Policies to support women’s paid work

Policies in developing countries to improve women’s access to paid work should also consider child welfare

Gianna Claudia Giannelli Giannelli, G (University of Florence and CHILD, Italy, and IZA, Germany)

Elevator pitch

Engaging in paid work is generally difficult for women in developing countries. Many women work unpaid in family businesses or on farms, are engaged in low-income self-employment activities, or work in low-paid wage employment. In some countries, vocational training or grants for starting a business have been effective policy tools for supporting women’s paid work. Mostly lacking, however, are job and business training programs that take into account how mothers’ employment affects child welfare. Access to free or subsidized public childcare can increase women’s labor force participation and improve children’s well-being.

Women’s jobs are more vulnerable than
                        men’s (percentage of total employment by gender), 2013

Key findings


  • Job training and wage subsidies encourage female paid employment in the formal sector.

  • Business training increases female labor force participation and raises the income of low-earning women already in business.

  • In low-income households, access to free or subsidized public childcare increases women’s labor force participation.

  • Child health and nutrition may improve with female paid employment if the money spent on children rises.

  • Female paid employment improves child schooling.


  • The positive effects of training and wage subsidies tend to disappear after one or two years.

  • Business training increases profitability only in the short term and is only effective in starting subsistence businesses.

  • The effects of childcare programs on female labor supply are not clear.

  • When women work, children’s well-being can decline, since mothers have less time for childcare.

  • Women in poorer households are more likely to work, but their labor income is not enough to increase school attendance of children.

Author's main message

Training programs in developing countries have been effective in boosting female employment and earnings. However, the impact depends on whether the programs target younger or older women, new entrants or women with labor market experience, or women in rural or urban areas, as well as other factors. Policies to enhance women’s access to paid work and support their efforts to start a business should also consider any potential negative spillover effects on child well-being, since women still bear most of the burden of childcare. Integrating public provision of childcare with training programs can make it easier for women to engage in paid employment.


For women everywhere, holding a decently paid job is the most direct way to achieve empowerment. In developing countries, however, many women work unpaid in family businesses, in their own low-profit businesses, or in firms for very low wages. In these countries, female employment and earnings are heavily constrained by women’s low levels of education. In some cases, formal education does not provide the skills that firms are seeking. In other cases, the education system has begun to invest in programs that meet labor market skill requirements, but women who are already out of formal education are left behind.

Job and business training programs are policy tools that aim to fill the skill gap for people who are out of the education system. The most effective labor market policies for supporting female paid employment take a lifecycle approach. Thus, policy measures designed to ease the transition from school to work for younger women ought to differ from those designed to help older women reconcile paid employment with household responsibilities and gender norms. For women with children, child well-being is a major concern, and policies can be designed to jointly enhance women’s paid employment and child development.

Discussion of pros and cons

Women’s status of employment differs across developing regions

In sub-Saharan Africa, North Africa, and South Asia, unpaid work accounts for well over a third of female employment (Figure 1). In Asia and sub-Saharan Africa, small low-income generating businesses are widespread. In Latin America, wage employment is more typical for women. Policy measures need to take into account women’s different status in employment.

Women’s predominant status in employment
                        differs by region, 2013 (percentage of total employment)

Status in employment

Wage and salaried workers (employees) hold paid employment jobs, with explicit (written or oral) or implicit employment contracts that promise a basic remuneration that is not directly dependent on the revenue of the unit for which they work. They are typically remunerated with wages and salaries but may be paid in commission from sales, piece-rates, bonuses, or in-kind payments, such as food, housing, and training.

Self-employed workers includes individuals who engage in work on their own or with one or a few partners. They hold jobs in which their remuneration depends on the profits derived from the goods and services they produce. Own-account workers work alone or with one or a few partners, without engaging employees on a continuous basis to work for them. Employers or entrepreneurs are workers who hire one or more people to work for them as employees on a continuous basis.

Contributing family workers (unpaid family workers) are unpaid workers who work in a market-oriented establishment operated by a related person living in the same household. They cannot be regarded as partners, because their degree of commitment to the operation of the establishment is not at a level comparable to that of the head of the establishment.

Vulnerable employment is the sum of own-account workers and contributing family workers. Vulnerable employment is used as an indicator for measuring progress toward achievement of the Millennium Development Goal of reducing poverty and hunger.

