投稿査読レター論文
共働き化が家事の外部化に与える影響の時系列変化
― 「家計調査」ミクロデータに基づく日本の調理食品を対象とした検証 ―
2026 年 46 巻 3 号 p. 317-320
詳細
2026 年 46 巻 3 号 p. 317-320
Household outsourcing refers to the use of market-based products or services for family care, such as preparing foods and childcare. Previous studies from Europe and North America established that, once household income and life stage are controlled for, the impact of dual-earner status disappears. However, they lack a perspective on temporal change, and the effects may differ depending on the originally constructed family norms. Using microdata from Japan’s “Family Income and Expenditure Survey,” this study examines the effect of dual-earner status on household outsourcing in Japan. The findings reveal two main points: (1) in contrast to existing research, a dual-earner effect exists in Japan; and (2) the gap in the household outsourcing between dual- and single-earner households has gradually narrowed. Finally, potential explanations and future research directions are discussed.
家事の外部化とは，料理や育児など家族のケアに関わる家事を市場化された製品で賄う消費行動である。先行して共働き化が進展した欧米を対象とした研究では，共働き化の影響は，世帯年収と世帯のライフステージを考慮すれば確認できないことが複数報告されているが，時系列変化に対する視点の欠如が指摘されてきたほか，元来的に構築された家族規範が異なれば影響は異なる可能性がある。本研究は，日本での共働き化が家事の外部化に与える影響を，1990～2020年の総務省「家計調査」のミクロデータで検証する。①先行研究とは異なり，日本では共働き化の影響は確認される一方，②共働き世帯と専業世帯の間の家事の外部化に関する差異が僅少化していくことを示し，考えられる可能性や研究課題を示す。
Household outsourcing refers to “the transfer of intimate tasks historically or normatively seen as being performed within the family and by family members to formal commercial establishments located outside of the family” (Epp & Velagaleti, 2014, p. 911). Recently, as dual-earner households have become more common in Japan, outsourcing of family care such as cooking and childcare has increased, as seen in the growth of the prepared and frozen food market. Household outsourcing has been examined as a consumption practice involved in family identity construction, taking the family as the unit of analysis (e.g., Epp & Velagaleti, 2014; Kastarinen et al., 2023).
Previous studies examining Western countries during a period of dual-earner household growth, similar to present-day Japan, have surprisingly shown that the effects of such households cannot be confirmed when controlling for household income and life stage (e.g., Nickols & Fox, 1983; Strober ＆ Weinberg, 1980; Weinberg ＆ Winer, 1983). However, these studies have been criticized for a lack of perspective on temporal changes (e.g., Rubin et al., 1990). Furthermore, the effects may differ depending on the originally constructed family norms. Therefore, using microdata from the “Family Income and Expenditure Survey” by the Ministry of Internal Affairs and Communications, this study examines the temporal changes in the effect of dual-earner households on household outsourcing in Japan.
This study uses microdata from the “Family Income and Expenditure Survey” by the Statistics Bureau of Japan, which covers nationwide households with two or more members, at five-year intervals from 1990 to 2020 (see Table 1(i)). This period is selected because it marks the time during which the proportion of dual- to single-earner households reversed in Japan, and the nature of housework underwent significant transformation (Hakuhodo Institute of Life and Living, 2019). The housework domain examined is meal preparation, which is measured using the expenditure share of prepared foods in total food expenditure. This is because in previous studies, meal preparation has been frequently referred to as an essential housework task strongly associated with family care, and prepared foods1) examined as a domain in which the dynamics of family care are manifested more sharply than in dining out (e.g., Epp & Price, 2008; Moisio et al., 2004; Nickols & Fox, 1983).
;%0A%09%09%09newWindow.document.open();%0A%09%09%09newWindow.document.write('<img src=%22./Graphics/46-3qjm02_t01.jpg%22>');%0A%09%09%09newWindow.document.close();%0A%09%09)
**…p<1%, Log-Likelihood=−176918.88, AIC=353885.8, BIC=354157.8, “—” indicates same as above
The variables are listed in Table 1(ii). Using Stata 18.0, a fractional logit regression analysis (Villadsen & Wulff, 2021) was conducted with the expenditure share of prepared foods (deflated by CPI) as the dependent variable. The independent variables include year dummies (each year=1, otherwise=0), a dual-earner household dummy (dual-earner=1, single- and non-employed households2)=0), age of household head dummies (40s, 50s, 60s, and 70s–90s=1, otherwise=0), household income (deflated by CPI), number of children (0–6, 6–15, and 15–18 years old), a 3-generation household dummy (3-generation=1, otherwise=0), and residential area dummy (5 major metropolitan areas=1, otherwise=0).
