery in Hong Kong was officially abolished by Queen Victoria in 1844, the practice of muijai remained largely unaffected until the 1940s when the muijai



Cheung and Lui 9

were replaced by live-in amahs, sworn spinsters and/or female refugees who arrived from Guangdong seeking domestic work. They enjoyed relative free-dom, even though the master–servant relationship was maintained. By the 1970s, when Hong Kong started to industrialize, more women opted to work outside the home rather than be confined to domestic work, as the former pro-vided a more independent lifestyle and higher social status. In 1973, the govern-ment implemented a policy allowing foreign nationals to work as “domestic helpers,” which further increased the number of families hiring domestic help.

In 2001, 10.1% of the households in Hong Kong hired domestic help (Hong Kong Census and Statistics Department, 2001). The most recent Population Census shows that 299,961 foreign domestic helpers are employed within the 2.37 million domestic households (Hong Kong Census and Statistics Department, 2013). Unlike other societies in which households hire part-time domestic help for specific tasks, households in Hong Kong typically hire one domestic helper for a range of services traditionally considered “women’s tasks.” These tasks include cleaning, tidying up the household, purchasing groceries, cooking, taking care of children, and washing and ironing clothes (Hong Kong Census and Statistics Department, 2001). Of the households engaging help, 87.9% hire foreign full-time, live-in domestic helpers, predom-inantly from the Philippines and Indonesia. Although this figure is very high compared with Western societies, studies analyzing the factors associated with hiring domestic helpers in Hong Kong and other Asian societies are lacking.

There are two primary reasons why hiring domestic help is becoming more common in Hong Kong. First, the globalized economy has created more job opportunities for women and made importing foreign domestic helpers less costly. Although, as in many other East Asian societies, the female labor participation rate in Hong Kong is still lower than the male labor force participation rate, the gender gap has narrowed significantly in recent years. In 1986, only 55.4% of women aged between 30 and 34 years were in the labor force (compared with 98.8% of men in this age range). The figure rose to 79.7% in 2009 (compared with 96.5% of men), reflecting the rise in dual-earner households. As indicated in the latest Census, only 35.8% of households were sole earner (Hong Kong Census and Statistics Department, 2013). The income gap between men and women is also narrowing. In 2009, the median monthly wage earned by local working women in Hong Kong was approximately US$1,300 (Hong Kong Census and Statistics Department, 2010). In contrast, domestic helpers’ median wage was well below US$500. Based on these figures, a household could earn substantially more if domestic work was outsourced and the wife worked outside the home.

Second, domestic labor remains “women’s work” despite women working long hours outside the home. On average, employed women work outside the




home 45 hours per week. Many of those who work as professionals, manag-ers, and service employees work overtime, clocking more than 50 hours per week (Hong Kong Census and Statistics Department, 2003). Despite this, working women are still expected to be responsible for housework. Many people in Hong Kong continue to believe in the woman-homemaker/man-provider role divisions (Chow & Lum, 2008). In 2002, more than half of the married women in Hong Kong undertook the majority of the domestic work in their households (Hong Kong Census and Statistics Department, 2003).

Method

Data

This study analyzes household survey data from the Hong Kong Family Life Survey 2009. The survey was conducted by the Centre for Chinese Family Studies of the Hong Kong Institute of Asia-Pacific Studies at the Chinese University of Hong Kong. A list of residential addresses was randomly sam-pled from the Frame of Quarters. This frame is maintained by the Hong Kong Census and Statistics Department and contains the most complete and up-to-date register of residential addresses in Hong Kong. Eligible respon-dents were Chinese residents in Hong Kong who were, or had been, married (or ever cohabited with a partner) and were 60 years old or younger. If there was more than one eligible respondent living in the sampled household, only one respondent was randomly selected for a face-to-face interview. We con-ducted interviews to collect information on the respondents and the respon-dents’ spouses. The units of analysis in this study are at the individual level. To avoid oversampling homemakers, interviews were mostly conducted on the weekends or at night on weekdays. In total, 1,177 respondents completed the interview. The contact response rate was 49.2%, which is comparable with other household surveys conducted in Hong Kong. We restricted our focus to the patterns of hiring domestic help for respondents who were mar-ried at the time of interview. Excluding those respondents who were currently cohabiting (18 respondents) or separated or divorced from their partner (185 respondents), our analytic sample totaled 974 respondents.

Measures

Hiring Domestic Help. The dependent variable of this study is dichotomous, indicating whether the respondent had hired domestic help to share the house-work (0 = no, 1 = yes). Respondents were asked who was primarily respon-sible for the following seven household tasks: (a) cooking, (b) washing



Cheung and Lui 11

dishes, (c) buying groceries, (d) washing and ironing clothes, (e) maintaining household appliances, (f) managing household expenses, and (g) cleaning and tidying up. If a respondent indicated that a domestic helper was primarily responsible for at least one of the tasks, we identified that household as hav-ing hired domestic help. If a respondent did not mention that a domestic helper was responsible for any of the seven tasks, we assumed that no domes-tic help had been hired. This assumption is reasonable because the tasks are the most common for domestic helpers in Hong Kong (Hong Kong Census and Statistics Department, 2001).4 We constructed an additional variable, “number of household tasks for which domestic helper is responsible,” for additional analysis. For those who had hired domestic help, the average num-ber of household tasks for which a domestic helper was responsible was 4.3.