Source: United Nations. The World’s Women 2005—Progress in Statistics. New York: UN Department of Economics and Social Affairs, Statistic Division, 2006; p. 161. Online at: http://unstats.un.org/unsd/demographic/products/Worldswomen/ww2005_pub/ww2005_complete_report.pdf; International Labour Organization. International Classification by Status in Employment (ICSE, 1993). Geneva: Department of Statistics, 2010. Online at: http://laborsta.ilo.org/applv8/data/icsee.html

Women’s status in employment can vary greatly across countries, even within the same region. Wage employment has the largest share in Brazil, Colombia, and the Dominican Republic, as it does in Jordan and Sri Lanka. But wage jobs may not be good jobs. The shares of own-account work and contributing family workers, the two most vulnerable types of employment, are the largest by far in India, Ghana, and Uganda, in descending order.

Training and wage subsidies

A lack of skills is one of the main obstacles to women’s greater participation in the labor market in low- and middle-income countries. Training programs offer women the opportunity to develop the job skills required by employers or needed to run their own business. Training programs for wage employment combine attendance on courses with on the job training. Wage subsidies can help match firms and workers. Programs offering business training for self-employment provide vocational training and, sometimes, cash grants for starting a business.

Training programs and wage subsidies can improve women’s wage employment under some conditions

Since the 1990s, training programs have been especially common in Latin America. Because of a shortage of skilled labor, firms in Argentina, Brazil, Chile, Colombia, the Dominican Republic, Honduras, Panama, Peru, and Uruguay favor these programs. Unemployment rates are high among the young people the programs are intended to reach, and youths who are employed often have low-quality, low-paid jobs in the informal sector. These training programs have been implemented mainly in urban areas, and women aged 21–27 make up a large majority of participants (the largest share is 74% in Colombia).

The impact of training on the target population has been analyzed through several experimental design studies using randomized controlled trials. The evaluation objectives in these studies are to determine whether a program has increased the employment probability and earnings of participants compared with outcomes for a group of individuals with similar characteristics who did not participate in the programs. Since randomization has to occur before a policy is implemented, only a handful of programs have the data needed for evaluation, and most of these are recent. Thus, the great majority of evaluations are based on non-experimental design studies. For the most part, however, the results for both experimental and non-experimental designs show positive effects on women’s employment and earnings.

One of the most successful training programs was implemented in Colombia between 2001 and 2005. The program included classroom instruction and business internships. For women who took the classroom courses and then interned in a firm, the probability of paid employment rose nearly seven percentage points, work per week rose almost three hours, and wages rose close to 20%—and as much as 33% for participants who were employed in the formal sector [2].

Colombia’s Youth in Action training program

The Youth in Action (Jóvenes en Acción) training program in Colombia in 2001–2005 increased female employment and earnings.

Source: Attanasio, O., A. Kugler, and C. Meghir. “Subsidizing vocational training for disadvantaged youth in Colombia: Evidence from a randomized trial.” American Economic Journal: Applied Economics 3:3 (2011): 188–220.

However, not all evaluations find beneficial effects for training programs. A training program in the Dominican Republic in 2004, for example, seems to have yielded minimal gains in female employment and earnings [3]. This less positive result might be due both to the design of the program (a two-month internship in the Dominican Republic program compared with three months in the Colombia program) and to the smaller sample size due to higher program drop-out rates. The school to work transition is particularly important.

Thus, one crucial ingredient for a successful training program is how well private training institutions match the skills transmitted in the classroom to the needs of firms. Through program-sponsored internships, firms can assess the skills of trainees without the commitment of a formal employment contract, and trainees can familiarize themselves with the job characteristics to determine whether the job is a good fit.

To promote the school to work transition, the World Bank launched the Adolescent Girls Initiative in 2008 through pilot programs in Afghanistan, Haiti, Jordan, Laos, Liberia, Nepal, Rwanda, and South Sudan. All the programs included job, business, and life training, and some also included a wage subsidy paid to employers. Results of a preliminary impact analysis of the training component in the program in Liberia show a 47% rise in employment and an 80% jump in average weekly income among project participants compared with young women in the control group [4].

The evidence for wage subsidies is not as positive. An evaluation study of the Adolescent Girls Initiative in Jordan found that wage subsidies are not very effective in increasing women’s employability [5]. The program gave vouchers to female college graduates to use to reimburse companies that hired them; the vouchers provided a subsidy equivalent to the minimum wage for up to six months. The vouchers increased the probability of employment, but mostly in informal jobs and only in the very short term. The positive effects disappeared four months after the voucher period ended [5].

Training and cash grants can increase women’s self-employment, especially when programs are more structured

Self-employed women in developing countries are generally engaged in very small-scale enterprises and follow rudimentary business practices. A number of programs organized by governments, NGOs, microfinance institutions, and the International Labour Organization (ILO) are intended to improve employment and income among self-employed women.

The largest program, global in scale, is the ILO’s Start and Improve Your Business, which trained 4.5 million women in about 100 countries from 2003 to 2010. Participants are taught how to develop new business projects (product selection, pricing, hiring, equipment and input purchases, financial planning) or expand their existing business (marketing, evaluating revenue and cost, accounting, and planning).

