81682




PROGRAMS PROMOTING YOUNG WOMEN’S EMPLOYMENT:
                    WHAT WORKS?




                       Elizabeth Katz
              Associate Professor of Economics
                 University of San Francisco


                       October 2008




            Launch of the Adolescent Girls Initiative
       October 10, 2008 ▪ 11:00am – 5:45pm ▪ MC13-121
                       The World Bank
                        1818 H St, NW.
                    Washington, D.C. 20433
2
                                Table of Contents


Executive Summary
1. Introduction
2. Issues: Gender and Youth Employment in Developing Countries
   2.1. International Labor Market Trends for Young Women
      2.1.1. Labor Force Participation
      2.1.2. Unemployment
      2.1.3. Idleness and Joblessness
   2.2. Gender-specific Challenges to Youth Employment
3. Programs: Which Youth Employment Initiatives Work Best for Young Women?
   3.1. Youth Employment Inventory
   3.2. Review of Selected “Best Practice” Programs
      3.2.1. The “Latin American” Model
      3.2.2. Adolescent Livelihoods Programs
4. Lessons: What Can New Youth Employment Initiatives Learn About How Best
   to Meet the Needs and Challenges Specific to Young Women?
   4.1. Gender Issues in Vocational Education and Training Programs
      4.1.1. Gender-sensitive, Participatory Needs Assessment
      4.1.2. Targeting and Outreach
      4.1.3. Adapting Curricula to the Needs and Skills of Young Women
      4.1.4. Helping Young Women Balance Domestic Responsibilities with Job
             Training
      4.1.5. Gender-sensitive Labor Market and Social Support Services
      4.1.6. Gender-disaggregated Impact Assessment
   4.2. Lessons Learned from Adolescent Livelihoods Programs
   4.3. Minimalist Approach or Integrated Adolescent Services?
5. Conclusions and Recommendations
References




                                          3
                                 Executive Summary


The World Bank is launching an initiative aimed at addressing the economic needs
of adolescent girls and young women in poor or post-conflict countries. As a part of
background research for this initiative, this paper undertakes a review of existing
policies and programs designed to promote labor force participation of young women
in developing countries. The goal of this paper is to unpack and assess what
elements of program design are essential to promoting young women’s transition to
the labor market.


For developing countries as a whole, 34% of young women are estimated to be
“jobless” – unemployed, out of the labor force, and not in school. There are several
gender-specific supply- and demand-side determinants of low employment rates and
earnings among young women, including girls’ disproportionate share of unpaid
domestic labor, lack of productive skills and contacts to help with a job search,
limited life skills, opposition from family or male partners, and employer preference
for hiring young men over young women in a range of industries and occupations.

In Latin America, a number of youth employment programs contract with
decentralized training entities to organize and offer vocational courses in which
beneficiaries can enroll. The courses, which contain both classroom and on-the-job
experiences, match local firms’ needs with the content of the training curriculum.
These programs’ documented success in promoting young women’s employment can
be attributed to the fact that they are exceptionally well-targeted, demand-driven,
and linked with private sector labor demand. The promotion of women’s equal
access, especially to training in non-traditional skills, combined with the provision
of additional stipends to cover childcare, most likely also plays a role in enhancing
the benefits of these programs for female youth. By offering a wide range of skill
training to young women and men on an equitable basis, providing additional
support to young mothers in the form of a childcare stipend, and delivering the
programs through decentralized mechanisms that can better capture the skills

                                           4
needs and work opportunities of local communities, the new “Latin American”
model of vocational education has succeeded in increasing young women’s
employment and earnings in a number of countries. The replicablity of these types
of programs to other regions might be limited, however, by several factors, including
the pre-existing institutional infrastructure (vocational training centers and formal
private sector) and a socio-cultural environment in which women’s mobility and
access to education and employment is relatively equal, compared to other low-and
middle-income regions.

An alternative approach to enhancing young women’s economic opportunities comes
largely out of the population and reproductive health field, where programming
targeted specifically to adolescent girls has only relatively recently recognized the
economic dimension of the developmental transition from childhood to adulthood.
Assessments of the “livelihoods” approach to adolescent programming, in which
employment and/or business skills are imparted as part of a package of services
addressing girls’ needs for social capital, mentoring, and access to health
information and services, suggest that while such programs are extremely
promising, there is much to be done in the way of strengthening the ability of NGOs
and other implementing organizations to effectively integrate and appropriately
sequence the diverse range of activities that these kinds of programs encompass.
This would imply that piloting of relatively modest projects that are flexible in
design and include technical assistance to allow providers to be as effective as
possible across the areas of intervention may be the best course of action at present.


There is no straightforward answer to the question of whether the promotion of
young women’s employment and economic empowerment in developing countries is
better served by programs that limit themselves to high quality vocational training
along the lines of the Latin American model, or by programs that simultaneously
address multiple constraints limiting young women’s labor market participation, as
with the “livelihoods” approach. The ability of “minimalist” programs to
dramatically improve young women’s prospects in the labor market implies that –

                                           5
in the socioeconomic context in which they were implemented – relevant skills
acquisition, coupled with links to local employers, is enough to get many of these
youth into the workforce with decent earnings. In the countries in which the
vocational training were particularly effective for young women, other potential
barriers to female youth employment, such as social isolation and domestic
responsibilities, are not strong enough to prevent a narrowly-focused set of
activities from achieving their objectives – which were also narrowly-focused on
labor market outcomes for participants.


