Document Type

Migration Policy Series

Publication Date



Balsillie School of International Affairs


Migrant remittances have become an important source of income for many developing countries, exceeding official development assistance. As a result, migration and remittance behaviour are becoming a growing focus of international attention. Understanding the processes and patterns of remittance behaviour can help shed light on their usage and impact, both on recipient households and on wider socio-economic development in migrant-origin countries. One key aspect of such an understanding is the gender dynamics of migration and remittance practices. Globally, there is evidence of the feminization of migrant flows, with women increasingly migrating as independent migrants in their own right. Female migrants maintain strong ties to family members in their home countries. These include significant flows of remittances, of both cash and goods, sent to family members at home.

Southern Africa has a long history of cross-border migration and associated flows of remittances. Although cross-border economic migration in the region has been dominated by male migrant labour to the South African mining industry, women have also engaged in movement across the region’s borders for purposes of seeking work. Evidence suggests that female migration in the region, especially to South Africa, has increased significantly over the past 10-15 years. Little is known about the nature of migrant women’s remittances and their impact on the households that receive them, nor about the changing patterns of male and female migration over the past decade.

SAMP devised the Migration and Remittances Surveys (MARS) to provide nationally-representative data on remittance flows and usage at the household level for five SADC countries: Botswana, Lesotho, Mozambique, Swaziland and Zimbabwe. SAMP-led research teams in each country conducted the survey using a standardized questionnaire and sampling strategy. Households were randomly selected and included in the survey only if they had members who were cross-border migrants working outside the country.

The MARS survey collected two different types of data: data on individual household members, both migrant and non-migrant, and data on migrant-sending households. In total, 4,700 household interviews were conducted in the five countries and information collected on over 30,000 people. In addition to questions about migrant destinations, occupations and demographics, questions were asked about remittance behaviour, the methods used for remittance transfer, the role of remittances in the migrant-sending household economy, and the impact of migrant remittances on migrant-sending households. Gender-related variables were included in the survey through a question asking the sex of individual household members, and also in terms of household headship, marital status, relationship to the head of household, and household type (e.g. female- or male-centered; nuclear or extended).

The overall survey findings have been presented in an earlier SAMP report (Migration Policy Series No. 44). Building on that earlier report, this report presents a gender breakdown and analysis of the MARS findings. As the Botswana sample included only a very small number of female migrants, Botswana has been left out of the analysis and so the report covers the four countries of Lesotho, Mozambique, Swaziland and Zimbabwe. Gender analysis reveals significant gender-based differences in migrant demographics, including divergent patterns and trends between male and female migrants in terms of their age and marital status. Yet it also reveals striking similarities between male and female migrants in terms of the nature, role and impact of their remittances.

The main findings in terms of migrant demographics and migration trends for men and women are as follows:

  • Cross-border migration in the region remains dominated by men. Overall, just over 15% of the migrants identified in the MARS sample were women. The proportion of women amongst migrants varies widely from country to country. Zimbabwe stands out as the country with by far the highest proportion of migrants who are female, at 43.6%, with Lesotho a distant second at 16.4%. In the other countries surveyed, the proportion of migrants who are female was found to be below 10%.
  • Although the lack of reliable benchmark data makes it impossible to quantify, the MARS data suggests that there has been an increase in female migration over the period 1994-2004. Compared to female migrants, a relatively higher proportion of male migrants had been migrating for periods of over a decade - except in the case of Zimbabwe, where extensive out-migration is a more recent phenomenon for both men and women.
  • The type of women who migrate appears to have undergone significant changes. Migration surveys conducted by SAMP in the late 1990s showed that female migrants were more likely to be older and to be married than male cross-border migrants. MARS findings show that today’s female migrants are younger and less likely to be married than male migrants.
  • The proportion of female migrants in the younger, 15-24 age bracket is significantly higher than the equivalent proportion for males in Lesotho, Mozambique and Swaziland. In these same three countries, men are correspondingly over-represented in the older, 40-59 age bracket compared to women. Zimbabwean male and female migrants’ age profiles were roughly equivalent, with the majority in the 25-39 age bracket.
  • In all four countries, and especially in Mozambique and Swaziland, the category containing the highest proportion of female migrants was ‘daughter’ – almost three-quarters of the female migrants in Mozambique and two-thirds in Swaziland. These countries’ female migrants thus most closely conform to the historical pattern of migration to South Africa being dominated by young, unmarried adults.
  • On the male side of the equation, migration appears to be increasing among older, married heads of household. Some 76% of male migrants from Lesotho and 61% of those from Swaziland are heads of their household. Male migration from Mozambique is still dominated by sons (49%); that from Zimbabwe is more or less equally divided between sons and household heads. This suggests that for men, migration is becoming a career path rather than just a temporary phenomenon at a particular stage in younger men’s lives, whereas young single women are engaging in migration practices traditionally found amongst young single men.
  • In Lesotho, Mozambique and Zimbabwe, female migrants revealed higher levels of divorce, separation, abandonment and widowhood than their male counterparts. This is especially true of Lesotho, where 24% of female migrants are widows and a further 20% divorced or separated. These women are likely to be the primary or sole breadwinner for their families.
  • Female migration and female household headship appear to be closely linked. Among female migrants, the proportion coming from female-centred households, having no husband or male partner, was 43% for Lesotho, 41% for Mozambique, 31% for Swaziland and 28% for Zimbabwe. In addition, 24% of the female migrants from Lesotho and 17% of those from Zimbabwe, the two countries with the highest proportions of women among their migrants, were themselves household heads. Male migrants hail predominantly from male-headed, nuclear or extended-family households.
  • Female migrants from Lesotho and Swaziland are better educated than male migrants. Men from these countries are over-represented in the categories of ‘none’ or ‘primary’ education, while women migrants are more likely than men to have some secondary schooling. Mozambique has the least educated migrants of the four countries, with close to three quarters of both male and female migrants having only primary education. Zimbabwean male and female migrants have roughly equivalent education profiles, with a highly educated migrant cohort in which over 75% have secondary education or above.

The geographical and economic profiles of male and female migration also display both similarities and divergences. Here, the main findings are as follows:

  • For both male and female migrants, the main destination is South Africa. Lesotho’s migration is almost entirely (99%) to South Africa, as is that of Swazi men. Some women migrants from Swaziland can be found working in countries beyond the region (13.5%). Mozambique sends small numbers of migrants, especially women, to Swaziland, Botswana and other SADC countries in addition to South Africa. Zimbabwe is again the exception, with only one third of its migrants (male and female) in South Africa and 40% working in countries beyond Southern Africa.
  • Perhaps the greatest difference between male and female migrants is in their activity and employment profiles. Minework is still the predominant form of employment for male migrants from Lesotho, Swaziland and Mozambique. Almost 80% of male migrants from Lesotho and two-thirds from Swaziland work on the South African mines. In the case of Mozambique, the figure is one-third. Male migrants from Mozambique also work in a range of non-mining occupations including skilled and unskilled manual labour (18%). Zimbabwe’s more educated male migrants work in professional and service occupations, while others are engaged in trade. Few men from the other three countries listed trade as an occupation.
  • In general, women migrants are spread across a wider range of occupations than their male counterparts. Relative to male migrants, female migrants are less likely to be in formal employment and more likely to be engaged in informal economic activity. Trading is a significant economic activity for female migrants from all four countries, with trade being particularly important as an occupation for women from Mozambique and Zimbabwe. Domestic service is a more significant form of employment for women from Lesotho and Swaziland. Informal sector production is another important occupation for female migrants. Agricultural, manual and ‘other service’ work occupy a small but significant number of migrant women. Among more skilled women, professional and office occupations are common, and 16% of Zimbabwe’s female migrants are employed in the health sector.