A randomized experiment conducted among some 1,200 women participating in the ILO program in urban Sri Lanka evaluated the impact of training classes plus a cash grant [6]. It considered both women running an enterprise and women out of the labor force who wanted to become entrepreneurs. Each sample was randomized into three groups: a control group, a group invited to attend training, and a group invited to receive training and a grant conditional on completing training. The women received the training and cash grants in September 2009, and the entrepreneurs and would-be entrepreneurs were followed up three times: after three to four months, seven to eight months, and 24 to 25 months. The study found that the training led to significant improvements in business practices; adding a cash grant made only a small short-term difference in the program’s impact. The impact on outcomes (profits, sales, and capital stock) was positive when training was accompanied by the cash grant, but the effects were limited and of short duration; the positive effect on profits, for example, disappeared after two years. In the case of women who were starting a business, the impact was more noticeable, if just as temporary: the combination of training and a cash grant increased the probability of starting a business by around 20–30 percentage points. After two years, however, those receiving the cash and training treatment were ten percentage points more likely to have opened a business at any point since the start of the program and nine percentage points more likely to have opened a business and then closed it: that is, the effect on business ownership was near zero. Training also improved the rate of profitability of new firms.

An experimental study in Ghana sought to distinguish between the effects of giving female enterprise owners a grant in cash or in kind (mostly items to sell, raw materials, and equipment such as sewing and knitting machines) [7]. As in Sri Lanka, the cash grants had a limited effect, a result the study attributed to a lack of oversight of how participants spent the money. In contrast, the in-kind transfers, whose selection was supervised by a project research assistant, significantly increased profits for a large group of firms.

In some cases, the success of a cash grant seems to depend on how that part of the program is implemented. A randomized trial in 2008 and 2010 of the Youth Opportunities Program, initiated in Uganda in 2006, yielded more positive evidence on the effectiveness of cash grants. The program was designed to help poor and unemployed young adults, most of them engaged in farming activities, become self-employed artisans. A third of the applicants were women. Only groups could apply for the cash grant. Business proposals had to include a period of training in non-agricultural skills by local artisans or small private training institutes. The need for group members to make collective decisions contributed to the success of the program. After four years, the incomes of participating women were 73% greater than the incomes of the control group; for men, the differential gain was 29% [8]. This study is a rare example of a longer-term (after four years) evaluation of a government program to ease the transition from agricultural employment to artisanal activities.

Uganda’s Youth Opportunities Program

Uganda’s Youth Opportunities Program of 2006–2010 was directed to men and women aged 16–35 living in rural villages. It had several important characteristics that may have contributed to its successes:

Source: Blattman, C., N. Fiala, and S. Martinez. “Generating skilled self-employment in developing countries: Experimental evidence from Uganda.” The Quarterly Journal of Economics 129:2 (2014): 697–752.

There are many other examples in developing countries of policy interventions directed specifically to groups, with mixed results. For example, self-help groups in rural India consist of 10–20 poor women who pool their savings and deposit them in a bank account to provide credit to women in need. Taking advantage of the networks established by these self-help groups in villages in Gujarat state, a policy experiment was set up to evaluate the effectiveness of vocational training provided to participants by the Self-Employed Women’s Association (SEWA). SEWA ran workshops on farming techniques intended for female farmers and also offered craft workshops. The impact evaluation estimated a 14 percentage point increase in the probability of non-farm employment for workshop participants compared with non-participants. Within two years, however, the study found no significant impact on income or on the probability of working outside the household [9].

Evidence on the effects of childcare programs is limited and varies by program and country

A substantial body of evidence shows that training programs are effective in easing young women’s access to paid employment. In most Latin American countries, the average age of women receiving employment training was 19–22, while women receiving business training were generally in their thirties and married with children. The employment effect of training for this older group is weaker, probably because women with children have household constraints that can be difficult to overcome.

For women with children, ensuring the well-being of their children is a vital concern. Since the time parents spend caring for their children is crucial for child development—and mothers still bear most of the burden of childcare—empowering women through work and ensuring child development may be conflicting objectives. In countries where well-paying jobs are available for women, child well-being may even rise as women join the labor market, because women may have more money to spend on meeting their children’s needs. However, in developing countries, the potential conflict between women’s work and child well-being is intensified by the large number of unpaid or badly paid jobs for women, pervasive child labor, and high drop-out rates among children. When mothers work, children—especially girls—take on more domestic chores, such as looking after younger siblings. A lack of care by mothers can also lead to poorer nutrition and health problems. A study for India shows that women in poorer households are more likely to work but their additional income does not seem sufficient to increase children’s school attendance [10].

