I. Introduction

Cabo Verde is currently undergoing a demographic transition with social and economic implications that will be defining for its development path going forward. Other countries that are also in the advanced stages of their demographic transitions in Africa include Mauritius, South Africa, Botswana and the Seychelles (Drummond et al. 2014).

This study analyzes Cabo Verde’s demographic transition from the perspective of gender equality. It also discusses some policy options that may contribute to attaining the potential economic and social dividends from the transition. With the drop in the fertility and mortality rates in the early stages of a demographic transition, the ratio of people younger than 15 and older than 64 to people between the ages of 15 and 64 decreases. This combination known as youth bulge due to the shape of the demographic pyramid may result in demographic dividends derived from changes in labor supply, savings, and human capital.

More specifically, in a demographic transition, labor supply increases as people in the relatively expanding 15-64 age group are more likely to be working than people in other relatively shrinking age groups (Bloom et al 2003). In addition, labor supply can also increase with the increase in the number of women in the workforce resulting from smaller family sizes being less taxing on women’s time. Savings tend to increase during a demographic transition leading to a youth bulge because working-age people tend to consume less than what they produce compared to the young and the old. Moreover, a smaller family size can also result in higher savings: resources are reallocated from investing in children to preparing for retirement. Finally, Bloom et al (2003) argues that the impact from a demographic transition on the level of investment in human capital is perhaps “the most significant and far-reaching,” with larger amount of resources being “invested” in education and health of a relatively smaller number of children.

Bloom and Finlay’s (2008) estimates of demographic dividends suggest that, between 1965 and 2005, demographic change contributed around 40 percent of the economic growth in South East Asia, 29 percent in the Republic of South Korea, and 26 percent in China.

However, whether demographic dividends are in fact realized depends on an enabling policy environment and the broader social, economic, and cultural context. For instance, the increase in productivity from the increase in the proportion of the population that is of working-age will depend on the absorption capacity of the labor market. This capacity rests on the ability of a growing economy to create job opportunities. Likewise, the quality of jobs – in terms of varying levels of informality, stability of contractual employment practices, wages levels, and fit between education and work - also matter a great deal for achieving demographic dividends. Thus, the International Labor Organization (ILO) estimates that nearly half the youth of developing economies is not achieving their full economic potential because of the lack of decent jobs. This, according to the ILO, represents a key impediment to sustainable development (ILO 2015). Gender equality has also an impact on demographic dividends because of its impact on the labor supply and/or on the duration of the demographic transition. Closing the gender gap in education by promoting female education, for instance, is often associated with lower fertility – i.e. narrower population pyramid bases - and diminishing family sizes (Kane 2004).

As Cabo Verde’s demographic transition slows, the size and potential of demographic dividends post-2015 will drop sharply due to “mechanical” reasons – i.e., the slowdown of the increase in the proportion of working-age population. Henceforth, it will largely be determined by the growth in productivity and the growth in the effective participation rate of the working-age population in the labor market. In this context, efforts towards closing gender gaps, in combination with efforts towards making economic growth more inclusive, will be increasingly important in enhancing otherwise diminished growth dynamics, reducing poverty, and improving lives of all, women and men. This study estimates, for instance, that closing the gender gap in labor force participation could have a direct impact on GDP of as much as 12.2 percent in Cabo Verde.

Besides being a fundamental human right and a development goal on its own, gender equality can multiply and extend the potential economic and social gains of a demographic transition. That is to say that ignoring gender inequality prevents an economy from growing at its full potential.

Indeed, referring to the relatively low level of female participation in its workforce, Goldman Sachs has raised the alarm over the “underutilization of assets” of Japan. Closing the gender gap in employment in Japan could boost the country’s GDP by close to 13 percent (Matsui et al 2014). In fact, since 1999, researchers at Goldman Sachs have advocated for the adoption of “womenomics” as a national priority to solve the demographic crisis of Japan. This includes boosting female employment with the implementation of supporting policies such as expansion of daycare and nursing care services, flexible work arrangements, and reevaluation of compensation and evaluation systems (Matsui et al 2014, 2010, and 1999; Lawson 2008; Daly 2007). Steinberg and Nakane (2012) emphasize the role of policies in reducing the gender gap in career positions and in providing better support for working mothers in Japan.