It is unlikely that a pure vocational training program could be optimally effective at
addressing the constraints on young women’s economic activity in most of the low-
income and post-conflict countries under consideration for inclusion in the World
Bank’s Adolescent Girls Initiative. In these countries, it is likely that the
constraints on young women’s employment are much more complex than simply
lack of relevant vocational skills. The model proposed for the Liberian pilot
program Economic Empowerment of Adolescent Girls in Liberia, in which job skills,
life skills, and entrepreneurship training with linkages to microfinance are being
combined with ancillary interventions addressing gender-based violence and
reproductive health, seems a reasonable starting place to assess the effectiveness of
a more integrated approach.




                                           6
1. Introduction


The World Bank is launching an initiative aimed at addressing the economic needs
of adolescent girls and young women in poor or post-conflict countries. Working
together with governments, donors, foundations, and private sector partners, the
Bank proposes to develop and test a core set of promising interventions to promote
the economic empowerment of adolescent girls and young women. The menu of
interventions, based on emerging good practices around the world, includes: i)
business development skills training and services; ii) technical and vocational
training, combined with job placement assistance; iii) access to micro finance; iv) life
skills training; v) mentoring; and vi) apprenticeship programs. As a part of
background research for this initiative, this paper undertakes a review of existing
policies and programs designed to promote labor force participation of young women
in developing countries.


It is well known that in developing countries, young women’s labor force
participation is influenced by a number of factors, including human capital
constraints, socio-cultural barriers to entry, demographic factors such as early
marriage and high fertility, and gender-based hiring preferences on the part of
employers. While programs that directly address marriage or fertility could
influence young women’s labor force participation, the focus of this paper is on
programs primarily addressing employment. Some programs for promoting young
people’s transition into the labor market take a minimalistic approach (eg.
concentrating on skills training alone). Other employment programs, particularly
those targeted to young women, simultaneously address multiple constraints
limiting participation (eg. lack of skills, limited mobility, child care needs, and lack
of sexual and reproductive health information). The goal of this paper is to unpack
and assess what elements of program design are essential to promoting young
women’s transition to the labor market.




                                            7
The paper is organized as follows: Section 2 provides an overview of some of the
major trends and issues facing young women in the labor market in an
international context. Section 3 describes a selection of “best practice” programs,
some of which are focused exclusively on employment training and others of which
take a more integrated approach to providing gender-targeted adolescent services.
Section 4 details the lessons learned from the implementation and evaluation of
these programs, and considers the circumstances under which a “minimalist” versus
comprehensive/integrated approach might be most effective. Section 5 concludes
and summarizes the policy and program recommendations.


   2. Issues: Gender and Youth Employment in Developing Countries


This section reviews the international data on youth labor markets, with special
attention to gender differences. It also considers some gender-specific challenges
facing young women with respect to both the supply of labor and demand for paid
employment.


   2.1 International Labor Market Trends for Young Women


      2.1.1. Labor Force Participation
   Young people1 reduced their labor force participation rates (LFPR) in every
   region of the world except the Middle East and North Africa between 1995 and
   2005 (see Table 1). The only countertrend was among young women in Latin
   America and the Caribbean and in the Middle East and North Africa, who joined
   the labor force at higher rates in 2005 than in 1995. As is the case for adults,
   female youth LFPR is everywhere lower than male youth LFPR, with the
   exception of East Asia, where more than 68% of young women are in the labor
   force.




                                          8
                                        Table 1




Source: ILO (2006)


      2.1.2. Unemployment


Conventional measures of unemployment (youth who are in the labor force but not
currently employed) indicate rates of between 7.5% for East Asia to 25% for the
Middle East and North Africa. In 2006, unemployment was significantly higher for
young women than for young men in the same regions that have experienced
increases in female youth labor force participation rates: Latin American and the
Caribbean and the Middle East and North Africa. In LAC, almost 22% of young
women are unemployed, compared to 14% of young men, and in MENA, close to one-
third of women in the 15-to-24 age group are unemployed, compared to less than
one-quarter of young men (see Table 2). In some countries, such as Lesotho, Saint
Lucia, and Suriname, female youth unemployment exceeds 40% (see Figure 1).




                                         9
                                  Table 2
          Youth Unemployment Rates by Gender and Region, 1996-2006




Source: ILO (2007)




                                     10
                                        Figure 1
                     Youth Unemployment by Gender, Selected Countries




Source: ILO (2007)


        2.1.3. Idleness and Joblessness


Perhaps a more useful measure of the strength of the youth labor market is the
proportion of the population that is neither in the labor force nor in school; that is,
the group that has left school but does not appear to be transiting into the world of
market work. Fares et al. (2006) define the “home status rate” (which is also
sometimes referred to as the “idleness rate”) as
HR = {N – [S&N]} / P
Where P = population
N = not in labor force = (P – L), where L = Unemployed + Employed
S&N = in school and not in the labor force


They further define the jobless rate as the home status rate plus the unemployment
rate:
                                           11
JR = HR + [U/P] = [P – E – S&N] / P
Where E = employed.