Gender Attitude. The respondents’ attitudes toward gender were measured using a three-item scale. The respondents were asked if they agreed with the following three statements: (a) “Men should pay the expenses of dating”; (b) “Husbands working outside the home with wives responsible for the house-work is the best household arrangement”; and (c) “Breadwinning is mainly a man’s responsibility.” The responses to each item ranged from 1 (strongly disagree) to 7 (strongly agree). The overall scale ranged from 3 to 21, with higher scores corresponding to more traditional gender attitudes (Cronbach’s alpha = .68). The alpha reliability for the gender attitude scale is comparable to similar scales used in past studies in other societies, which varied between .6 and .7 (e.g., Cubbins & Vannoy, 2005). The results of factor analysis showed that all three items load onto one factor.

Men’s gender attitudes are indicated by the responses of the male respon-dents, whereas women’s gender attitudes are indicated by the responses of the female respondents. As only one respondent was interviewed in each household, and only respondents were asked to answer the three “gender atti-tude” questions, we only have the husband’s gender attitude in the male sub-sample and the wife’s gender attitude in the female subsample.

Income. The respondents were asked to report their own and their partner’s monthly income. The variables “Husband’s income” and “Wife’s income” were then constructed with reference to the respondent’s gender. These vari-ables were in 13 ordered categories, ranging from 0 (no income) to 13 (HK$60,000 or above, equivalent to US$7690 or above).

Other Covariates. Other covariates related to our focal variables were included in our final analysis as control variables. These covariates included the hus-band’s age (measured in years) and education (measured in years of formal




schooling), wife’s age and education, number of children, and house ownership (which is related to the average flat size in Hong Kong; see Jim & Chen, 2009).

Analytic Strategy

Logistic Regression Models. The primary objective of this study was to exam-ine the association between individual-level gender attitudes, women’s income, and hiring domestic help. The hiring of domestic help is a dichoto-mous variable. Logistic regression models were used to predict the log-odds of hiring a domestic helper. Two logistic regression models were specified. First, the main effect model was fitted to the data to examine the association between couple’s income, respondent’s gender attitude, and hiring domestic help, controlling for other covariates but without considering the interaction effect. In the second model, an interaction term, the product of the wife’s income and the respondent’s gender attitude, was added. Similar to studies that have collected data from only one member in a couple (Baxter, 2002; Baxter et al., 2009; Oropesa, 1993; Rogers & Amato, 2000), we only have information about the respondents’ gender attitudes but not their spouses’. Following the design used in a number of other studies (Baxter, 2002; Baxter et al., 2009), we estimated the two logistic regression models for the male and female subsamples separately to specify the effect of men’s and women’s individual-level gender attitudes on hiring domestic help.

In addition to the logistic regression models, we tested our interaction effect hypothesis with a Poisson model predicting the number of house-work items for which a domestic helper is responsible. The results of the logistic regression models (predicting whether a domestic helper is hired) and Poisson models (predicting the number of items for which help is received) are almost the same. As we are most concerned with the decision to hire domestic help, and given space constraints, we only show and dis-cuss the results of logistic regression here, but the results of the Poisson model are available on request.

Missing Data. There are no missing values for the dependent variable in this study. However, there are missing values for some independent variables. The multiple imputation method is recommended for obtaining unbiased estimates (Acock, 2005). Thus, we used a multiple imputation procedure with chained equation to impute the missing values, creating 20 complete data sets with imputed values for the missing data. Logistic regression mod-els were fitted to the imputed data. To check whether our conclusion was robust to the choice of imputation strategy, we also analyzed the data with single imputation and the listwise deletion method. The substantive



Cheung and Lui 13

conclusion remains the same for all three strategies of handling missing data. We only include the results from multiple imputation analysis as it is the most commonly recommended strategy in the literature.

Results

Descriptive Statistics

Table 1 presents the descriptive statistics for all variables used in our analysis and the number of cases with missing values for each variable. Overall, 12.1% of the married couples in Hong Kong had hired a domestic helper.

Approximately 51.9% of the married couples in the sample lived in apart-ments they owned. On average, the married couples had 1.7 children. The mean age of the husbands was 49.5 years and that of the wives was 45.0 years. On average, the husbands had 10.2 years of formal schooling and the wives had 9.8 years. The average income for the husbands was between HK$8,000 and HK$9,999, whereas the average income for the wives was less than HK$6,000. This income disparity probably arose because a sub-stantial proportion of wives in Hong Kong work part time only or do not work outside the home. The average income of the working wives in the sample was between HK$8,000 and HK$9,999. To avoid multicollinearity, we did not include the employment status of husbands and wives in our analysis. The gender attitude scale ranged between 3 and 21 (higher scores meaning more traditional gender attitudes). The average scores for gender attitude in the male and female subsamples were 13.8 and 12.6, respectively, slightly higher than the midpoint of the scale.

There are no missing values for the dependent variable, although there are a number of cases with missing values for house ownership (four cases), number of children (three cases), male respondents’ gender attitudes (seven cases), and female respondents’ gender attitudes (nine cases). There are more cases with missing values for the age, education, and income variables, mostly because some respondents did not report the age, years of schooling, and income of their spouses. Of the respondents sampled, 17.6% and 9.2% did not report the husband’s or the wife’s income, respectively. To take the effect of missing data into account, we estimated the logistic regression with the imputed data (m = 20).