Daly (2007) estimates that increasing female participation and closing the gap between male and female employment in the Eurozone accounts for 0.4 pp of the region’s 2.1 percent trend growth between 1995 and 2005. Women entering the labor force represented half the rise in the Eurozone’s total employment rate between 1995 and 2005 and that policies encouraging and supporting that shift constituted “the single-biggest driver of Eurozone’s labour market success, much more so than ‘conventional’ labour market reforms.”³

In contrast to Japan and Europe, Cabo Verde is at a much earlier stage of its demographic transition. The adoption of “womenomics” would therefore only reinforce and expand the potential of the demographic dividend and the cumulative effects on long-term growth and poverty reduction. Indeed, the fact that 33 percent of households headed by women are poor (among men, 25.7 percent are poor) and that the share of poor households headed by women increased from 51 percent in 2001/2002 to 56.3 percent in 2007 (INE 2004 and 2007) indicate that the gender gap in poverty has widened and that Cabo Verde’s sustainable poverty reduction strategies must increasingly place women at their center.

Cabo Verde’s has made significant advances in promoting gender equality in some areas but there is still a lot of catching up to do on the economic participation front. This is reflected in its being ranked 50^th out of 145 countries in the World Economic Forum’s Global Gender Gap Index (2015). The same index broken down by categories places Cabo Verde in first place, next to Finland in terms of health and survival, but in 115^th in terms of economic participation and opportunity.

A case study of Cabo Verde generates important lessons for other countries in Africa that are in the earlier stages of their respective demographic transitions and are looking for ways to manage the potential dividends resulting from these transitions. As many countries in Africa are facing increasing rates of youth unemployment and increasing gaps in gender equality, these are increasingly urgent issues (United Nations Economic Commission for Africa, African Union, African Development Bank and United Nations Development Programme 2015).

Section 2 describes the current demographic transition in Cabo Verde. Sections 3 measures the potential demographic dividend in Cabo Verde and section 4 estimates the potential impact on GDP of closing the gender gap in labor force participation. Section 5 discusses labor force participation and time use and section 6 looks at employment conditions from a gender perspective. Finally, section 7 presents a conclusion and a discussion of policy options and recommendations.

II. The Demographic Transition in Cabo Verde

Cabo Verde’s National Institute of Statistics (Portuguese abbreviation, INE) estimates that the share of the population between 15 and 64 years old will rise from 62 percent in 2000 to 67.5 percent in 2030. This increase will be accompanied by a significant decline in the proportion of the population under the age of 15 from 32 percent to 24 percent, and an increase in the population older than 64 from 6 percent to 8.5 percent.

INE’s estimations indicate further that most of the growth in the share of people between 15 and 64 has already occurred between 2000 and 2015. With the projections that this group’s growth will nearly stagnate after 2025, the youth bulge is expected to peak at around 2030.

This demographic shift implies a significant increase in the proportion of people of working age and a decrease in the proportion of dependents, in particular of dependents younger than 15. Indeed, in the mid-1970s, shortly after independence, it is estimated that there was one person of working age (15 to 64) for each dependent (younger than 15 or older than 64); this ratio was below the average in sub-Saharan Africa. It is projected that this ratio has doubled by 2015, i.e., 2 people at working age for each dependent. This level is comparable to the average of middle-income country, and is almost twice as large as the average of the Sub-Saharan African region (Figure 1).

View Full Size

Number of People at Working Age by Number of Dependents

Citation: IMF Working Papers 2016, 169; 10.5089/9781475524246.001.A001

Source: Own calculation based on WDI (2015) and INE (2012)Note: (P) indicates projections. Working age is from 15 to 64 years old and dependents are younger than 15 or older than 64 years old.

• Download Figure
• Download figure as PowerPoint slide

With the increasing number of potential workers per dependent, Cabo Verde’s economy has had and will continue to have the opportunity to benefit from the demographic bonus. This is based on the premise that adults of working age tend to produce more through their labor than they consume, while children and older people tend to consume more than they produce. The exact age range of the economic lifecycle that defines periods of net production and net consumption may vary depending on countries and generations. For this analysis, this study considers the age range for when production is higher than consumption to be the traditional 15 to 64 years old.

II. Sources of Growth

D. Sources of income per capita growth, 1990-2030

Between 1990 and 2013, income per capita in Cabo Verde grew at an average of 5.9 percent per year according to the WDI (2015). The breakdown of growth indicates that, between 1990 and 2013, productivity was the largest source of growth in income per capita. In fact, growth of income per capita in the labor force contributed, on average, about 75 percent, or 4.4 p.p. (percentage points), of total growth of income per capita; growth in the labor force participation rate contributed 6 percent, or 0.33 p.p.; and growth in the share of the population between 15 and 64 years old contributed 20 percent, or 1.2 p.p. (Table 1, column 1).