As Table 3 indicates, joblessness and home status are much more highly correlated
for young women than for young men, suggesting that variation in female
joblessness across countries is more closely tied to variation across countries in
female home time.2


                                        Table 3
             Youth Home Status and Jobless Rates by Region and Gender
Region                          Home Status (“Idleness”)           Jobless Rate
                                        Rate
                                 Female          Male         Female          Male
EAP                                27             9             32             22
ECA                                10             7             26             22
HIC                                6              2             17             14
LAC                                27             7             35             16
MENA                               26             3             36             22
OHIC                               3              1             12             10
SA                                 39             10            57             18
SSA                                24             14            34             20
All Developing Countries           23             8             34             21


Source: Fares et al. (2006), Appendix Table 1


   2.2 Gender-specific Challenges to Youth Employment


The World Development Report 2007: Development and the Next Generation
identifies four principal factors contributing to the vulnerability of youth in the
labor market (World Bank 2007a):
   (1) Large cohorts of new entrants and higher female participation rates, driving
         up labor supply and placing downward pressure on youth employment and
         wages;
   (2) Poor access to information and credit markets, leading to premature exit
         from school and perpetuating skill mismatches;

                                           12
   (3) Restrictive labor market institutions, such as unemployment insurance,
      employment protection laws, and the minimum wage, some of which have a
      disproportionate effect on youth;
   (4) Social institutions and norms, including religious beliefs and limitations on
      mobility, which hinder the full participation of girls in particular in skill
      acquisition and work.


In addition to these factors, there are several gender-specific determinants of low
employment rates and earnings among young women. On the supply side, girls are
responsible for a disproportionate share of unpaid domestic labor, either in their
parents’ or their own home (Ritchie et al 2004). In countries with early marriage
and/or childbearing, this may include primary responsibility for childcare (Okojie
2003). Other constraints on young women’s labor market participation might
include a lack of productive skills and contacts to help with a job search; limited life
skills; and opposition from family or male partners (Ruiz Abril 2008).


An often under-recognized barrier to young women’s employment is their relative
lack of the kind of social capital that can help them gain information about and
access to jobs. As discussed in a recent publication from the Center for Global
Development entitled “Girls Count,” research in Sri Lanka and Thailand suggests
that young men and women have access to very different social networks for
information about jobs and job training opportunities; these social networks may be
critical to their success in the labor market. In Thailand, where both parents and
youth have high expectations that out-of-school youth of both sexes will earn money,
girls consult their parents about work issues, while boys and young men cite
colleagues as their primary sources of information. Because young men rely on
broader social networks to advance their employment interests, they have more
information to improve their job prospects (Levine et al. 2008).


Data from three of the International Labor Organization’s School-to-Work Surveys
(STWS) give an indication of the cross-country and gender differences for the
                                           13
reasons young people are not in the labor force. In China, half of all female youth
surveyed cited housework and/or childcare responsibilities as the primary reason for
not working for pay; in Kosovo, almost a third of young women were out of the labor
force due to ill health or disability; while in Syria, one third were denied permission
to work by their families.
                                       Table 4




Source: ILO (2006)


On the demand side, employers in a range of countries reveal a striking preference
to hire young men rather than young women (ILO 2008). For formal sector
employers, this may be related to concerns regarding the likelihood of early
marriage and/or pregnancy on the part of young female workers, which could
generate high turnover rates or costly maternity leaves (Brewer 2004).
Findings from the STWS are revealing of employer perspectives on hiring youth in
general, and young women in particular. In Egypt, for example, 34 percent of a
sample of almost 350 employers reported a preference for hiring men for
professional and managerial occupations, and 69% expressed a similar gender
preference for production and manual occupations. In the Egyptian case, male bias
in hiring preferences is present both the industrial and service sectors of the
economy (see Table 5).




                                          14
                                       Table 5
             Egypt: Preferred Hiring by Sex, Industry and Occupation




Source: El Zanaty and Associates (2007)


Another recent employer survey in Sierra Leone found that employers were biased
against females, and the bias increases with age of employees (World Bank 2007b).
More than twenty percent of the 668 employers surveyed stated that they perceived
18-to-24 year-old men as more skilled than young women, and 36 percent perceived
male youth as harder working than females. On the other hand, female workers
were considered slightly more reliable, trustworthy, and cooperative.


It is important to note that employer preference for young female workers does exist
in some countries and in some industries. Women tend to be concentrated in
electronics, textiles and garments, where low labor costs are a crucial part of
international competitiveness. Women’s wages are typically lower than men’s, and
employers perceive women as more productive in the types of jobs available in the
export sector. Reasons that employers cite for the latter include: women’s putative
“nimble fingers;” their obedience and being less prone to worker unrest; their being
suited to tedious work; and their reliability and trainability relative to men
(Braunstein 2000). Specific examples include the garment industry in Bangladesh
(Amin et al. 1998), export agribusiness in Mexico (Arizpe and Aranda 1981), coffee

                                          15
and tobacco farmers in Honduras (Buvinic 1998); the cut flower industry in Ecuador
(Newman 2002), and the export processing zone in Madagascar (Glick and Roubaud
2006).


3. Programs: Which Youth Employment Initiatives Work Best for Young Women?


         3.1 Youth Employment Inventory


The World Bank’s Youth Employment Inventory (YEI) identifies forty-five out of
291 youth employment programs around the world which a ranking of “1”
(“positive”) for women’s access. (Two hundred and thirty programs reported having
“gender-neutral” access; the remaining 14 are considered to have “negative” gender
access practices, or the equality of access on the basis of gender is not known.)
Relatively few of these gender “best practice” programs have had impact
evaluations that would allow for reliable measurement of the degree to which they
have achieved their objectives: fifteen (one-third) have no evaluation information
available at all on outcomes or impact; 20 (44%) have evaluations which include
basic information on the gross outcomes of the intervention (e.g. number of
participants who found a job after the intervention, improvement in earnings of
participants) without considering net effects (i.e., there is no control group). The
YEI does include ten programs with the highest gender equality access ranking for
which estimates of net impact on labor market outcomes such as employment and
earnings in the labor market (using control groups to measure impact); six of these
also have been subject to some kind of cost-benefit analysis.