Results of Logistic Regression Models

Table 2 presents the results of the logistic regression models predicting the log-odds of hiring domestic help in Hong Kong. The main effect model




considered the association between hiring domestic help and husbands’, wives’, and household characteristics. The interaction effect model addition-ally considered the interaction effect of husbands’ or wives’ gender attitudes and wives’ income on hiring domestic help. As we did not have information on the spouse’s gender attitude, we separated our models for the male and female subsamples to estimate the effect of husbands’ and wives’ gender atti-tudes on hiring domestic help.

In the main effect model, the number of children was positively associated with hiring domestic help for both the male (logit coefficient = 0.590, p < .05) and female (logit coefficient = 0.542, p < .01) subsamples. The husband’s income was also positively and significantly associated with hiring domestic help for both the male (logit coefficient = 0.174, p < .05) and female (logit coefficient = 0.220, p < .001) subsamples. The wife’s education was posi-tively associated with hiring domestic help, but was only significant in the male subsample (logit coefficient = 0.129, p < .05). The wife’s income was positively and significantly associated with hiring domestic help for both the male (logit coefficient = 0.281, p < .001) and female (logit coefficient = 0.185, p < .001) subsamples. In general, couples with higher socioeconomic status and more children were more likely to hire a domestic helper.

The respondents’ gender attitudes were not significantly associated with hiring domestic help in the male (logit coefficient = 0.065, p = .248) and female subsamples (logit coefficient = 0.043, p = .299). These findings are similar to those of past studies. However, the interaction term of the

Table 1. Means and Standard Deviations of All Variables Used in This Study (Raw and Imputed Data, N = 974).

Raw data Imputed data (m = 20)

n M SD n M SD

Hiring domestic help 974 0.121 0.326 974 0.121 0.326House ownership 970 0.519 0.500 974 0.518 0.500Number of children 971 1.732 0.991 974 1.733 0.991Husband’s age 915 49.527 9.494 974 49.640 9.546Husband’s education 906 10.161 4.784 974 10.128 4.799Husband’s income 803 4.254 3.424 974 4.305 3.371Husband’s gender attitude 384 13.805 3.849 391 13.785 3.846Wife’s age 907 44.966 8.508 974 44.902 8.507Wife’s education 920 9.775 4.439 974 9.767 4.429Wife’s income 884 1.899 2.920 974 1.974 2.895Wife’s gender attitude 574 12.598 4.023 583 12.612 4.024



15

Tab

le 2

. Lo

gist

ic R

egre

ssio

n M

odel

s: H

ouse

hold

Cha

ract

eris

tics,

Cou

ple

Cha

ract

eris

tics,

and

Hir

ing

Dom

estic

Hel

p.

Mal

e sa

mpl

e (n

= 3

91)

Fem

ale

sam

ple

(n =

583

)

M

ain

effe

ct m

odel

Inte

ract

ion

effe

ct m

odel

Mai

n ef

fect

mod

elIn

tera

ctio

n ef

fect

mod

el

Lo

git

coef

ficie

ntSE

Logi

t co

effic

ient

SELo

git

coef

ficie

ntSE

Logi

t co

effic

ient

SE

Hou

seho

ld c

hara

cter

istic

s

Hou

se o

wne

rshi

p0.

552

0.45

90.

498

0.46

40.

531

0.37

50.

533

0.37

5

Num

ber

of c

hild

ren

0.59

0*0.

245

0.61

3*0.

252

0.54

2**

0.19

50.

544*

*0.

196

Hus

band

’s c

hara

cter

istic

s

Hus

band

’s a

ge−

0.01

00.

056

0.00

10.

059

−0.

008

0.03

8−

0.00

90.

038

H

usba

nd’s

edu

catio

n0.

049

0.05

00.

052

0.05

00.

114

0.05

90.

114

0.05

9

Hus

band

’s in

com

e0.

174*

0.07

00.

195*

*0.

073

0.22

0***

0.05

90.

220*

**0.

059

H

usba

nd’s

gen

der

attit

ude

0.06

50.

056

−0.

061

0.07

7—

——

—W

ife’s

cha

ract

eris

tics

W

ife’s

age

−0.

046

0.05

8−

0.05

60.

061

−0.

008

0.03

9−

0.00

80.

040

W

ife’s

edu

catio

n0.

129*

0.06

50.

128*

0.06

50.

041

0.06

00.

041

0.06

0

Wife

’s in

com

e0.

281*

**0.

064

−0.

135

0.18

50.

185*

**0.

053

0.13

00.

134

W

ife’s

gen

der

attit

ude

——

——

0.04

30.

041

0.02

60.

056

Inte

ract

ion

term

W

ife’s

inco

me

× h

usba

nd’s

ge

nder

att

itude

——

0.03

3*0.

015

——

——

W

ife’s

inco

me

× w

ife’s

ge

nder

att

itude

——

——

——

0.00

50.

011

Inte

rcep

t−

5.84

9***

1.73

0−

4.40

8*1.

803

−6.

589*

**1.

461

−6.

