The distribution of male-headed and female-headed households in Gauteng


In South Africa, three out of every five households have a male head while two out of every five have a female head (Statistics South Africa, 2021). The Gauteng province has a similar ratio. (This refers to households that have adult members and therefore do not include child-headed households). However, as we show in this article, this ratio varies considerably between municipalities and, at a more local level, between wards. We also show that, when mapped, there is a visible relationship between the ratio (either more male-headed households or more female-headed households per ward) and the socio-economic circumstances of the area. This raises a series of puzzles:

(1) What are the associations between the sex of the household head and socioeconomic outcomes?

(2) Why do some areas have a majority of households with male heads while others have a majority of households with female heads?

(3) What processes produce households with male heads and what processes produce households with female heads? This article offers an initial exploration of these puzzles.

Sex of household head and socioeconomic outcome

While not all female-headed households are worse off than their male-headed counterparts, there is a general association between the sex of the household head and wellbeing. Research shows that female-headed households are more likely to experience poverty when compared to male-headed households (Chant, 2004; Goebel et al., 2010; Rogan, 2014). To take a specific outcome, children from female-headed households have been found to fare less favourably in education outcomes compared to others (Ndagurwa and Nzimande, 2016; Newlin, Reynold, and Nombutho, 2016). These kinds of associations are complex to unravel. The sex of the head of household is just one variable that can be associated with the development of households. Some research has shown that the sex of the household head was not as strongly associated with development outcomes as other variables, such as the income of the household head (Pashapa and Rivett, 2017). However, the sex of the household head can make a material difference in particular respects. One reason is that female-headed households have been found to contain, on average, fewer income earners whose earnings tend to be lower than those of their counterparts from male-headed households (Posel, 2001). Another is that female-headed households are more likely than male-headed households to have members that are all unemployed (Nwosu and Ndinda, 2018). More generally, women tend – on average – to earn less than men as a result of biases in the labour market. In summarising the complex association between development outcomes and the sex of the household head, we can observe that the sex of the household head can make a material difference to the welfare of the household because of social biases in the treatment of women and because of the structure of such households.

Male-female household headship at municipal level

Nationally, most households are headed by males. Research done by Jhamba and Mmtali (2015) found this to be the case in about 60% of households. In the Gauteng province, data from the GCRO Quality of Life (QoL) survey 6 (2020/21) showed that the percentage of households headed by males is also roughly 60%. Although this is consistent with the national picture, there are notable variations in the ratios of male-to-female households between municipalities in Gauteng (Figure 1).

Figure 1: Distribution of male- and female-headed households in Gauteng and the municipalities

As shown in Figure 1 above, three of the nine Gauteng municipalities – Emfuleni, Lesedi, and Midvaal – have household headship patterns that are very different from the provincial average. The Midvaal municipality has the highest proportion of households headed by males (69% male- and 31% female-headed). By contrast, 52% of the households in Emfuleni municipality are headed by males while females head 48%. At 53% male and 47% female headship, Lesedi municipality has the second-highest proportion of female-headed households in Gauteng.

Male-female headship patterns within municipalities

In Figure 2, we show a more granular picture through ward averages.


Figure 2: Map showing ward-level prevalence of male and female household headship in Gauteng for households with adult members, QoL 6 Survey 2020/21

While most households in Gauteng are headed by males, there are many areas in the province where female headship is more prevalent than male headship. In Emfuleni local municipality, shown to have the highest proportion of female household headship in the province (Figure 1), there is a well-defined clustering of female headship in small areas in the low-income townships of Evaton and Sebokeng – North of Vanderbijlpark (O) – where 60 to 86% of households are headed by females. Meanwhile, the high-income suburbs of Vereeniging in Emfuleni municipality – east of Vanderbijlpark (O) – have as high as 75% to 97% of households headed by males (Figure 2).

The metropolitan municipalities also show a clustering of female household headship, especially in low-income areas. The southern parts of Johannesburg, particularly those in or near Soweto (K), have greater proportions of households headed by females, from 60% rising to as high as 75% (Figure 2). Meanwhile, in the northern areas of Johannesburg, larger proportions of households are headed by males. Tshwane also displays well-defined geographic clustering of male-female household headship. In areas generally associated with lower socioeconomic status, like Hammanskraal (A) and Soshanguve (B), female-headed households are more prevalent than male-headed households. This contrasts with suburbs in the southern parts of Tshwane, like Centurion (D), where most households are male-headed. In Ekurhuleni, areas around Katlehong (L) have higher proportions of female-headed households.

The headship patterns shown in this map raise several implications for how to understand household demography in Gauteng, and South Africa more generally. The localised nature of concentrations of female headship, which in some wards reaches 86% of all households, shows the importance of a small-scale analysis of household demography. In the context of gender-related poverty, inequality, and violence, household headship becomes relevant to policies related to social development and policing, among others.

Sex of household head and income

The Quality of Life survey data allow us to further interrogate the associations between the sex of the household head and socioeconomic indicators like income. When we plot the distribution of each type of household against income levels, we can see that although both male- and female-headed households are represented in all income bands, female-headed households have higher over-representation in lower-income bands compared to male-headed households. When we combine the lower five income bands, we can see that 65% of female-headed households earn R3 200 monthly or less while 47% of male-headed households earn R3 200 or less.

Figure 3: Percentage distribution of male and female headed households by average monthly income

The data represented in Figure 3 provide one interpretation on the geographical variation seen in Figure 2. Residential areas in townships and inner cities have households that have higher proportions of female heads, have lower incomes, and – in many cases – both of these characteristics together.

