Unfolding Patterns of Unpaid Household Work in Latin America

ABSTRACT

Although Colombia, Mexico, Peru, and Uruguay show similar empirical patterns in terms of time women devote to unpaid work, they also present important variations in how unpaid work is distributed between men and women. Using time-use surveys for the 2007–10 period, this study finds a uniform pattern across the four countries regarding the main individual-level variables related to the allocation of unpaid work. When decomposing the gender gap in hours devoted to unpaid work, most of the difference cannot be attributed to variations in observable characteristics of men and women: the unexplained part of the gap is the dominant part. Results suggest that both the strength of traditional gender roles and existing welfare architecture are relevant factors in understanding variations in how unpaid work is distributed between men and women in these four countries. The results reaffirm that powerful interventions are needed to shift gender norms about unpaid work.

KEYWORDS:

• Unpaid household work
• determinants
• Latin America
• time-use surveys

JEL Codes:

• C81
• D13
• C83

ACKNOWLEDGMENTS

The authors thank Pilar Manzi and Martin Brun for excellent research assistance, as well as the two anonymous reviewers for their insightful comments.

SUPPLEMENTAL DATA

Supplemental data for this article can be accessed at doi:10.1080/13545701.2017.1344776. The underlying research materials for this article can be accessed at https://sites.google.com/site/mceciliarossel/publications/journals.

Notes

1 This statistic is based on a simple average of seventeen countries, based on data from CEPALSTAT (ECLAC 2014).

2 This criterion for classifying activities as work and nonwork is not free from debate as, for example, it cannot be straightforwardly applied with the same results in different contexts and cultures (Wood 1997).

3 The 19th International Conference of Labour Statisticians, held in 2013, considers own-use provision of services and volunteer work in households producing services as activities beyond the 2008 SNA production boundary but inside the general production boundary. This provision of services covers: (i) household accounting and management, purchasing, or transporting goods; (ii) preparing or serving meals, household waste disposal, and recycling; (iii) cleaning, decorating, and maintaining one’s own dwelling or premises, durables, and other goods, and gardening; (iv) childcare and instruction, transporting and caring for elderly, dependent, or other household members, and domestic animals or pets.

4 Unpaid work also includes activities related to helping other households or the community in a broader sense, as well as some specific activities related to procuring inputs and producing for own use (Antonopoulos 2009).

5 The other way to collect data on time use consists of the diary approach: respondents are asked to report their activities for a 24-hour period. This approach is mainly used in European countries. The surveys conducted in Argentina (2005, 2013) also use the diary approach (Government of the City of Buenos Aires 2007; Instituto Nacional de Estadística y Censos [INDEC] 2013).

6 For a review on the main features and trade-offs in Latin American time-use surveys, see Valeria Esquivel et al. (2008).

7 International efforts for uniform classification of activities in time-use surveys include the International Classification of Activities for Time Use Statistics (ICATUS) and Clasificación de Actividades de Uso del Tiempo para América Latina (CAUTAL). Other examples of international efforts of data harmonization are the recommendations for Harmonized European Time Use Surveys (HETUS), developed by Eurostat; and the Multinational Time Use Study (MTUS), based currently at Oxford University.

8 All surveys are available online at the countries' respective national statistics offices' websites. For Colombia, see http://www.dane.gov.co/; for Mexico, see http://www.inegi.org.mx/; for Peru, see https://www.inei.gob.pe/; and for Uruguay, see http://www.ine.gub.uy/.

9 On theoretical grounds, the surveys used in this article do not allow for simultaneity of tasks, although three of the databases include a small percentage of individuals with total hours of work higher than 135 (0.3 percent in Colombia and Mexico, and 2.4 percent in Uruguay).

10 If zeros in time-use data arise from a mismatch between the reference period of the data and the period of interest, then a Tobit model may not be adequate. For methodological discussions on this issue in relation to time-diary data, see Jay Stewart (2009) and Gigi Foster and Charlene M. Kalenkoski (2013). In previous versions of this paper we also reported OLS results, which were very similar to Tobit results and led to the same conclusions.

11 The Gender Values Index is based on the following questions from the World Values Survey, which asks the interviewee to declare if she or he agrees or disagrees with each phrase: (1) “When jobs are scarce, men should have more right to a job than women”; (2) “If a woman earns more money than her husband, it’s almost certain to cause problems”; (3) “Having a job is the best way for a woman to be an independent person”; (4) “When a mother works for pay, the children suffer”; (5) “On the whole, men make better political leaders than women do”; (6) “A university education is more important for a boy than for a girl”; (7) “On the whole, men make better business executives than women do”; and (8) “Being a housewife is just as fulfilling as working for pay.”

12 Comparisons should be done very cautiously, given the differences regarding the collection of data on time use.

13 The highest gender gap in total work hours corresponds to Peru, where paid work hours are relatively high for women, and women’s labor participation is high. Although market work is high among women in Peru, 17 percent of women workers are unpaid family workers (ECLAC 2014). Strictly speaking, all market work (including unpaid work) is considered paid work in this article (following standards for the construction of labor market indicators).

14 As Peruvian men do not show the inverted U pattern, data for them are not included in the Supplemental Online Appendix.

Additional information

Notes on contributors

Verónica Amarante

Verónica Amarante is Director of the Montevideo Office of the Economic Commission for Latin American and the Caribbean (ECLAC, United Nations). She is also Professor in the Department of Economics at Universidad de la República in Uruguay. She holds a PhD in Economics from Sussex University. Her research focuses on poverty, inequality, labor markets, and social policies.

Cecilia Rossel

Cecilia Rossel is Assistant Professor in the Political and Social Sciences Department at Catholic University in Uruguay. She is a sociologist and holds a PhD in government and public administration (Instituto Universitario Ortega y Gasset – Universidad Complutense de Madrid). Her research focuses on the political economy of social policy, and welfare regimes and inequality in Latin America.