360*

**1.

547

Not

e. N

= 9

74; i

mpu

ted

data

m =

20.

In t

he m

ale

sam

ple,

hus

band

’s a

nd w

ife’s

cha

ract

eris

tics

refe

r to

the

res

pond

ent’s

and

spo

use’

s ch

arac

teri

stic

s, r

espe

ctiv

ely.

In t

he

fem

ale

sam

ple,

hus

band

’s a

nd w

ife’s

cha

ract

eris

tics

refe

r to

the

spo

use’

s an

d re

spon

dent

’s c

hara

cter

istic

s, r

espe

ctiv

ely.

*p <

.05.

**p

< .0

1.**

*p <

.001

(tw

o-ta

iled)

.




-.20

.2.4

.6.8

Logi

t Coe

ffici

ent o

f wife

's in

com

e

3 5 7 9 11 13 15 17 19 21Husband's gender attitude

95% C.I. Logit Coef.

Figure 1. Logit coefficient of wife’s income under different levels of husband’s gender attitude.

husband’s gender attitude and the wife’s income is positive and significant (logit coefficient = 0.033, p < .05) in the interaction effect model with the male subsample. This indicates that the husband’s gender attitude and the wife’s income affect the likelihood of hiring domestic help interactively, not independently. In contrast, the interaction term of the wife’s gender attitude and her own income is positive but not significant. We do not have statistical evidence for an interaction effect between the wife’s gender attitude and her income on hiring domestic help.

Interpreting the Interaction Effect

To effectively show the impact of the wife’s income and husband’s gender attitude on hiring domestic help, as shown in the interaction effect model, the size of logit coefficients are derived by combining the logit coefficients of constituent terms and the interaction term. Figure 1 presents the logit coeffi-cient of the wife’s income on hiring domestic help under different levels of the husband’s gender attitude. As the figure illustrates, a husband’s gender attitude magnifies the positive association between his wife’s income and



Cheung and Lui 17

hiring domestic help. Although the main effect model indicates that the wife’s income was significantly associated with hiring domestic help, the interac-tion effect model shows that this positive association was not significant if her husband had an egalitarian gender attitude. For an egalitarian husband, his wife’s income was not associated with hiring domestic help.

Although gender attitude was not significantly associated with hiring domestic help in the main effect model, the interaction effect model shows that a husband’s gender attitude was positively and significantly associated with hiring domestic help if he had a high-income wife. As Figure 2 illus-trates, only if the husband had a low-income wife, his gender attitude was not significantly related to the decision to hire domestic help. Therefore, ignoring the interaction effect, one may overlook the association between a husband’s gender attitude and hiring domestic help for those with a high-income wife.

To show the difference between the interaction and the main effect models more intuitively, predicted probabilities of hiring domestic help were derived from the above models. Table 3 presents the predicted probabilities of hiring domestic help for couples with different combinations of the husband’s gen-der attitude and wife’s income, which are derived from the interaction model

-.20

.2.4

.6Lo

git C

oeffi

cien

t of h

usba

nd's

gen

der a

ttitu

de

1 2 3 4 5 6 7 8 9 10 11 12 13Wife's income

95% C.I. Logit Coef.

Figure 2. Logit coefficient of husband’s gender attitude under different levels of wife’s income.




Table 3. Comparison of Predicted Probabilities of Hiring Domestic Help for Couples With Different Levels of Husband’s Gender Attitude and Wife’s Income (Interaction Effect Model).

Predicted probabilities of hiring domestic help

Wife’s income

HK$0

(not employed)HK$8,000

to HK$9,999HK$18,000

to HK$21,999HK$30,000

to HK$34,999

Husband’s gender attitude

6 4.61% 5.55% 6.66% 7.98%9 3.88% 6.19% 9.74% 15.00%

12 3.26% 6.91% 14.04% 26.42%15 2.74% 7.70% 19.81% 42.22%18 2.30% 8.58% 27.20% 59.79%

Note. Predicted probabilities are derived from the interaction effect model in Table 2. Other covariates held at their sample means.

with the male subsample in Table 2. As indicated, there was an increasing trend for hiring domestic help with increases in the wife’s income. A hus-band’s traditional gender attitude also magnified the positive effect of the wife’s income on hiring domestic help. For a husband with a high-income wife, his gender traditionalism was positively associated with hiring domes-tic help. For an average-income wife, the husband’s gender attitude did not affect the probability of hiring domestic help.

For couples with a traditional husband and a high-income wife, the pre-dicted probability of hiring domestic help could be as high as 59.79%. Couples with an egalitarian husband and a high-income wife had probabili-ties of hiring domestic help as low as 7.98%, which is slightly lower than the percentage of hiring domestic help for the overall sample.

Table 4 presents the predicted probabilities of hiring domestic help for couples with different combinations of the husband’s gender attitude and the wife’s income, which are derived from the main effect model in Table 2. For couples with a low-income wife, the two models do not offer substantially different predictions. Both models indicate that these couples have low pre-dicted probabilities of hiring domestic help. It is clear, however, that the pre-dictions from these two models are substantially different for couples with a high-income wife. The main effect model underestimates the probabilities of hiring domestic help for couples with a traditional husband and a high-income wife but overestimates the probabilities of hiring domestic help for couples with an egalitarian husband and a high-income wife.