Factors that promote female headship

The pathways through which households come to be female-headed are diverse. It would be incorrect to suggest that female-headed households are a deviation from the norm (of male-headed households). Female headship can be a choice, as in situations where women seek autonomy from men (Hunter, 2010). That said, some households are female-headed because of factors that female heads would not choose such as the death of male partners, or fathers who do not take responsibility for children. We know from demographic processes that in general, males have higher mortality than females across all age groups. As a result, females survive their male partners and end up as heads of households, as in the case of those headed by elderly women. Some households have fathers that do take responsibility for their children but live elsewhere, possibly to secure livelihoods. Meanwhile, some women move to particular places for work with their children. Finally, some households are female-headed because they comprise same-sex couples.

Some of the wards of Figure 2 are shaded blue as a result of quite specific processes. For instance, many of the residents of Meadowlands were among the first occupants of Soweto (K), having been forcibly relocated there from Sophiatown in the 1950s. In 1954, the National Party government passed the Native Resettlement Act (No 19 of 1954) which allowed the state to remove Black African residents from any area within and close to the magisterial district of Johannesburg which required the relocation of Sophiatown residents to Meadowlands (Lodge, 1981). Given the age of this group of residents, many male partners have died leaving behind their female partners.


This Map of the Month article sought to analyse the geographic patterns of male and female household headship in Gauteng. We used data from the GCRO QoL 6 survey of 2020/21. The results showed that most of the households in the province are headed by males, but that male-female household headship patterns are not uniform across the province. Some municipalities like Emfuleni and Lesedi have higher proportions of female-headed households while Midvaal has the least. We also found variations in household headship within municipalities. The main finding in this regard is that low-income areas tend to have greater proportions of female-headed households while male headship is more prevalent in areas associated with higher incomes.

Note on methods

This work is based on data from the GCRO QoL Survey 6 (2020/21). In this analysis, several questions (variables) were used, and these pertain to the sex of the respondent, the relationship of the respondent to the household head, monthly household income, municipality, and ward. The computation of headship patterns made use of survey weights to allow for generalisation to the total number of households in Gauteng with at least one adult member. It is important to note that some households may be potentially treated as female-headed while they are male-headed, or male-headed while they are female-headed because they comprise same-sex couples, but the potential impact this has on the headship patterns reported in this analysis is arguably minimal. We can use other countries with readily available data on same-sex couples as benchmarks. In the United States of America, 1.8% of all unions are same-sex, and in Norway 1.9% of all unions are same-sex (Rault, 2023; Zahl-Olsen and Thuen, 2023). These two countries can be considered to have progressive acceptance of same-sex unions comparable to South Africa. Assuming similar proportions in the QoL Survey 6 (2020/21), this will translate to a maximum of 39 households that are potentially misclassified as male/female-headed when the opposite is true. Therefore, the patterns of male and female household headship reported in this piece closely reflect the state of households in the Gauteng province. All analyses were carried out in Stata 17.0, and results were exported to a Geographic Information System (GIS) software for mapping, and the Datawrapper platform for plotting in graphs.


Chant, S., (2004). Dangerous Equations? How Female-headed Households Became the Poorest of the Poor: Causes, Consequences and Cautions. IDS Bulletin 35, 19–26. Available at:

Goebel, A., Dodson, B., Hill, T., (2010). Urban Advantage or Urban Penalty? A Case Study of Female-headed Households in a South African City. Health & place 16, 573–580.

Hunter, M. (2010) Love in the Time of AIDS: Inequality, Gender, and Rights in South Africa. Indiana University Press.

Lodge, T. (1981). ‘The Destruction of Sophiatown’, The Journal of Modern African Studies, 19(1), pp. 107–132. Available at:

Ndagurwa, P. and Nzimande, N. (2016). ‘The Impact of Family Structure on Schooling Outcomes for Children in South Africa’, in Children in South African Families: Lives and Times.

Newlin, M., Reynold, S. and Nombutho, M.M.-W. (2016). ‘Children from Child-headed Households: Understanding Challenges that Affect in their Academic Pursuits’, Journal of Human Ecology, 54(3), pp. 158–173. Available at:

Nwosu, C. and Ndinda, C. (2018). ‘Gender-based Household Compositional Changes and Implications for Poverty in South Africa’, Journal of International Women’s Studies, 19(5), pp. 82–94.

Pashapa, T. and Rivett, U. (2017). ‘Gender of Household Head and the Digital Divide in South Africa’s Settlements’, Gender, Technology and Development, 21(3), pp. 232–249. Available at:

Posel, D.R. (2001) .‘Who are the Heads of Household, What do They do, and is the Concept of Headship Useful? An Analysis of Headship in South Africa’, Development Southern Africa, 18(5), pp. 651–670. Available at:

Rault, W. (2023). Same-sex unions in high-income countries: more widely recognized and more frequent. Population and Societies.

Rogan, M. (2014). Poverty may have Declined, but Women and Female-headed Households Still Suffer Most. Grahamstown [en línea] http://www. econ3x3. org/article/poverty-may-have-declined-womenand-female-headed-households-still-suffer-most.

Statistics South Africa, (2021). ‘General Household Survey, 2021’ [Accessed 27-02-2023].

Zahl-Olsen, R., & Thuen, F. (2023). Same-sex Marriage Over 26 Years: Marriage and Divorce Trends in Rural and Urban Norway. Journal of Family History, 48(2), 200–212.

Recommended citation: Ndagurwa, P., Naidoo, L. & Miles-Timotheus, S. (2023). The distribution of male-headed and female-headed households in Gauteng. Map of the Month. Gauteng City-Region Observatory. March 2023.

Review and edits: Dr. Richard Ballard, Graeme Götz, Gillian Maree, and Christian Hamann.

Map design: Jennifer Murray


The GCRO sends out regular news to update subscribers on our research and events.