The results are reported in Tables 1(ii) and (iii). First, the effect of dual-earner households is consistently confirmed based on the coefficients of the interaction between the dual-earner household and year dummies, and the average marginal effect for each year (i.e., the percentage change in the dependent variable resulting from a one-unit change in each variable). However, second, the effect varies across years. That is, the effect increased relative to 1990 until 2010, peaked in 2015, and fell below the 1990 level by 2020, thus indicating that the effect became marginal.
First, the effect of dual-earner households is consistently identified in Japan over the 1990–2020 period, during which the shift to dual-earner households was advancing. Drawing on qualitative research (e.g., Epp & Velagaleti, 2014; Thompson, 1996), which demonstrates that the use of outsourced household products often involves processes of negotiation and compromise regarding mothers’ roles, this finding can be interpreted as follows: under the persistent family norm that continued to assign housework to women, the household division of labor was slow to change even in dual-earner households, thereby making their tendency to rely on prepared foods as a partial supplement to housework more pronounced. This possibility warrants further detailed examination.
Second, the effect of dual-earner households changes over time, and a time lag exists in the growth of the share of prepared foods between dual- and single-earner households. This result may reflect changes in time-saving attitudes and family norms, even among single-earner households, as well as quality improvements through technical innovation.
Many previous studies reviewed in the introductory section adopted the household production function approach and focused on the physical time-saving effects in dual-earner households, where the time available for housework tends to be relatively limited. However, this approach cannot fully explain why such a tendency has also become common among single-earner households.
Not only efficiency but also the process through which family norms are reconstructed through consumption practices need to be considered to explain the spread among single-earner households. Qualitative research indicates that family norms are not fixed (being family) but constructed through consumption practices (doing family) (e.g., Epp & Price, 2008; Holt, 1998; Kastarinen et al., 2023), and reveals that consumers lessen conflicts over outsourced products by legitimizing their use (e.g., Epp & Velagaleti, 2014; Thompson, 1996). Drawing on these studies, the widespread adoption even among single-earner households reflects the reconstruction of household-related family norms, supported by legitimizing discourses regarding prepared foods—such as “wisely saving effort rather than cutting corners” or “just a handmade touch will do.” Future research should examine how meanings related to “homemade” (Moisio et al., 2004) are reinterpreted as they intersect with outsourced products, and how legitimizing discourses regarding their use affect the reconstruction of family norms (Humphreys, 2010).
The theoretical contributions of this study are twofold. First, by examining temporal changes in household outsourcing in Japan, the work identifies the phenomena overlooked by earlier studies using the household production function approach and conducting only single-period analyses (e.g., Nickols & Fox, 1983; Strober & Weinberg, 1980; Weinberg & Winer, 1983). Second, drawing on qualitative research on family identity construction (e.g., Epp & Price, 2008; Epp & Velagaleti, 2014; Kastarinen et al., 2023), the present study highlights the need to conceptualize household outsourcing as a cultural phenomenon involving the reconstruction of family norms.
In addition, the negative coefficient of the residential area dummy also warrants further examination. Higher car ownership and larger dwelling spaces in rural areas may facilitate bulk purchases of frozen and prepared foods, whereas the greater variety of food options in urban areas may limit the extent to which households rely on such foods.
Finally, this study has limitations in that it captures only part of increasingly diverse family forms, thus leaving room for interpretation of the results. Other outsourcing modes, such as dining out, may also yield different results, thus requiring caution. Nevertheless, using newly released large-scale data, this study provides an empirical examination of Japan, clarifies current patterns, and identifies future research directions, which constitute its contributions.
The author is grateful to the anonymous referees for their valuable comments on this paper. The author also acknowledges the Statistics Bureau of Japan, Ministry of Internal Affairs and Communications, for providing the microdata. The data used in this study were independently constructed and analyzed by the author based on the provided microdata, and any remaining errors are solely the responsibility of the author. This research was supported by JSPS KAKENHI Grant Number 22K01764.
The authors have no conflicts of interest to disclose regarding this manuscript.
1) “Foods partially or fully prepared through cooking normally done at home or in restaurants” (“Classification of Income and Expenditure Items,” Statistics Bureau, MIC), such as frozen, ready-to-eat packaged, and chilled foods, as well as prepared dishes, boxed lunches (bento), and rice balls (onigiri).
2) In Japan, population aging has increased the share of older adult households, many of whom have become non-working after retirement. To capture the overall view, they are included in the single-earner household category.
Yuichiro Hidaka
Professor, College of Business Administration, Ritsumeikan University. Completed the Doctoral Program (Ph.D. in Commerce) at the Graduate School of Business Administration, Kobe University. Previously served as Associate Professor, Faculty of Business Administration, Yamanashi Gakuin University, and Professor, Graduate School of Humanities and Social Sciences, Okayama University, before assuming the current position.
https://ror.org/00hhkn466 