         3.2 Review of Selected “Best Practice” Programs


As indicated in the YEI, it is a challenge to identify youth employment programs
targeting women which have had sufficiently rigorous impact evaluations to be able
to draw lessons for future work. This section focuses on a small number of projects


                                          16
for which good impact evaluations have been conducted, and which appear to have
had quite positive outcomes for the young women who participated.


             3.2.1. The “Latin American” Model


Two projects from Latin America – ProJoven (Peru) and Jóvenes en Acción
(Colombia) – followed the earlier models of youth employment programs in Chile
and Argentina by contracting with decentralized training entities to organize and
offer training courses in which beneficiaries could enroll. The courses, which
contained both classroom and on-the-job experiences, matched local firms’ needs
with the content of the training curriculum.


In the case of ProJoven, which provided vocational training to approximately 42,000
sixteen-to-twenty-four year olds in ten major cities between 1996 and 2003, equal
participation of women in its training courses was strongly encouraged, especially
for traditionally male-dominated occupations. Over the eight years of the program’s
operation, women made up over half (52-56%) of the beneficiaries (Diaz and
Jaramillo 2006). Women with children under the age of five received a double
monthly stipend to cover transportation, meals and medical insurance.
Expenditures for mothers’ subsidies represented 6.6% of total stipends, or less than
1.5% of ProJoven’s total budget (Ñopo et al. 2007). After three months of training,
beneficiaries who passed competency tests graduated to a three-month internship
with a local firm.


Using a matched control group methodology, a gender evaluation of ProJoven found
no substantial differences between women and men from training on hourly wages,
but important gender differences in outcomes for employment rates and
occupational segregation (Ñopo et al. 2007). The program’s impact on employment
rates was much greater for women than for men: twelve months after completing
their internships, employment rates for female beneficiaries were almost 6% higher
than for a matched control group, and after 18 months, the difference grew to
                                         17
15.2%. For young men who participated in the program, average employment rates
were actually 11% lower than the control group of non-beneficiaries.


The impact on hours worked also varied significantly by gender, with male
participants working almost 15 more hours per week than the control group, and
women increasing their average number of hours worked per week by almost 6.
The impact on hourly earnings was very similar for men and women: both earned
an average of 20% higher than the control group 18 months after program
participation. Perhaps the most striking difference of program impact between men
and women was in total monthly earnings, which combines the employment and
labor income effects: 18 months after completing the program, female participants
had increased their monthly earnings by nearly 93% relative to the control group,
while men’s earnings were only 11% higher.


Another important finding for ProJoven was that the program reduced occupational
segregation among its targeted population. The value of the Duncan Index, which
can be interpreted as the percentage of working women that would have to switch
from female-dominated to male-dominated jobs to achieve an equal distribution of
men and women across occupations, fell from .63 at baseline to .46 eighteen months
after the program. The comparable indicators for the control group were .68 at
baseline and .73 eighteen months later. The authors of the impact evaluation
conclude that “labor-training programs that promote equal gender participation
have disproportionately positive effects on outcomes for women trainees in a labor
market with substantial gender differences” (Ñopo et al. 2007).


The second Latin American youth employment training program, Jóvenes en
Acción, was introduced in Colombia between 2002 and 2005. It provided 3 months
of in-classroom training and 3 months of on-the-job training to 80,000 young people
between the ages of 18 and 25 in the two lowest socio-economic strata of the
population. The total cost of the program was US$70 million, or US$875 per
person. The training courses provided vocational skills in a diverse number of
                                         18
occupations, including taxi and bus drivers, office assistants, call center operators,
medical assistants, preschool teacher assistants, cashiers, textile operators, and
carpentry, plumber, and electricians’ assistants. On-the-job training was provided
by legally registered (formal sector) companies, which provided unpaid internships
to the participants. Trainees received a daily stipend of $2.20, which was
eventually raised to $3 for women with children under 7 (Attanasio et al. 2008).


A randomized experimental design was used to evaluate the impacts of Jóvenes en
Acción on employment and earnings. Labor supply effects for women were large
and significant: controlling for both training institution fixed effects and pre-
treatment characteristics, female trainees’ probability of paid employment 19-21
months after completing the program increased by five percentage points relative to
the control group, and they worked an average of 1.1 more days per month and 2.5
more hours per week than women in the control group. (Young men trainees’
employment probabilities and quantity of labor supplied were not statistically
significantly different than those of the control group.) Women who participated in
Jóvenes en Acción earned approximately 18% more than women who didn’t receive
the training, while men realized earnings gains of eight percent relative to the
control group (Attanasio et al. 2008).


The results for Jóvenes en Acción imply that training has had positive labor market
impacts for both men and women, but possibly through different mechanisms. For
women, there is a clear effect on employment, days, and hours, as well as wage and
salary earnings. For men, the picture is different; the estimates imply an increase
in the average wage and salary earnings among treated employed workers, but no
effect on employment rates. For women, a third of the earnings increase can be
attributed to increased employment.


Cost-benefit analysis of Jóvenes en Acción suggests a large net gain, especially for
women. The internal rate of return is estimated to be 25% for women and 16% for



                                           19
men, if gains are assumed to permanent and constant over remaining working life
of participant (Attanasio et al. 2008).