Table 1.

Sources of Growth in Income Per Capita, 1990-2030

Source: Own calculation based on WDI (2015) and demographic projections of INE (2012) Note: (P) indicates projections.

Table 1.

Sources of Growth in Income Per Capita, 1990-2030

Average Annual Growth
		(1) 1990-2013	(2) 1990-2000	(3) 2001-2013	(4) 2014-2030 (P)
GDP per capita		5.9%	8.5%	3.9%	−
Breakdown of GDP per Capita Growth:
	GDP per capita in the labor force	4.4%	7.5%	2.0%	−
	Labor Force Participation Rate	0.3%	0.4%	0.3%	−
	Share of Population 15-64 years old	1.2%	0.6%	1.6%	0.2%

Source: Own calculation based on WDI (2015) and demographic projections of INE (2012) Note: (P) indicates projections.

View Table

Table 1.

Sources of Growth in Income Per Capita, 1990-2030

Average Annual Growth
		(1) 1990-2013	(2) 1990-2000	(3) 2001-2013	(4) 2014-2030 (P)
GDP per capita		5.9%	8.5%	3.9%	−
Breakdown of GDP per Capita Growth:
	GDP per capita in the labor force	4.4%	7.5%	2.0%	−
	Labor Force Participation Rate	0.3%	0.4%	0.3%	−
	Share of Population 15-64 years old	1.2%	0.6%	1.6%	0.2%

Source: Own calculation based on WDI (2015) and demographic projections of INE (2012) Note: (P) indicates projections.

Cabo Verde’s income per capita grew on average 8.5 percent from 1990-2000 and 3.9 percent during the period of 2001-2013. The slowdown in the latter period is partly due to the financial crisis of 2007/2008 and its impact on world economic growth in subsequent years.

Productivity appears to have been the main engine of growth from 1990 to 2000, contributing on average 88 percent (or 7.5 p.p.) of the growth of total income per capita. In the period from 2001 to 2003, the growth in the share of the population aged 15 and 64 contributed more strongly to growth in income per capita. While the contribution of the growth in the labor participation rate appears to remain stable over the whole sample, the contribution from the growth in the share of the population between 15 and 64 to total growth of income per capita climbed from 7.6 percent in 1990-2000 to 41.4 percent in 2001-2013. This increased contribution is due to the demographic change that intensifies in the latter period with the average growth in the share of population between 15 and 64 jumping to 1.6 percent from 0.6 percent in the previous period (Table 1, columns 2 and 3).

INE estimates and projects a slowdown in the average growth of the share of the population between 15 and 64 to 0.2 percent between 2014 and 2030 (Table, column 4). Assuming that the 2001-2013 growth rates of income per capita in the labor force and of the labor force participation rate continues for 2014-2030, the country’s growth of income per capita would fall from an average of 3.9 percent in 2001-2013 to 2.5 percent in 2014-2030. That means that a drop of 1.4 p.p. in growth would be directly related to the slowdown in the demographic transition.

This analysis of the demographic projections - with all its caveats - leads to an important conclusion: with the demographic transition slowing, the size and potential of demographic dividends post-2015 will be determined largely by the growth of productivity and the growth in the effective participation rate of the working-age population in the labor market.

Policies that are able to target productivity and encourage overall labor force participation effectively will have a greater likelihood of successfully extending and even amplifying the gains associated with the demographic transition. In particular, promoting gender equality and increasing the labor force participation of women can have a significant impact on GDP growth.

III. Impact on GDP from Closing the Gender Gap in the Labor Force

In 2014, labor force participation rate of women was 51.2 percent of the total female working-age population in Cabo Verde; for men, it was 65 percent of the total male working-age population. Note that national employment figures from INE surveys define the working-age population as 15 or older (as opposed to 15 to 64); thus, for the calculations in this section, I follow INE’s definition.

Raising the labor force participation rate of women to match that of men would imply increasing the total labor force participation rate by 12.2 percent or a total of 26,428 workers. The impact on GDP would be substantial: assuming everything else is kept constant in equations 1 and 2, it would follow that closing the gap in the labor force participation rate would increase the labor force participation rate and thus the GDP by 12.2 percent.