Cheung and Lui 19

Discussion and Conclusion

This article investigates the interaction effect of individual-level gender atti-tude and wife’s income on the decision to hire domestic help in Hong Kong. Some previous studies fall short of finding a significant association between gender attitude and hiring domestic help, while wife’s income is a consistent factor in the decision to hire domestic help in a number of studies (Baxter et al., 2009; Bittman et al., 1999; Oropesa, 1993; Spitze, 1999; van der Lippe et al., 2004). We found that a husband’s traditional gender attitude and his wife’s high income sharply increase the likelihood of hiring domestic help, controlling for other covariates. However, the associations with hiring domestic help are conditional on each other. Not surprisingly, a husband’s gender traditionalism is not associated with hiring domestic help if he has a low-income wife, because it is not financially feasible. For a man with a high-income wife (that is, for whom hiring domestic help is most financially pos-sible), the husband’s gender attitude is significantly associated with the decision to hire domestic help. In contrast, the positive association between a wife’s income and hiring domestic help is magnified if her husband has a traditional gender attitude. In couples with an egalitarian husband, the wife’s income is not associated with hiring domestic help. As argued elsewhere (Lui, 2013), hiring domestic help can be a strategy to avoid shifting house-work responsibilities to the husband. Under circumstances in which a hus-band is unwilling to take on more domestic responsibility, his wife’s income is important for her to “buy out” the burden of housework.

Table 4. Comparison of Predicted Probabilities of Hiring Domestic Help for Couples With Different Levels of Husband’s Gender Attitude and Wife’s Income (Main Effect Model).

Predicted probabilities of hiring domestic help

Wife’s income

HK$0

(not employed)HK$8,000

to HK$9,999HK$18,000

to HK$21,999HK$30,000

to HK$34,999

Husband’s gender attitude

6 1.91% 4.30% 9.38% 19.25%9 2.31% 5.16% 11.14% 22.42%

12 2.78% 6.19% 13.19% 25.94%15 3.35% 7.40% 15.56% 29.81%18 4.03% 8.83% 18.26% 33.98%

Note. Predicted probabilities are derived from the main effect model in Table 2. Other covariates held at their sample means.




In the family devotion framework, couples with a traditional husband and a high-income wife tend to contract out domestic duties to ensure that they meet the intensified needs of motherhood. Hiring help does not imply that these wives are relinquishing their motherhood role; rather it reflects that the family devotion schema has expanded the definition of good mothering. This schema acknowledges that mothers have to delegate some domestic duties, particularly those that do not require higher thinking skills or emotional involvement, to domestic helpers so that they can better perform their roles as mothers (Anderson, 2000). With tension increasing between family and work roles, working women, who are caught in work–family conflict and unable to shift the burden to their husbands, may choose to outsource household tasks to a domestic helper if they can afford to do so. Thus, the increasing popularity of hiring domestic help maintains the hierarchical and gendered division of household labor in Hong Kong. In general, the use of domestic help is not as common in societies where egalitarian household arrangements are easier to achieve. As shown in this study, couples with more egalitarian gender attitudes are less likely to hire domestic help. Groves and Lui (2012) note that men with traditional gender attitudes often pay for domestic help as a “gift” to their wives so that they can legitimately avoid sharing the housework. An egalitar-ian husband would be more willing to negotiate shared housework (Davis & Greenstein, 2009), and thus domestic help might not be necessary. Alternatively, as Ehrenreich and Hochschild (2003) point out, individuals with egalitarian attitudes are more reluctant to hire domestic help to avoid exploiting other women. This may also partly explain our findings, because the image of hiring domestic help is for some people still closely linked with the image of the exploitation of the muijai in Hong Kong (Constable, 2007; Lui, 2013).

The increasing trend of hiring domestic help is often interpreted as a response to the rising rate of female labor force participation and family–work conflict for women in postindustrial societies. This study shows that the esca-lating trend of hiring domestic help in Hong Kong, especially among middle-class families, is not only driven by these conflicts or by the increasing affordability of domestic help for well-resourced families. It is also driven by a social context in which a traditional gender ideology prevails and husbands tend not to share housework even when wives are working outside the home. Men’s gender expectations play an important role in outsourcing domestic tasks. Fung and Ma (2000) find in a household survey that Hong Kong people generally hold attitudes favoring traditional gender role arrangements. They argue that in Hong Kong, people are still “very stereotypic regarding gender relationships” (p. 67). Indeed, the image of the breadwinner is central to men’s gender performance in Hong Kong. Choi et al. (2012) find that the gendered division of household labor is an important element of the construction of



Cheung and Lui 21

men’s self-identity in Hong Kong. With these traditional attitudes about gen-der, most families in Hong Kong do not have egalitarian household arrange-ments (Cheung & Choi, 2013). By highlighting the role of gender attitudes in domestic outsourcing, we argue that the increasing trend of hiring domestic help in Hong Kong, and probably other Asian societies, should not be inter-preted merely as a response to the increasing female labor force participation rate and family–work conflict but also as a reflection of enduring traditional-ism in how gender is understood and performed. Past studies have shown a trend towards liberalizing gender attitudes in Chinese societies (Chia, Allred, & Jerzak, 1997). Given the findings of this study, this liberal shift of gender attitudes may have important implications for the future trend of hiring domes-tic help in Hong Kong. If gender attitudes in Hong Kong continue to liberal-ize, fewer married couples may hire domestic help despite an increase in the female labor force participation rate over time. Our findings suggest that in addition to women’s employment and income status, researchers and policy makers must take men’s gender attitudes into account when projecting local demand and interpreting the socioeconomic patterns of hiring domestic help.

