A Gendered Analysis of Individual-Level Asset Poverty in Ecuador

Abstract

This study uses individual-level wealth data to explore the gender dimensions of asset poverty among the principal adults in Ecuadorean households, the first such study in a developing country. The study departs from conventional practice by analyzing not only sole heads but also partnered men and women heads and show systematic differences by gender, marital status, and household type. Among both sole and partnered heads, women are more likely to be asset poor than men. Further, in contrast to developed countries, asset poverty rates are much higher among partnered men and women than among sole men and women heads, largely because of structural factors that put those in consensual unions at a disadvantage in accumulating assets. The gender gap, however, is much larger among sole heads. In Ecuador, the risk of asset poverty is mainly associated with low levels of education, type of employment, and not having received an inheritance.

HIGHLIGHTS

• This study uses individual-level asset data to explore the extent of asset poverty in Ecuador.

• Women in Ecuador are more likely to be asset poor compared to men.

• The gender asset gap is more prominent among sole heads of household.

• Asset poverty rates are the highest among women in consensual unions.

• Education, employment, and inheritance significantly explain asset poverty.

KEYWORDS:

• Asset
• poverty
• wealth
• Ecuador
• feminization of poverty
• vulnerability

JEL Codes:

• I32
• D31
• J16

SUPPLEMENTAL DATA

Supplemental data for this article can be accessed at https://doi.org/10.1080/13545701.2021.1995019https://doi.org/10.1080/13545701.2021.1995019.

Notes

1 See Cheryl Doss (2013) for a summary of both the theoretical and empirical bargaining power literature.

2 Estimate based on EAFF survey, subsequently described.

3 On alternative definitions see Melvin L. Oliver and Thomas M. Shapiro (1990) and Michael R. Carter and Julian May (1999).

4 Several studies have shown that in developing countries livestock can provide an instrument of liquidity, particularly in times of need (Dercon 1998; Moll 2005). Investing in livestock is often equivalent to holding a savings account or insurance.

5 The Ecuador survey was undertaken as part of the Gender Asset Gap Project, which also collected individual-level asset ownership data in Ghana and India. For details of the project, see https://sites.google.com/view/genderassetgap. The Ecuador data set is available at: flacsoandes.edu.ec/pagina/60759-documentos-de-encuesta-activos-flacso-florida.

6 For the purposes of the survey, a household was considered to include all those who lived together and pooled some income and consumption. Most households surveyed, 68.5 percent, were made up by a main couple; 24.8 percent consisted of a sole woman and 6.7 percent, a sole man head.

7 Figure S1 shows total wealth, semi-liquid, and liquid wealth values at different quantiles of the distribution along with the asset poverty lines.

8 We run a sensitivity test to the choice of time period by estimating poverty rates for different reference periods (from three-month to nine-month; see Table S3, Online Appendix). Depending on the measure considered, the asset poverty gaps vary from the six-month gap by 1–4 percentage points.

9 For partnered men and women heads, we run independent marginal effects and random-effects probit regressions to account for unobserved household-specific effects. The results do not differ significantly. We report marginal effects for ease of interpretation.

10 “Non-income earners” includes those who do not report a primary or secondary occupation and those who are unpaid family workers. “Daily wage or domestic worker” is used as the reference group.

11 Among the eighty-three partnered individuals who reported “domestic worker” as their primary occupation, there were only two men.

12 We have also run separate regressions for heads who are married and in a consensual union (Tables S5 and S6, Online Appendix). While the signs tend to be similar for the two groups and by sex, the coefficients for married heads are more frequently significant across measures of wealth. Among the notable exceptions is with respect to household composition. The number of young children increases the odds of asset poverty for married women by the semi-liquid and liquid poverty measures, and for men in consensual unions by the semi-liquid measure. Also, the number of working teens and other adults increases the odds of asset poverty for married men by the semi-liquid measure; this is the only regression where this variable is significant. Also, being white or mestizo as opposed to indigenous or Afro-Ecuadorian also increases the risk of asset poverty for women in consensual unions once housing wealth is excluded; for married women, being white or mestizo decreases the odds for total wealth.  

Additional information

Notes on contributors

Boaz Anglade

Boaz Anglade is Postdoctoral Researcher at the University of Florida’s Institute for Food and Agricultural Sciences. His research interests focus on gender disparities in wealth, poverty, and diffusion of agricultural technologies in developing countries.

Pilar Useche

Pilar Useche is Associate Professor of Latin American Studies and Food & Resource Economics and Affiliate Professor of the School of Natural Resources and Conservation at the University of Florida. Her current research centers on sustainable and equitable economic development, technology adoption and disease management, and innovative methods and metrics in applied economic research.

Carmen Diana Deere

Carmen Diana Deere is Distinguished Professor Emerita of Latin American Studies and Food & Resource Economics at the University of Florida and Distinguished Professor at the Latin American Faculty of Social Science (FLACSO) in Ecuador. She was co-PI of the Gender Asset Gap Project, and directed its Ecuador study. Her research interests center on women’s property rights and ability to accumulate assets.