The actual full impact of closing the gender gap on GDP would depend on several factors, including the absorption capacity of the economy and the labor productivity of the new entrants.

One must also consider that as women enter the labor force, their children and dependents would not have anyone to look after them during the hours women are out of the house. The role of women is central in home-making in Cabo Verde: according to the Time Use Survey (INE 2012), close to 90 percent of women report spending an average of almost 60 hours a week doing unpaid work in the home. One third of that is spent on taking care of children and dependents; the majority of hours spent taking care of children is dedicated to children between the ages of 0 and 3.

Supportive policies for working mothers, including day cares, are important to encourage female labor force participation but they are also central to preventing the potentially negative social and economic impact of unattended young children and dependents. I will discuss this point further in section 7.

Aguirre et al, (2012) attempt to quantify some dampening factors on growth when measuring the economic contribution of new entrants in the labor force. They consider two factors: (1) a “productivity drag” produced by a gap in productivity between incumbents and new entrants and (2) the possibility that new entrants could be pushed into part-time jobs because of family reasons or a limited absorption capacity of the economy. They assumed that factor (1) could dampen GDP growth by around 30 percent and factor (2) by another 13 percent. Even after taking into account these dampening factors, at 7.4 percent the total boost in growth from closing the gender gap would still be substantial.

These results serve to underline that women tend to be one of the most “underutilized assets” in the economy. However, labor force participation only tells part of the story; it is also necessary to understand where the major gaps in labor force participation lie and if there are gender gaps in employment conditions and more broadly that prevent women from participating fully in the economy.

IV. Labor Force Participation and Time Use

According to the most recent national survey (INE 2015), the gap between the female and male participation rate in the labor force is much larger in rural areas, while it is also observed in urban areas.

In 2014, the average labor force participation rate among urban residents was 59.1 percent for women against 69.8 percent for men. In rural areas, these figures averaged 35.7 percent and 55.5 percent respectively (Table 2). The gender gap is thus almost twice as large in rural areas, where labor force participation for both men and women is also much lower than in urban areas.

Table 2.

Labor Force Participation by Age and Gender, 2014

Source: INE 2015

Table 2.

Labor Force Participation by Age and Gender, 2014

	Urban			Rural			Total
Age	Men	Women	Total	Men	Women	Total	Men	Women	Total
15-19	13.7	12.4	13.1	19	9.8	14.7	15.9	11.3	13.7
20-24	67.7	58.8	63.2	56.7	30.5	44.9	63.7	49.8	57
25-29	83.6	73.4	78.6	71.5	47.2	60.4	80.1	66.6	73.6
30-34	88.1	80.9	84.5	77.7	60.8	69.1	85.1	75	80
35-39	90.3	83.8	86.9	82.3	59.3	70.5	88.3	77.9	82.9
40-44	91.9	80.6	85.8	78.4	57	67.3	87.7	73.7	80.2
45-49	89.8	79.4	84.9	78.2	54.8	65.7	86.3	70.5	78.6
50-54	79	69.5	74.3	78.1	46.5	60.7	78.8	61.5	69.8
55-59	78.5	46.1	60.9	61.4	44.8	51.4	73.1	45.6	57.6
60-64	57.2	29.1	41	52.1	26.7	34.7	55.8	28.2	38.9
65+	19.8	7.2	12.3	15.3	5.2	9.1	17.8	6.3	10.9
Total	69.8	59.1	64.3	55.5	35.7	45.3	65	51.2	57.9

Source: INE 2015

View Table

Table 2.

Labor Force Participation by Age and Gender, 2014

	Urban			Rural			Total
Age	Men	Women	Total	Men	Women	Total	Men	Women	Total
15-19	13.7	12.4	13.1	19	9.8	14.7	15.9	11.3	13.7
20-24	67.7	58.8	63.2	56.7	30.5	44.9	63.7	49.8	57
25-29	83.6	73.4	78.6	71.5	47.2	60.4	80.1	66.6	73.6
30-34	88.1	80.9	84.5	77.7	60.8	69.1	85.1	75	80
35-39	90.3	83.8	86.9	82.3	59.3	70.5	88.3	77.9	82.9
40-44	91.9	80.6	85.8	78.4	57	67.3	87.7	73.7	80.2
45-49	89.8	79.4	84.9	78.2	54.8	65.7	86.3	70.5	78.6
50-54	79	69.5	74.3	78.1	46.5	60.7	78.8	61.5	69.8
55-59	78.5	46.1	60.9	61.4	44.8	51.4	73.1	45.6	57.6
60-64	57.2	29.1	41	52.1	26.7	34.7	55.8	28.2	38.9
65+	19.8	7.2	12.3	15.3	5.2	9.1	17.8	6.3	10.9
Total	69.8	59.1	64.3	55.5	35.7	45.3	65	51.2	57.9