Several limitations of this study should be noted. Although we do not have statistical evidence that women’s attitudes toward gender affect the decision to hire domestic help in a manner different to men’s attitudes, our limited sample size may have precluded detection of the effect. In future studies, a larger sample will be needed to further investigate whether husbands’ and wives’ gender attitudes associate differently with hiring domestic help. In addition, we only collected information from one member of each couple in the survey and could thus only include either the husband’s or the wife’s gender attitude in our statistical analysis. Although individual-level data are common in the literature on the division of housework and domestic out-sourcing (Baxter, 2002; Baxter et al., 2009; Oropesa, 1993), couple-level data are desirable if the effects of the husband’s and the wife’s gender attitudes on hiring domestic help are to be estimated simultaneously (e.g., van der Lippe et al., 2013). In addition, the nature of cross-sectional data hindered us from drawing a solid causal inference. Our data cannot rule out the possibility of the reverse causation that by hiring domestic help, couples may be able to maintain and reinforce their traditional gender attitudes. Longitudinal data are needed to draw a causal inference.

Our findings on the relationship between gender attitudes, women’s income, and hiring domestic help may be culturally specific. The relationship may further depend on other attitudinal or cultural variables, such as how much men and women think outsourcing domestic help is culturally desirable (Baxter et al., 2009), and on structural factors, such as how easy, affordable, and common it is to hire domestic help in a given context. Nevertheless, our




empirical knowledge of how social contexts affect decisions to hire domestic help is very limited. Comparative studies, such as between Asian societies and Western societies, could empirically examine how the factors involved in hiring domestic help may change in different social contexts. With these limi-tations in mind, our findings still call for scholarly attention to be paid to the fact that the relationship between women’s income, men’s gender attitudes, and the hiring of domestic help is interactive and not as simple as discussed in the literature.

Finally, while we attempt to address the relationship between the women’s income and hiring domestic help, many topics of domestic outsourcing in Asia are still not adequately addressed. For example, long work hours are common in East Asia. This leaves a limited amount of quality time for cou-ples to spend at home together. Official figures show that it is common for women in Hong Kong to work overtime and spend over 50 hours at work per week (Hong Kong Census and Statistics Department, 2003). As policies to promote work–family balance are not well implemented in Hong Kong, along with some other Asian countries, domestic outsourcing is a popular strategy for women to regain time previously spent doing housework. Although the relationship between time availability and outsourcing domes-tic tasks is debatable (Sullivan & Gershuny, 2013), there is a lack of empirical studies examining the relationship between hiring domestic help and time availability in the region. Further studies on the complex relationship between hiring domestic help and the amount of work, household, and leisure time are needed.

Acknowledgments

The authors would like to thank Professor Kwok-fai Ting for sharing the survey data generously. The authors are grateful to the editor and anonymous reviewers for their comments and suggestions on earlier drafts of this work. Many thanks are also given to Professor Frank Furstenberg from University of Pennsylvania, Professor Julie Brines from University of Washington, and colleagues from the Asia Research Institute, National University of Singapore for their encouragement and useful comments.

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publi-cation of this article.



Cheung and Lui 23

Notes

1. As a solution, some families buy “labor saving devices.” Van der Lippe et al. (2004) suggest that buying dishwashers and hiring help save women time. However, some studies argue that time spent on housework has, in fact, increased because technological household appliances simply increase the standard of housework and the way in which women perform household tasks (Bittman, Rice, & Wajcman, 2004). Boydston (1990) further argues that instead of provid-ing a solution to women’s housework burden, these technologies merely idealize the home as a work-free space for men. These household technologies may thus not serve as substitutes for the need for hiring domestic help. This article does not intend to debate whether households use this strategy to save time but rather the focus is on how gender ideology and income affect the hiring of domestic help.

2. In addition to labor income, other resource indicators such as home owner-ship and financial assets are correlates of domestic outsourcing (Cohen, 1998). In Hong Kong, as most flats are small and under 70 m2 in area (Jim & Chen, 2009), apartment size may play a role in the decision of hiring local (part-time) or foreign (live-in) domestic helpers. Hiring a foreign domestic helper is not a feasible strategy if the couple lives in a shoebox apartment. Residents in larger apartments are thus more likely to hire a full-time domestic helper than those in small apartments. For the literature on financial assets and domestic outsourc-ing, readers are referred to Cohen (1998), van der Lippe et al. (2004), and Zick, McCullough, and Smith (1996).

3. Quoting Constable (2007), “the report of the Commission on Mui Tsai in Hong Kong and Malaya defines a muijai as a young girl, around eight or ten years old, who was transferred from her natal family . . . to another family with the inten-tion that she . . . be used as a domestic servant, not in receipt of regular wages and [not] at liberty to leave the employer’s family of her own free will or at the will of her parents” (p. 45). Muijai were often treated as the property of their masters.