The authors of the Jóvenes en Acción impact evaluation study speculate that there
are three possible reasons for the differential returns to training for women relative
to men. First, the higher return could result from the fact that women have lower
levels of formal education to begin with. However, baseline gender differences in
years of education were very small: 9.9 for women compared to 10.1 for men
(Attanasio et al. 2008, Table 2). Second, women with children received an
additional stipend for child care, which may have freed up additional time for
women to devote to training. Indeed, women did tend to enroll in courses requiring
longer hours. Third, women were more motivated and were more responsible
during the classroom and internship phases of the program; dropout and expulsion
rates were both lower for women than for men.


Part of the success of the Jóvenes en Acción program for female youth in particular
may also be attributable to the fact that it was run by known NGOs and located in
the young person’s neighborhood. This might have been especially important for
girls with small children at home and who had little experience or confidence to
venture far from their homes. The NGOs spent a lot of informal time in general
counseling, and a lot of effort on discipline. Thus, while Jóvenes was an
employment program, it also provided social work services for the students.
According to World Bank staff affiliated with the project, the life skills component,
which comprised 20% of training time at some sites, was highly praised by the
participants, and the internship component was particularly important to women
than men since they are less linked to labor markets (Wendy Cunningham, personal
communication).


Both ProJoven and Jovenes en Acción – along with similar projects in other Latin
American countries – could be considered traditional vocational education programs
in the sense that their objectives were largely limited to labor market outcomes,
                                          20
without consideration of other aspects of youth welfare. Their success in promoting
young women’s employment can be attributed to the fact that they were
exceptionally well-targeted, demand-driven, and linked with private sector labor
demand. The promotion of women’s equal access, especially to training in non-
traditional skills, combined with the provision of additional stipends to cover
childcare, most likely also played a role in enhancing the benefits of these programs
for female youth.


The replicablity of these types of programs to other regions might be limited,
however, by several factors. First, the decentralized nature of these programs
relied on a pre-existing, well-developed national network of private and not-for-
profit vocational training centers, as well as community-based non-governmental
organizations, with which to contract. While the quality of the individual training
institutes varied, the basic infrastructure was already in place with respect to
facilities, personnel, and links to local employers. Even where participating NGOs
may not have had prior experience in offering vocational training, the selection
criteria of the program helped to assure that met certain quality standards. Many
developing countries lack such vocational training infrastructure. Second, the
formal private sector in these middle-income countries is large and vibrant enough
to be interested in participating in such a scheme; none of the evaluation reports
mentioned lack of private firm participation as a limiting factor. In countries where
the formal sector is small, creating such linkages to provide on-the-job training may
present more of a challenge. A third set of factors limiting replicability of the Latin
American youth employment training model are more social in nature, related to
the relative lack of restrictions on young women’s mobility and their ability to
interact with men in public. The gender-integrated, training center-based approach
that has had so much success in Latin America would be unfeasible to implement in
societies where young women are largely confined to their parents’ or husband’s
home, and/or where prohibitions exist on the intermingling of unrelated men and
women in the public sphere.


                                          21
             3.2.2. Adolescent Livelihoods Programs


An alternative approach to enhancing young women’s economic opportunities comes
largely out of the population and reproductive health field, where programming
targeted specifically to adolescent girls has only relatively recently recognized the
economic dimension of the developmental transition from childhood to adulthood.
The rationale for integrating “livelihoods training” into population work is
summarized by Mensch et al. (2004):
      Broadly conceived, the “livelihoods approach” to adolescent
      programming attempts to develop technical and life skills while
      influencing social networks and improving access to savings, loans,
      and markets. In settings where young women’s movements are
      restricted to the domestic arena, providing safe spaces outside the
      home is expected to promote mobility and independence and give girls
      greater visibility in the community. By increasing contact with others
      outside the family, including both female peers and adults who can
      function as mentors, social and interpersonal capacities may be
      advanced and communication skills developed. Finally, livelihood
      programs offer acceptable settings for supplying information about
      reproductive health.

A report by the International Center for Research on Women (ICRW) also makes
the case for integrating reproductive health and livelihood programs:

      Increasingly, donors and governmental and non-governmental
      organizations faced with the challenge of addressing and prioritizing
      adolescent needs are recognizing that various aspects of young people’s
      lives are intricately connected and may benefit from simultaneous
      attention. In particular, there is increased understanding that
      reproductive and sexual behavior for adolescents is closely linked with
      their educational and economic options. Teen pregnancy, abortion, and
      exposure to HIV/AIDS and other sexually transmitted diseases (STDs)
      have enormous social and personal impact in terms of educational and
      work opportunities. Conversely, entry into the labor force and
      economic options during the teen years are critical in determining not
      only future opportunities for social and economic mobility, but also
      exposure to health risks, fertility outcomes, and overall well-being
      (Esim et al. 2001).


The ICRW report distinguishes between linked reproductive health/livelihood
programs which are implemented by “sector-specific” organizations and those
                                          22
carried out by “multi-service organizations.” Sector-specific organizations are those
with expertise and experience in either reproductive health or economic
empowerment, which might incorporate interventions outside of their usual domain
for specific target groups of adolescents. For example, an NGO in the reproductive
health field might implement a skills development programs for teenage mothers,
or a labor rights organization might initiate a sexual health program with outreach
to a specific workplace, such as garment factory workers. Alternatively, sector-
specific organizations interested in meeting the multiple needs of youth might
establish linkages through referral systems, drawing on other organizations’
complementary skills and experience. In contrast, multi-service or community-
based organizations provide multiple programs and activities for their target
population, but these programs are often not explicitly linked, or even offered to the
same group of beneficiaries (Esim et al. 2001).