Source: INE 2015

Looking at the figures by age provides a proxy for how the labor force participation rates behave over the average economic lifecycle of a typical person in Cabo Verde. The country has changed rapidly in the last 10 to 15 years, and this is and will continue to be reflected in significant cohort effects. With that limitation in mind, the analysis of the figures suggests two additional facts about labor force participation: (1) women spend fewer years than men in the labor force; and (2) the patterns over the lifecycle of the evolution of the gender gap in rural and urban settings are different.

As Table 2 and Figure 2 show, the overall gender gap - urban and rural combined - increases over the lifecycle up to the age group 60 to 64; after that, it drops sharply. Even at the peak of female participation, at 30 to 34 (urban and rural combined), the gender gap persists.

View Full Size

Labor Force Participation by Age Group, Gender and Urban and Rural Residence

Citation: IMF Working Papers 2016, 169; 10.5089/9781475524246.001.A001

Source: INE 2015

• Download Figure
• Download figure as PowerPoint slide

The charts in figure 2 suggest that women tend to enter the labor force later, between the age of 25 and 29, and leave the labor force earlier, between the age of 55 and 59, than man.⁵ This implies a much shorter period of labor-force participation for women: the average work-life of a man is around 40 to 44 years against an average of around 30 to 34 years for a woman.⁶ It is important here to note that this does not imply that women work less than man. In fact, given what we know about very considerable unremunerated work in the house, they may often work more (although this is neither reflected in labor force participation nor the GDP).

There are also differences in rural and urban areas that are worth mentioning. The gender gap in urban areas exists and tends to be fairly stable from the age of 20 to 54, after which it grows very sharply, mostly driven by the drop in the labor force participation of women. In rural areas, the gender gap appears to widen much earlier in the lifecycle, after the age of 34. These differences could be the result of patterns of migration, very common among Cabo Verdeans looking for better opportunities,⁷ but other structural issues.

Some of these differences in the gender gap between rural and urban areas are driven by differences between the two cohorts of women. Women in rural areas tend to enter the labor force later and leave it earlier than women in urban areas. The average work-life estimates are 30 years on for urban women and 15 years for rural women. The peak in the labor force participation rate is 60.8 percent for rural women at the age of 30 and 34, and 83.8 percent for urban women at the age group 35 to 39.

Policies aiming at encouraging higher female labor force participation need to pay particular attention to the large gender gap and the low labor force participation of women in rural areas. And while the process of urbanization has likely been very important to the increased participation of women in the labor force, more can be done in terms of increasing the average work life of women residing in urban areas. A later – relative to man – entry in the labor force could imply more limited career opportunities for women, and a diminished ability to accumulate on-job skills, which has a potential impact on wage differentials between men and women. Such a gender gap is important for growth and poverty reduction: in Cabo Verde, more than one in four people are considered poor and about one in three households headed by a woman is poor. Breaking cycles of poverty will inevitable require breaking women’s poverty.

As previously noted, the smaller female labor force participation rate does not mean that women work less than man or contribute less to the economic welfare of the country. Indeed, the Time Use Survey (INE 2012) indicates that women spend on average 1.6 times more time than man doing unpaid work, including household tasks, taking care of dependents (including child-rearing), supporting other families, and volunteer work. Much of the gender gap is driven by differentials in time spent on household tasks and taking care of dependents. Women in Cabo Verde spend on average about three and a half hours more than men on unpaid work each day. In addition, more women report being engaged in unpaid work (90 percent) when compared to men (73 percent). The survey also indicates that both the number of hours dedicated weekly to unpaid work by women and the gap in the number of hours between men and women increases with decreasing standards of living, levels of education, and an increasing number of dependents.