4. In the original scale we also included the outsourcing of child care in the survey. However, as the use of child care and paid domestic help for housework are conceptually different and affected by different factors (Bittman et al., 1999), we chose to focus on the use of paid domestic help in this study.

References

Acock, A. C. (2005). Working with missing values. Journal of Marriage and Family, 67, 1012-1028.

Anderson, B. J. (2000). Doing the dirty work? The global politics of domestic labour. London, England: Zed Books.

Atkinson, M. P., & Boles, J. (1984). WASP (wives as senior partners). Journal of Marriage and the Family, 46, 861-870.

Baxter, J. (2002). Patterns of change and stability in the gender division of household labour in Australia, 1986-1997. Journal of Sociology, 38, 399-424.




Baxter, J., Hewitt, B., & Western, M. (2009). Who uses paid domestic labor in Australia? Choice and constraint in hiring household help. Feminist Economics, 15(1), 1-26.

Berk, S. F. (1985). The gender factory: The apportionment of work in American households. New York, NY: Plenum Press.

Bianchi, S. M., Milkie, M. A., Sayer, L. C., & Robinson, J. P. (2000). Is anyone doing the housework? Trends in the gender division of household labor. Social Forces, 79, 191-228.

Bittman, M., Matheson, G., & Meagher, G. (1999). The changing boundary between home and market: Australian trends in outsourcing domestic labour. Work, Employment & Society, 13, 249-273.

Bittman, M., Rice, J. M., & Wajcman, J. (2004). Appliances and their impact: The ownership of domestic technology and time spent on household work. British Journal of Sociology, 55, 401-423.

Blair-Loy, M. (2005). Competing devotions: Career and family among women execu-tives. Cambridge, MA: Harvard University Press.

Blair, S. L., & Lichter, D. T. (1991). Measuring the division of household labor gender segregation of housework among American couples. Journal of Family Issues, 12, 91-113.

Blumstein, P., & Schwartz, P. (1983). American couples: Money, work, sex. New York, NY: Morrow.

Boydston, J. (1990). Home and work: Housework, wages, and the ideology of labor in the early republic. New York, NY: Oxford University Press.

Brayfield, A., & Hofferth, S. L. (1995). Balancing the family budget: Differences in child care expenditures by race/ethnicity, economic status, and family structure. Social Science Quarterly, 76, 158-177.

Bulanda, R. E. (2004). Paternal involvement with children: The influence of gender ideologies. Journal of Marriage and Family, 66, 40-45.

Cheung, A. K. L., & Choi, S. Y. P. (2013). Economic insecurity and husband-to-wife physical assault in Hong Kong: The role of husband’s power motive. In W. J. J. Yeung & M. T. Yap (Eds.), Economic stress, human capital, and families in Asia (pp. 105-127). Dordrecht, Netherlands: Springer.

Chia, R. C., Allred, L. J., & Jerzak, P. A. (1997). Attitudes toward women in Taiwan and China. Psychology of Women Quarterly, 21, 137-150.

Choi, S. Y. P., Au, W. T., Wong, W., Liong, C. C. M., Wong, F. Y., Lo, S. K. W., & Chao, K. C. (2012). Research report: Exploratory study on gender stereotyping and its impacts on male gender. Hong Kong: Equal Opportunities Commission.

Chow, N., & Lum, T. (2008). Trends in family attitudes and values in Hong Kong. Hong Kong: Central Policy Unit, Hong Kong SAR Government.

Cohen, P. N. (1998). Replacing housework in the service economy: Gender, class, and race-ethnicity in service spending. Gender & Society, 12, 219-231.

Coltrane, S. (1998). Gender and families. Lanham, MD: Rowman & Littlefield.Coltrane, S., & Ishii-Kuntz, M. (1992). Men’s housework: A life course perspective.

Journal of Marriage and the Family, 54, 43-57.



Cheung and Lui 25

Constable, N. (2007). Maid to order in Hong Kong: Stories of migrant workers. Ithaca, NY: Cornell University Press.

Crompton, R., & Lyonette, C. (2006). Work-life “balance”in Europe. Acta Sociologica, 49, 379-393.

Cubbins, L. A., & Vannoy, D. (2005). Socioeconomic resources, gender traditional-ism, and wife abuse in urban Russian couples. Journal of Marriage and Family, 67, 37-52.

Davis, S. N., & Greenstein, T. N. (2009). Gender ideology: Components, predictors, and consequences. Annual Review of Sociology, 35, 87-105.

DeVault, M. L. (1991). Feeding the family: The social organization of caring as gen-dered work. Chicago, IL: University of Chicago Press.

Doucet, A. (2001). “You see the need perhaps more clearly than I have”: Exploring gendered processes of domestic responsibility. Journal of Family Issues, 22, 328-357.

Ehrenreich, B., & Hochschild, A. R. (2003). Global woman: Nannies, maids, and sex workers in the new economy. New York, NY: Metropolitan Books.

Evertsson, M. (2014). Gender ideology and the sharing of housework and child care in Sweden. Journal of Family Issues, 35, 927-949. doi:10.1177/0192513X14522239

Fung, A., & Ma, E. (2000). Formal vs. informal use of television and sex-role stereo-typing in Hong Kong. Sex Roles, 42, 57-81.