The Population Council, which has been at the forefront of the “livelihoods
approach” to adolescent programming, has evaluated several programs in South
Asia and Africa that have successfully combined vocational training with
reproductive health and other interventions. The best documented of these are:
   1. The “Action for Slum Dwellers’ Reproductive Health, Allahabad” (ASRHA),
      which offered vocational training and savings opportunities for girls 14-19 in
      Allahabad, India (Mensch et al 2004);
   2. “Adolescent Girls’ Adventure” (Kishori Abhijan), a program in rural
      Bangladesh which offered livelihood skills (life-skills lessons, savings account
      options, access to credit, and vocational training) to 15,000 adolescent girls in
      three districts (Amin and Suran 2005); and
   3. Tap and Reposition Youth (TRY), a Kenyan savings and microcredit project
      targeted to urban out-of-school adolescent girls and young women aged 16-22
      (Erulkar and Chong 2006).


ASRHA was a project conducted in the slum areas of Allahabad in the northern
Indian state of Uttar Pradesh in 2001-2002, which integrated livelihood activities
                                          23
for adolescent girls aged 14-19 into CARE’s reproductive health program. Literate
girls with their parents’ permission were identified as peer educators, who attended
a 6-day reproductive health training course followed by a 2-day peer-education
training course. These peer educators then recruited eligible girls into 27 groups of
approximately 20 each (total N = 525), which received reproductive health training,
vocational counseling, savings formation information, and follow-up support.
Vocational training, which focused on handicrafts skills such as sewing, weaving,
and jewelry making, was provided by the project, and also by NGOs and
government agencies. In addition, the girls received counseling and assistance for
creating savings accounts at banks or post offices.


An unmatched control group of adolescent girls in another slum ward participated
in the ASRHA reproductive health activities, but not the livelihood components.3
This allowed the evaluation team to determine the marginal effects of vocational
and financial training on the outcomes of interest, which included gender-role
attitudes, mobility, self-efficacy, reproductive health knowledge, work expectations,
and time use. Labor market outcomes such as employment and wages were not
considered. The impacts of the program were found to be limited to those variables
that measured skills directly or for issues addressed directly by the intervention.
Only 10% of beneficiaries reported earning income from selling products they’d
made. Perhaps the most significant outcome was that intervention participants
showed a considerable increase in reproductive health knowledge, relative to control
respondents who attended the reproductive health classes that lacked the
additional livelihoods component (Mensch et al. 2004).


Kishori Abhijan, a UNICEF-funded initiative on adolescent girls’ livelihoods, was
implemented in 2001-2002 by two Bangladeshi development NGOs: the Bangladesh
Rural Advancement Committee (BRAC) and the Center for Mass Education and
Science (CMES). The program had three components: (1) mentoring to develop self-
esteem and leadership skills; (2) training in gender and gender discrimination,
health and nutrition, and legislation and legal rights; and (3) acquisition of
                                          24
livelihood skills, including links with existing facilities for establishing savings
accounts and obtaining credit. Using a propensity score matching methodology to
evaluate the difference-in-differences impact of the program, Amin and Suran
(2005) found that program participants had significantly increased their labor
market participation relative to matched non-participants during the two years that
the program was active in the communities. Logistic regression analysis of changes
in paid work status between 2001 and 2003 suggests that the odds ratio of working
for pay increased by between 1.9 and 3.2, controlling for district, age, wealth,
education, credit history, and marital status.4 Put differently, matched participants
increased their paid employment from about 11% before program implementation to
between 27.7 and 53.9% ex post, compared to the matched control group’s rise in
paid work from 10% to 20.3%.


The Kenyan Tap and Reposition Youth program is an example of a livelihoods
intervention with a highly vulnerable population. In the slums of Nairobi, 78% of
girls aged 15-17 are not in school and 58% do not live with their parents. Twenty-
one percent of sexually active girls aged 15-19 reported exchanging sex for money or
gifts, and the HIV infection rate for girls aged 15-24 was eight percent (Erulkar et
al. 2006). The K-Rep Development Agency, which is the oldest and largest
microfinance institution in Kenya, decided in 1998 to work in collaboration with the
Population Council to expand its services to adolescent girls and young women. The
TRY model evolved over the life of the program, growing from a minimalist savings
and credit model to one that also provided its clientele with social support and an
individual, voluntary savings option.


During the first phase of the project (1998-2000), K-Rep essentially replicated its
Asian microcredit model for the urban teen girls it was targeting: groups of 25
received training, contributed to group savings, and applied for microloans for small
business undertakings. As repayment rates dropped and groups began to dissolve,
modifications were made in the second phase (2001-2004). Specifically, TRY
expanded on its social support aspect by adding adult mentors to work with the
                                           25
groups in parallel with credit officers. But high dropout rates continued due to the
demands of the savings and lending program. Most recently, beginning in 2004, the
program has further evolved to include a “Young Savers Club,” which allows for
voluntary individual savings and maintains group activities.


The high rate of program attrition from TRY, combined with significant residential
mobility of poor young urban women, compounded the usual challenges of selection
bias for the impact evaluation of this project. Baseline data were collected from 326
matched pairs of young women; by the time the evaluation survey was administered
one to three years later, only 68% of those pairs (n=222) could be located. Moreover,
attrition from the sample is perfectly correlated with attrition from the program; all
girls lost to follow-up were TRY dropouts. Keeping in mind the upward bias this
non-random attrition creates for the program impact assessment, Eralkar and
Chong (2005) find statistically significant differences in household assets, earnings
from paid work, and savings between TRY participants and the matched control
group. Interestingly, the asset and earnings effects were only significant for the
older group of beneficiaries (aged 20-22), while savings rates increased only for the
younger group (aged 16-19).