The gender gap in the amount of time dedicated to unpaid work also increases with age, up to the age of 64: the share of women reporting spending time doing unpaid work is quite high over the entire lifecycle, increasing marginally from 87.6 percent for the age bracket of 10 to 14 to 93.9 percent for the age bracket of 45 to 64 (Figure 3). For men, the share clearly declines with age, from 84.6 percent (10-14) to 64.2 percent (45-64). Such a widening gender gap also exists with respect to the number of hours spent weekly doing unpaid work. At the peak gap (age 45-64), women report spending time doing unpaid work 1.5 times more frequently than men. The 45 to 64 female age bracket spends on average about 5h and 10 minutes more than men on unpaid work each day (Figure 4).

View Full Size

Share of Respondents Who Report Time Spent Doing Unpaid Work, by Gender and Age Group

Citation: IMF Working Papers 2016, 169; 10.5089/9781475524246.001.A001

Source: INE 2012

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Average Minutes Dedicated to Unpaid Work Weekly, by Gender and Age Group

Citation: IMF Working Papers 2016, 169; 10.5089/9781475524246.001.A001

Source: INE 2012

• Download Figure
• Download figure as PowerPoint slide

While it is not possible to draw a clear causal connection between female labor force participation and hours dedicated to unpaid work by age group because of data limitation, the Census 2010 reports that 22.5 percent of female respondents – against 4 percent of male respondents – report family tasks as a limiting factor in their access to the labor force (Cabo Verde’s Millennium Development Goals (MDG) National Report 2014). The type of unpaid work probably also matters for the ability to work and participate fully in the labor force. Taking care of infants when no day care services are accessible and/or affordable or attending to dependents who need special attention are much less flexible than other home-making tasks. Another possible underlying reason for the lack of a clear causal connection is that family tasks may drive women to work on more flexible arrangements, which tend to be associated with informality or vulnerability. With the available data, one cannot say if unpaid work causes lower labor force participation and/or poorer employment conditions but the simple fact that women spend significantly more hours per day than men doing unpaid work at all ages strongly suggests that unpaid work constitutes an important cause of the gap between women and men in terms of their ability to realize their full employment potential in the formal labor market.

V. Labor Market Characteristics

The effective participation of both women and men in the labor force depends on their ability to get employment and have decent jobs. The larger the effective participation is, the larger its impact on long-term GDP growth.

Currently, 49 percent of the total working-age population -in and out of the labor force- is employed (INE 2015). This rate is the smallest, 19.1 percent, among women aged 15 to 24 and highest, 74.1 percent, among the men aged 35 to 64.

Unemployment, with a national average of 15.8 percent of the labor force, affects mostly young people in urban areas among whom the unemployment rate is as high as 45.9 percent (15 to 19 years). While unemployment is overall higher among men (16.3 percent) than women (15.2 percent), the highest rates of unemployment is observed among women between the ages of 15 and 19 (48.9 percent) and 20 and 24 (38.1 percent) residing in rural areas. The largest gender gaps in the unemployment rate are observed in the younger age groups, where the levels of unemployment are also the highest. Indeed, unemployment among women can be twice as high as among men (see Figure 5).

View Full Size

Gender Gaps in Unemployment Rate and Unemployment Rates for Women, Urban and Rural

Citation: IMF Working Papers 2016, 169; 10.5089/9781475524246.001.A001

Source: INE 2015

• Download Figure
• Download figure as PowerPoint slide

High rates of unemployment increase the likelihood of underemployment, which is often characterized by low wages and poor conditions and tends to be associated with informal sector activities. The disproportionately high level of unemployment among young women therefore tends to create distortions early on in life that can amplify gender inequalities over the working life in terms of job opportunities in the formal market but also in terms of wages and career opportunities. The important role of women as heads of households greatly increases the risk of these distortions being transmitted onto future generations, increasing the risk of low social mobility and persistent cycles of poverty.

Informality is high in Cabo Verde: the large majority, close to 59 percent, of the employed reports having no labor contract (INE 2015). Evidence suggests that women are the majority of those working in informality, which suggests that the decline in the gender gap in employment with age could be masking gender disparities in other important aspects of employment.

Among employed women, close to one fourth works in retail and another 13 percent works in family homes. Both sectors have very high levels of informality marked by the absence of a contract. Indeed, the great majority of people working in family homes, 92.2 percent, reports having no labor contract, with only 3.1 percent reporting having any kind of contract. In retail, 77.8 percent of the respondents report having no labor contracts. Workers in agriculture, which employs another 8.5 percent of employed women, report having no contracts 95.4 percent of the time.