Gershuny, J. (2000). Changing times: Work and leisure in postindustrial society. Oxford, England: Oxford University Press.

Gershuny, J., Bittman, M., & Brice, J. (2005). Exit, voice, and suffering: Do couples adapt to changing employment patterns? Journal of Marriage and Family, 67, 656-665.

Gershuny, J., & Robinson, J. P. (1988). Historical changes in the household division of labor. Demography, 25, 537-552.

Groves, J. M., & Lui, L. (2012). The “gift”of help: Domestic helpers and the main-tenance of hierarchy in the household division of labour. Sociology, 46, 57-73.

Gupta, S. (2006). Her money, her time: Women’s earnings and their housework hours. Social Science Research, 35, 975-999.

Gupta, S. (2007). Autonomy, dependence, or display? The relationship between mar-ried women’s earnings and housework. Journal of Marriage and Family, 69, 399-417.

Hanson, S. L., & Ooms, T. (1991). The economic costs and rewards of two-earner, two-parent families. Journal of Marriage and the Family, 53, 622-634.

Hochschild, A. R. (1989). The second shift. New York, NY: Viking.Hochschild, A. R. (1997). The time bind: When work becomes home and home

becomes work. New York, NY: Metropolitan Books.Hong Kong Census and Statistics Department. (2001). Thematic household survey

report (Report No. 5). Hong Kong: Author.Hong Kong Census and Statistics Department. (2003). Thematic household survey

report (Report No. 14). Hong Kong: Author.




Hong Kong Census and Statistics Department. (2010). Women and men in Hong Kong: Key statistics 2010. Hong Kong: Author.

Hong Kong Census and Statistics Department. (2013). 2011 Population census—Main report: Volume 1. Hong Kong: Author.

Hook, J. L. (2006). Care in context: Men’s unpaid work in 20 countries, 1965-2003. American Sociological Review, 71, 639-660.

Jaschok, M. (1988). Concubines and bondservants: A social history of Chinese cus-tom. Hong Kong: Oxford University Press.

Jim, C., & Chen, W. Y. (2009). Value of scenic views: Hedonic assessment of private housing in Hong Kong. Landscape and Urban Planning, 91, 226-234.

Killewald, A., & Gough, M. (2010). Money isn’t everything: Wives’ earnings and housework time. Social Science Research, 39, 987-1003.

Kluwer, E. S., Heesink, J. A. M., & Van de Vliert, E. (1996). Marital conflict about the division of household labor and paid work. Journal of Marriage and the Family, 58, 958-969.

Lee, W. K. M. (2002). Gender ideology and the domestic division of labor in middle-class Chinese families in Hong Kong. Gender, Place and Culture, 9, 245-260.

Lui, L. (2013). Re-negotiating gender: Household division of labor when she earns more than he does. London, England: Springer.

Oropesa, R. S. (1993). Using the service economy to relieve the double burden female labor force participation and service purchases. Journal of Family Issues, 14, 438-473.

Presser, H. B. (1994). Employment schedules among dual-earner spouses and the division of household labor by gender. American Sociological Review, 59, 348-364.

Rogers, S. J., & Amato, P. R. (2000). Have changes in gender relations affected mari-tal quality? Social Forces, 79, 731-753.

Soberon-Ferrer, H., & Dardis, R. (1991). Determinants of household expenditures for services. Journal of Consumer Research, 17, 385-397.

Spitze, G. (1999). Getting help with housework household resources and social net-works. Journal of Family Issues, 20, 724-745.

Sullivan, O., & Gershuny, J. (2013). Domestic outsourcing and multitasking: How much do they really contribute? Social Science Research, 42, 1311-1324. doi:10.1016/j.ssresearch.2013.05.004

Task Force on Population Policy. (2003). Report of the Task Force on Population Policy, 2003. Hong Kong: Hong Kong SAR Goverment.

Tichenor, V. (1999). Status and income as gendered resources: The case of marital power. Journal of Marriage and the Family, 61, 638-650.

Tichenor, V. (2005). Maintaining men’s dominance: Negotiating identity and power when she earns more. Sex Roles, 53, 191-205.

Treas, J., & de Ruijter, E. (2008). Earnings and expenditures on household services in married and cohabiting unions. Journal of Marriage and Family, 70, 796-805.

van der Lippe, T., Frey, V., & Tsvetkova, M. (2013). Outsourcing of domestic tasks a matter of preferences? Journal of Family Issues, 34, 1574-1597.



Cheung and Lui 27

van der Lippe, T., Tijdens, K., & de Ruijter, E. (2004). Outsourcing of domestic tasks and time-saving effects. Journal of Family Issues, 25, 216-240.

Warren, T. (2003). Class and gender-based working time? Time poverty and the divi-sion of domestic labour. Sociology, 37, 733-752.

Windebank, J. (2001). Dual-earner couples in Britain and France: Gender divi-sions of domestic labour and parenting work in different welfare states. Work, Employment & Society, 15, 269-290.

Zick, C. D., McCullough, J., & Smith, K. R. (1996). Trade-offs between purchased services and time in single-parent and two-parent families. Journal of Consumer Affairs, 30, 1-23.



hiring domestic help © the author(s) 2015 in hong kong ... · domestic helpers increased from...

Documents