The experience of TRY over a 10-year period offers an unusual opportunity to learn
lessons about how to adapt program models to the specific needs and constraints of
adolescent girls and young women. (In this sense, it is a “best practice” not because
of reliably measured outcomes, but because it is an example of responsive
programming in an emerging field.) In urban Kenya, for the majority of young
women, entrepreneurship and repeated borrowing were not primary concerns.
Rather, their fundamental needs were related to acquiring social capital (including
accessing support groups and mentors), maintaining physical safety, and having the
opportunity to save their money in a safe, accessible place. When these needs are
met, entrepreneurship and use of credit opportunities may follow (Erulkar et al.
2006).


                                          26
   4. Lessons: What can new youth employment initiatives learn about how best to
      meet the needs and challenges specific to young women?

Considering the very different approaches to employment programming for young
women reviewed above, it may be useful to distinguish between lessons learned
from narrower vocational education programs, such as those that have been carried
out in Latin America, and recommendations coming out of experiences that combine
employment skills development with other interventions.



   4.1 Gender Issues in Vocational Education and Training Programs


A new generation of youth employment skills training programs, particularly in
Latin America, has largely addressed some of the criticism of earlier vocational
education and training programs, many of which lacked attention to gender
inequities. The large, centralized training institutions of the past focused
exclusively on male skill areas and few girls went on to study vocational subjects,
other than those “feminine” skills for lower paid jobs, such as childcare, secretarial
or domestic work (Brewer 2004). By offering a wide range of skill training to young
women and men on an equitable basis, providing additional support to young
mothers in the form of a childcare stipend, and delivering the programs through
decentralized mechanisms that can better capture the skills needs and work
opportunities of local communities, the new “Latin American” model of vocational
education has succeeded in increasing young women’s employment and earnings in
a number of countries.

What more could these youth training programs do to enhance the benefits for
women? Throughout the project cycle, the following five sets of activities are likely
to increase the marginal benefit of vocational education for female youth:

      4.1.1. Gender-sensitive, Participatory Needs Assessment

Depending on resources, programs can make use of existing demographic, labor
market, and/or micro level living standards data, or implement special data
                                          27
collection efforts focused on understanding the issues and obstacles facing youth.
An excellent template for such a needs assessment is the International Labor
Organization’s School-to-Work Transition Survey (SWTS). The SWTS is a
statistical tool designed by the ILO to assist countries in improving the design of
youth employment policies and programs. Specifically, the survey allows for
analysis of both quantitative and qualitative variables that define the relative ease
or difficulty of young people’s transition from school-to work-life. Among such
variables are young people’s education and training experience; their perceptions
and aspirations in terms of employment; the job search process, barriers to and
supports for entry into the labor market; the preference for wage employment or
self-employment; attitudes of employers towards hiring young workers; and
working conditions and earnings. As of 2008, surveys have been conducted and
reports issued for Indonesia, Vietnam, Sri Lanka, Kosovo, Syria and Egypt (ILO
2008). The World Bank’s Girls’ Vulnerability Assessment for Liberia and Youth
Employment Study for Sierra Leone are other excellent examples of baseline data
collection and analysis that allow program planners to identify potential areas of
productive intervention, as well as potential gender-specific challenges to
implementing youth employment programs in a particular context (World Bank
2007b and 2007c).


      4.1.2. Targeting and Outreach

In the design of an information and communication campaign for a youth
employment program, special attention should be given to materials and
outreach/dissemination methods that will reach young women where they are. In
Brazil, promotional materials for a technical training program in Bahia and
Pernambuco in 1991-92 made use of explicit gender equity language and images of
women in non-traditional fields to promote women’s participation in the project.
The number of female applications doubled, and there was a largely successful job
placement after training (Fawcett and Howden 1998). Programs that are open to
both men and women should also set gender-specific targets regarding the desired
number of female participants, which can be monitored as the program is rolled out.
                                          28
   4.1.3. Adapting Curricula to the Needs and Skills of Young Women

In the Latin American demand-driven youth training programs, the decentralized
training institutes are generally given a great deal of autonomy with respect to the
development and delivery of actual course content. While this is consistent with the
idea of matching training to local labor market conditions, more could be done to
address the particular training needs of young women. In the case of traditionally
female occupations, increasing the technical level of training has the potential to
improve women’s productivity and the applicability of the training to labor market
needs in such areas as textiles, tourism, office work and health care. One
innovative approach is what has been called the “stepwise” method of relating
similarities in the occupational background and existing skills of participants with
the skill needs in more productive sectors in order to identify possible new training
areas. With this approach, training programs can identify and incorporate more
technical skills into traditional women’s training curricula. An example of the
stepwise method comes from the national training institute in Colombia (SENA),
which identified the growth-oriented, export sectors of graphic arts and jewelry
manufacturing, and provided technical assistance to women already skilled in
related areas of sewing, handicrafts, and garment manufacturing to establish small
manufacturing operations.

With respect to promoting women’s inclusion in training for traditionally male
occupations, one option is to offer “pre-training” courses to women in order to
familiarize them with the basic concepts, terminology and tools of the specialized
training. For example, a metalworking training program for women in Brazil and a
plumbing and electronic appliance repair training program in the Dominican
Republic both provided courses in math, measurements, and design to women who
were interested in enrolling (Fawcett and Howden 1998).