Observable “underemployment,” characterized by working less than 40 hours but being willing to work more hours (INE 2015), affects as much as 30.1 percent of the population; this rate is particular high in rural areas (48.2 percent) and among women (33.9 percent). The highest rate is among women residing in rural areas where the average underemployment rate reaches 57.9 percent; in some counties it is as high as 80.1 percent (e.g. women residing in S. Miguel).

Finally, Cabo Verde’s 2014 MDG Report states that women tend to face greater job vulnerabilities – defined as self-employment and being family-workers – than men. In 2011, 55.4 percent of women had vulnerable jobs against 22.4 percent of men.

Broadly, the data show that gender disparities affect women’s labor force participation but also women’s employment conditions. Policies aiming at promoting gender equality therefore need to go beyond closing the gender gap in labor force participation rates and also focus on gender gaps in employment conditions.

VI. Conclusion and Policy Implications

There is much potential for increasing GDP growth in the coming years from the change in demographics that Cabo Verde is experiencing. GDP growth will come from the increase in the share of the working-age population. But a greater potential resides in a possible increase in the labor force participation of young people and women.

In the case of women, who are the focus of this study, the data shows that the labor force participation rate tends to be smaller than that of men, and that closing the gender gap in labor force participation could raise the workforce by more than 26,000 people and boost GDP growth by as much as 12.2 percent.

Closing other employment-related gender gaps also has the potential to raise productivity and GDP growth. There is much room for improvement in labor productivity among women through enhancing formal employment opportunities for them. Indeed, the data shows that women have a shorter working-life span than man – i.e., they enter the labor force later and leave it earlier; young women face the highest levels of unemployment; and women face higher levels of informality, underemployment, and job vulnerability than men.

The reasons underlying these many gaps need to be further explored, but the evidence suggests that taking care of family members and the home plays an important role in explaining many women’s late entry in the labor force. This and the high levels of unemployment at a young age women face in turn increase the likelihood of informality, underemployment, and job vulnerability, creating a vicious cycle of poverty and poverty traps.

According to Elborg-Woytek et al. (2013), improving women’s effective participation in the labor market entails well-designed and integrated sets of policies – from fiscal policies, including the use of tax credits or benefits to low-wage earners and expenditure measures targeting child support, to policies aiming at increasing the demand for women’s labor and promoting the use of flexible work arrangements. The overall conclusion, however, is that no single policy alone can be effective in closing all gender gaps because of the multiple, interconnected social, economic, and cultural aspects that underlie these gaps.

For instance, Cabo Verde’s 2014 MDG report shows that unemployment tends to be higher among women than men with a post-secondary education, which implies that the education level of girls and women does not automatically translate into equal access to job opportunities. Nonetheless, education remains one of the pillars of efforts to promote gender equality in terms of higher wages, a greater likelihood of working outside the home, lower fertility rates, reduced maternal and child mortality, and better health and education, including health and education of future generations (Lawson 2008).

Cabo Verde’s MDG report suggests further that part of the difficulty in dealing with this and other gender-disparities is related to the lack of a systematic gender approach in planning and managing of public policies and programs. Indeed, gender-based inequalities are among the obstacles in the way of achieving almost all MDGs.

In tackling labor force participation gaps and their strong link to unpaid work and home-making tasks in Cabo Verde, the 2014 MDG report urges to better integrate care in social policies and reframe care services as a societal responsibility rather than exclusively that of the care recipient’s family (and typically their female members). The report further emphasizes the need for policies that promote shared responsibility in the care of dependents and the house, better paid and unpaid work balance, and a higher social value associated with unpaid work.

Support to child and elderly care plays an important role in female labor participation. Elborg-Woytek et al. (2013) cite successful reforms in Japan, the United States, and South Africa. Community child-care centers that target low-income neighborhoods have been found to be effective in increasing maternal employment in a number of countries (World Bank 2012). In studying Chile’s Crece Contigo Childcare Program, however, Manley and Vasquez (2013) find evidence that the childcare program has not had a clear effect on women’s rates of participation. They suggest many reasons for the apparent ineffectiveness of the program, including cultural and traditional reasons and the need for a broader range of social safety-net policies.

Beyond care services, Cabo Verde’s 2014 MDG report also emphasizes the need for policies that are gender-based and aim at improving employment opportunities for women such as professional training integrated with policy initiatives for economic diversification including in agribusiness. The report also points to the importance of data disaggregated by gender for designing, monitoring, and evaluating policies.