For women with children, therefore, government programs that invest in training may need to bear the additional cost of ensuring good quality childcare. Some training programs have partially addressed this problem by easing access to childcare services or by providing childcare subsidies to mothers, as in the Youth in Action program in Colombia. However, while female employment is receiving increasing attention by governments in developing countries, interventions to improve the supply of childcare services lag behind. And most international and national institutions have separate programs for female employment and child well-being instead of focusing on integrating them.

A common arrangement for pre-school children is community daycare, which has been implemented in Latin American countries including Bolivia, Colombia, Peru, and Venezuela. For example, the program Hogares Communitarios de Bienestar Familiar, introduced in Colombia in 1986 to target nutrition and child development, established some 80,000 home-based nurseries that provided supplemental nutrition and psychosocial stimulation. A “community mother” (madre comunitaria) would cook and take care of up to 15 children aged 0–6 in her house. The program subsidized the nurseries, providing food and the pedagogic material, as well as a monthly payment to the madre comunitaria. An impact evaluation of this program has demonstrated that it significantly increased children’s height. The effects on female labor supply, however, are not clear [11].

The evidence on the effects of childcare programs on female labor supply in developing countries is limited and mixed, varying by program and country. In Argentina, for example, a large-scale program to construct pre-primary school facilities seems to have had a positive impact not only on pre-primary school attendance but also on maternal employment [12]. Despite the complexity of the relationship between maternal employment and child welfare and the ambiguity of some of the evidence, the importance of childcare services for women’s employment can hardly be denied.

Limitations and gaps

Although the agricultural sector accounts for more than 60% of female employment in South Asia and sub-Saharan Africa, most evaluations have examined programs in urban areas. The limited evidence on the impact of programs in rural areas remains an important gap in knowledge of the effectiveness of such programs.

A methodological challenge of impact evaluations is the difficulty of estimating whether the monetary benefits of a program exceed its cost. For example, in measuring women’s net earnings gains from participation in a program, one key question is whether gains are permanent or decline over the lifecycle. In the former case, gains would of course be higher. Another problem is the assumption about the number of years worked over the lifecycle. If women have interrupted careers because of children, the gains will be lower. The welfare loss of raising funds for the program through taxation should also be accounted for. All these aspects are difficult to quantify in one-time experiments.

While most programs are not gender-specific, women tend to participate and benefit more. Many programs, however, do not take account of women’s additional constraints to entering or staying in the labor market, including the burdens of childcare and domestic work.

Finally, more understanding is needed of the potential channels through which female employment policies affect child well-being so that policies can be designed that jointly consider mothers and children. Drawing on experience with conditional cash transfer programs, which make welfare payments conditional on meeting certain requirements, such as child nutrition and schooling, employment policies could incorporate incentives for improving the well-being of children as well as mothers.

Summary and policy advice

If one of the main objectives of training programs is to fill the skill gap, programs that offer a short course of two or three months, for a few hours a day, and without any follow-up support, are unlikely to improve the productivity of workers. This failure is confirmed by studies that find that training programs increase women’s employment probability but rarely lead to increased earnings over the longer term.

But despite some limitations, training programs, wage subsidies, and cash grants for starting a business have been shown to be viable tools for promoting female paid employment under some conditions. In Latin America, these programs have tried to increase wage employment in the formal sector, while in South Asia and Africa, the effort has been concentrated on training to ease the transition from unpaid work, mainly in agriculture, to starting a small business.

The experience of Latin American countries in training programs for wage employment shows that a successful design implies an intensive and focused training course and an efficient market of training providers that can match trainees with firms requiring the skills transferred by the program. Colombia’s experience teaches that this is an efficient way to improve young women’s employment and earnings.

In urban areas of South Asia and Africa, training programs that help women start a business seem to be effective in speeding women’s entry into self-employment. And while in-kind grants have been shown to improve the performance of existing businesses, training and cash grants have not had a positive impact.

In rural areas, simply providing training does not improve women’s labor market outcomes. Uganda’s Youth Opportunities Program teaches that women’s earnings can continue to increase over the long term if a program is more structured, first providing training and then grants to self-organized and self-controlled groups who submit promising business plans.

Job and business training policies alone, however, are not enough. Women need additional assistance to ease the burdens of family and home care. The demand for childcare services is growing rapidly in developing countries, and governments are taking important steps to provide publicly funded childcare. To guide policy development, new impact evaluations are needed that explicitly look at the effects of job and training programs on child well-being as well as on women’s earnings and labor market participation.


The author thanks an anonymous referee and the IZA World of Labor editors for many helpful suggestions on earlier drafts.

Competing interests

The IZA World of Labor project is committed to the IZA Guiding Principles of Research Integrity. The author declares to have observed these principles.

© Gianna Claudia Giannelli