Given this gender difference in occupation and employment, any similarities or differences between men and women in their remittance practices, and in the extent to which their households depend on those remittances, are of interest. The MARS findings outlined below demonstrate the extent and significance of remittance income to recipient households.

  • For most migrant-sending households, migrant remittances form the main source of household income, although male migrants’ remittances are more likely to be the primary or sole source of income for their households. Lesotho has the highest incidence of households reporting remittance earnings, followed by Zimbabwe, Mozambique and Swaziland. Households reporting remittance income from male and female migrants respectively in each of the four countries was: Lesotho 96% (M) and 90% (F); Zimbabwe 85% (M) and 78% (F); Mozambique 77% (M) and 65% (F); and Swaziland 63% (M) and 64% (F).
  • The amounts of money remitted by female migrants are significantly lower than those of male migrants, in part reflecting women migrants’ lower levels of income and employment security. Gender differences are most stark in Mozambique and Lesotho. Zimbabwe’s more gender-equivalent migration profile is again borne out in the remittance data, with men and women remitting similar amounts. The median annual values of remittances received by male migrant-sending households were: R9,600 in Lesotho; R2,011 in Mozambique; R2,400 in Swaziland; and R1,093 in Zimbabwe. For households sending female migrants, median remittance receipts were: R3,600 in Lesotho; R302 in Mozambique; R1,800 in Swaziland; and R1,093 in Zimbabwe.
  • Despite remitting less than men, Lesotho’s female migrants still remit larger sums than female migrants from any of the other three countries: twice as much as second-placed Swaziland, three times as much as women from Zimbabwe, and ten times as much as women from Mozambique. This probably reflects the higher incidence of household headship among Lesotho’s women migrants, possibly along with higher earnings than women from other countries (e.g. as domestic workers rather than informal traders).
  • Lesotho’s migrant-sending households displayed the highest dependence on remittance earnings, reporting fewer alternative sources of income. By contrast, many households in the other three countries had remittances as part of a bundle of incomeearning strategies, including wage work, casual work and formal or informal business, although remittances remain their primary source of income. Multiple sources of income were found especially among households sending female migrants. Female migrants remit lower sums, making other household income sources a necessity. Women are also less likely than men to be household heads, which means that they are often members of households with other working adult members.
  • The proportion of female migrants sending home goods is slightly higher than the equivalent proportion of male migrants, especially in Zimbabwe (72% of women, 62% of men) and to a lesser extent Lesotho (23% of women, 20% of men). In Mozambique and Swaziland, male and female migrants were equally likely to remit goods, at 65% and 16% respectively. Zimbabwe and Mozambique had the highest incidence of non-monetary remittances, whereas monetary remittances were much more significant in Lesotho and Swaziland.

The significance and impact of both male and female migrants’ remittances is evident in the contribution of remittances to household expenditure and the stated importance of remittances by recipient households:

  • Household expenditure data show that the main household purchases for both male and female migrant-sending households are the basic commodities of food, domestic fuel and clothing, and fundamental services such as schooling, health care and transport.
  • While the rank order of items purchased is broadly similar or even identical for male and female migrant-sending households, gender differences emerge in the actual proportion of households reporting a particular expenditure. In Lesotho, in almost every category, expenditure in the past month was reported by more male than female-migrant households. Gender-based patterns are more mixed in Mozambique, although lower proportions of female migrant-sending households reported expenditure in the key categories of food, clothing, medical expenses, education and transport. In Swaziland, there is no clear or consistent overall difference based on migrant gender. Zimbabwe displays the strongest similarity between expenditure in male and in female migrantsending households, consistent with findings from the rest of the survey in that country. Certainly in Lesotho and Mozambique, female migrant-sending households do thus appear to be poorer than male migrant-sending households.
  • There are also important gender differences, as well as differences between countries, in the estimated amounts of monthly expenditure on particular categories of expenses. Women migrants from Lesotho and Mozambique come from households with lower monthly expenditures, in almost every category, than households with male migrants. Swaziland displays the opposite gender pattern. Swazi households where the migrant members are female spend more in each category than households with male migrant members. Zimbabwe again stands out as the country with strongest gender similarity, suggesting that its male and female migrants come from similar sorts of households in socio-economic terms.
  • As the primary source of income for the majority of households, remittance earnings are vital in enabling households to meet their basic needs. Food is the most common annual expenditure of remittance money in all four countries and in both male and female migrant-sending households. Second in all countries is either clothing or school fees. Clothing or school fees also rank third in all countries except Swaziland, where purchases of agricultural inputs rank above clothing. Remittances do not appear to be spent on non-essential or luxury items; but nor are they commonly directed towards savings or investment in business or other productive activities. They are, however, significant in sources of investment in children’s education.
  • Remittance-receiving households confirmed the significance of remittances to food purchases. The most consistent importance rating, across countries and migrant genders, is food, with school fees and clothes also rated highly by many. There are some gender differences, with men’s remittances seemingly more crucial to the purchase of basic livelihood items such as food, than women’s. Given that men are older, more likely to be married, and more often the heads of households than female migrants, it is perhaps surprising that this gender difference is not greater.
  • Remittances of goods are also focused on basic household commodities. The ‘typical’ male or female migrant sends home money, which their households use to buy food and other basic goods and services, and brings home clothing, food and other goods. Some consumer goods and other ‘luxury’ items (e.g. electronic goods) are also sent home, as they are more readily available and cheaper in South Africa.
  • In addition to making regular remittances, migrants send home money in times of need, or to meet unexpected costs (such as funerals). Some gender differences are evident in these emergency remittances, although this is not consistent across all four countries. In Lesotho and Mozambique, a higher proportion of male migrants send money in times of need, whereas in Swaziland female migrants are more likely to do so. In Zimbabwe, once again, there is very little difference based on the gender of the migrant.
  • Emergency remittances are clearly important to the households receiving them. They are seen as important or very important by over 90% of migrant-sending households in each of the four countries, with only very small differences on the basis of migrant gender. Emergency remittances appear to be especially significant to households in Swaziland and Zimbabwe.

Perceptions of the overall impact of migration reinforce the overall positive contribution made by migrants:

  • There is a generally favourable view of cross-border migration. Respondents in Zimbabwe are the most positive. Close to 90% regard the impact of migration as either positive or very positive, with only a small difference based on the gender of the migrant. Respondents in the other three countries were broadly positive, although more so for male than for female migration. Close to 70% of the male migrant-sending household respondents in Lesotho, Mozambique and Swaziland regard migration as having positive or very positive impacts. The respective values for each country’s female migrant-sending households were 59%, 53% and 64%.
  • Perceptions of the positive impacts of working in another country reinforce the findings from income, expenditure and deprivation data i.e. that migrants support their households, improve living conditions and provide household income. Female migration is seen as providing the same sorts of benefits as male migration.
  • While the economic benefits of migration are recognized, so too are some of its personal and social costs. These include loneliness, being away too long, and placing too much responsibility on family members left behind. The broad patterns are the same, irrespective of whether the migrant is male or female.

Remittances clearly play a vital role in supporting Southern African households. Not only do migrants, whether male or female, demonstrate an unusually high tendency to send money home to their families, but those remittances are fundamental in enabling families to meet their everyday needs. Remittance behaviour and the role of remittances in the household economy differ only slightly based on the gender of the migrant. This demonstrates that women’s migration, while lower in volume than male migration, is nevertheless highly important to the migrant-sending household. Given that so many female migrants come from female-centred households, with no husband or male partner, women’s migration is especially significant to such households as the primary – often only – source of household income.

The MARS data also suggest that differences between male and female migration, and between male and female migrants, are starting to diminish. Certainly young, unmarried women appear to be engaging in ‘economic’ migration more than they did previously, while male migration is extending into broader spheres of economic activity, both formal and informal, as well as into older age cohorts. If the patterns and trends identified here are both valid and sustained, women’s cross-border migration in the region looks set to increase in extent and socio-economic significance.