Promoting gender equality also entails policies that support changes in culture and social behavior. For instance, 1,702 out of 2,693 children not enrolled in primary school were girls in 2010 (MDG report 2014). The 2014 MDG report suggests that, although the reasons need to be better understood, the gender of the child may well determine a family’s decision to have the child help the family as opposed to sending the child off to school.

Girls’ education is a problem faced by many countries. Kane (2004) discusses the question of girls’ education and finds that, especially among poor families, the opportunity cost of sending girls to school is often considered high and increasing as the quality of the education and expected returns from studying declines. The inability of young women to find jobs only exacerbates that return calculation. Costs associated with fees, transportation, uniforms, and books play also an important role; the demand for girls’ education in some cases is very sensitive to the distance to school because of their need to work at home. Lawson (2008) discusses girls’ versus boys’ education sensitivity to household income, citing a study of India that estimates that an increase of 1 percent in per capita household income increases the likelihood of girls’ enrollment in middle school by 4 pp, compared to 1pp for boys. Cultural reasons are in some cases as important as economic reasons as demonstrated by research conducted in Guinea and Ethiopia (Kane 2004).

Considering the evidence presented in this study, Cabo Verde would gain from adopting an integrated set of policies that especially target the reasons underlying late entry in labor force and higher levels of unemployment among young women. Particular attention should be given to women in rural areas and the underlying causes for the larger gender gap and especially low labor force participation. Two main set of policies could be effective: one that would focus on decreasing the time spent by women doing unpaid work and one that would focus on increasing the employability and hiring of women.

The first set of policies focusing on freeing time for formal employment could include expanding the capacity and affordability of care services, policies and awareness campaigns to break traditional views of gender-roles, targeted cash transfer policies, and other development initiatives such as increased access to water and electricity. In fact, access to running water has played a key role in the development of large economies such as the United States precisely because it has freed up especially female labor. Gordon’s (2012) analysis of long-term growth in the US includes an assessment of the importance of living standards, a measure that includes running water in the house, as a contributor to what he calls the second industrial revolution beginning in 1870: “Every drop of water for laundry, cooking, and indoor chamber pots had to be hauled in by the housewife, and wastewater hauled out. The average North Carolina housewife in 1885 had to walk 148 miles per year while carrying 35 tons of water…. There was no more important event that liberated women than the invention of running water and indoor plumbing, which happened in urban America between 1890 and 1930.” (Gordon 2012, 7)

Limited fiscal space, which is a severely binding constraint especially for developing nations, can be circumvented when possible through greater private sector involvement. Matsui (2014), for instance, emphasizes the importance of deregulation in promoting the participation of the private sector in daycare services in Japan, where fiscal constraints decreased the ability of the state to provide such services. In fact, the promotion of businesses offering care services in Cabo Verde would have the potential to create job opportunities and absorb a large number of young people, especially women.

The second set of policies, which would focus on employability of women and on increasing the demand for female labor, may include policies that promote flexible work arrangements, female entrepreneurship, specific training, access to financing, awareness campaigns to increase gender diversity in businesses, and the removal of discriminatory practices and red-tape. Indeed, Gonzales et al. (2015) find legal restrictions and regulations that establish discriminatory gender differences against women to be a significant deterrent to female labor participation across a variety of countries.

Significant changes in the dynamics of female labor participation, possibly in tandem with cash-transfer initiatives targeted to low income households and improvements in the quality of education and professional training for women, would likely increase the expected return from schooling for girls, especially poorer girls. This could have a positive impact on enrollment ratios of girls, especially in rural areas.

Full participation in the labor market also depends on the ability of the economy to create new jobs and absorb new entrants into the labor market. The evidence suggests that Cabo Verde’s recent history of strong growth has had limited impact in terms of job creation at national level. The tourism-related hotel and restaurant sector, for instance, which has grown 3.5 times faster than the overall economy since 2000, employed only 7.3 percent of the population in the labor force in 2014. Going forward, growth strategies will need to be particularly inclusive if they are to maximize the potential benefits of the demographic transition.

Overall, this study concludes that limited female participation in the labor market constitutes a significant loss of opportunity for Cabo Verde to grow, to improve people’s lives, and to lift men as well as women out of poverty. With the aging of the population, the path that will lead to the goal of Cabo Verde turning into a middle-income country by 2030 will inevitably and in fact increasingly, include promoting gender equality on all economic, social, and political fronts.