Another important issue related to curriculum relates to the degree to which youth
programs should be training participants for paid employment or for self-
                                          29
employment. One of the uses to which a good needs assessment can be put is to
help determine the relative opportunities and growth potential for women in the
local labor market, and whether the returns are likely to be higher for preparing
girls to become employees or entrepreneurs. In the latter case, an independent
microcredit component and/or other follow-on support may be important to helping
young women get their small businesses off the ground. For example, Project
Baobab in Kenya, which targets low-income rural young women, provides free
business skills training (entrepreneurship training along with a life-skills training
program) and small grants for business start-ups for some of those who are trained
in secondary schools and vocational centers. As part of the training, students create
individual business plans that focus on enterprises like selling second-hand clothes,
raising bees or chickens, or tailoring. These business plans are submitted to a
committee from the local business community, who evaluate them and recommend
the most promising to receive US $100 grants from Project Baobab
(http://www.projectbaobab.org/).


In addition to technical skills, youth training programs that want to help young
women find and keep jobs should include both job readiness skills (eg. effective
workplace communication and interpersonal skills, including conflict resolution,
active listening, negotiation, and getting along with co-workers; occupational
stereotyping; assertiveness; managing stress and discrimination; and childcare
issues) as well as job search skills (eg. information on how to find job openings and
opportunities, writing a resume and cover letter, interview skills, and networking
with employers) as part of the course curriculum.


   4.1.4. Helping Young Women Balance Domestic Responsibilities with Job
          Training


In the implementation phase of the project cycle, it is important that provisions be
made to recognize many young women’s high opportunity cost of time and their
duties with respect to unpaid household labor. For example, the duration of the
                                          30
training program is an important factor for women’s participation; shorter courses
and/or financial assistance for longer training programs should be made available.
Similarly, course schedules and locations should be available that are compatible
with young women’s household responsibilities, such as preparing the mid-day
meal. Finally, for young mothers who want to take part in employment training, it
is essential that provisions be made to help them with childcare.


Several different models of childcare support have been tried in Latin America and
the Caribbean. These include:
      Provision of stipends to young mothers for home- or center- based care –
       including credits for use of community programs like hogares comunitarios in
       Colombia, or securing enrollment spots for trainees in government childcare
       program in Chile;
      Offering the vocational training in community centers with on-site childcare
       – this was the model used in an Uruguayan program; and
      Combining early childhood education with adolescent programming, which is
       the approach taken by the NGO SERVOL in Trinidad and Tobago. This
       latter model is particularly innovative, because it uses the onsite preschool as
       a training site for the adolescents in the jobs program (Griffith 2002).


       4.1.5. Gender Sensitive Labor Market and Social Support Services


A further set of recommendations for improving the gender outcomes of youth
employment programs concerns the incorporation of post-training services to help
trainees actually find jobs that match their new skills set. These outplacement and
support services might be particularly important for young women with no previous
labor market experience and weak social networks to help them get started.
Examples of these kinds of services include career counseling, guidance, job
placement, mentoring/coaching, technical assistance, and provision of market
information. Two youth employment programs in rural India offer examples of how
this can be accomplished. The Baatchit Project complemented its labor market and
                                          31
entrepreneurial skills training with a career guidance component, which created
awareness among jobseekers about available career options, job vacancies and
facing interviews, and offered assistance with suitable job placements. The
Bharatiya Yuva Shakti Trust, a microenterprise program targeted to un- and
underemployed youth, works with a mentoring system whereby beneficiaries
receive one-on-one guidance, counseling, and monitoring and supervision as they
establish their small business (Brewer 2004).


   4.1.6. Gender-disaggregated Impact Assessment


As with the baseline needs assessment, it is important that all monitoring and
evaluation activities, as well as impact assessment efforts, collect data that allow
for analysis of how well the program is working for both women and men.
Reflecting on some of the unexpected findings in the Latin American impact
evaluations, this means not only keeping track of individuals’ employment status
and earnings after they complete the program, but also of sector and occupation (to
test the impact on labor market segregation, for example). Deeper analysis of
program impact on gender-specific outcomes such as time use, decision-making
power in the home, expenditure patterns (including investments in children’s
human capital), and domestic violence would necessitate both baseline and ex post
data collection on these variables as well.


Ideally, the impact evaluation will be designed early in the project cycle, and
incorporate some aspect of randomization or matched control group selection.
Where possible, treatment differentiation (for example, inclusion or exclusion of a
life skills component) can help distinguish between the marginal impacts of
program components. A good example of this is the proposed impact assessment
plan for the World Bank’s Economic Empowerment of Adolescent Girls project in
Liberia, which incorporates a rolling enrollment design such that the control group
will consist of those girls who are selected into the program, but for whom the
training courses do not yet have a space available (World Bank 2007c).
                                          32
   4.2. Lessons Learned from Adolescent Livelihoods Programs


In her summary of the lessons learned from the Population Council livelihoods
initiatives, Amin (2008) notes the four following common features that have
contributed to positive outcomes:
   1. Participants are organized into small groups that meet frequently;
   2. Peer educators provide skills training and group leadership;
   3. There exists a strong social support/mentoring component; and
   4. A phased approach, which begins with entry-level programs that provide a
      safe and supportive space where young people can gather; offer opportunities
      for individual, voluntary savings; and impart training in life skills, financial
      literacy, and health education. Later phases of the program can include more
      demanding options such as goal-oriented savings, vocational and business-
      skills training, and micro-credit.


Based on qualitative case studies of nine integrated adolescent reproductive health
and livelihood programs in Colombia, India, and Kenya, researchers from the
International Center for Research on Women concluded that, while client demand
for these types of programs is evident across diverse socioeconomic settings,
      The field of linked programming is clearly in its own adolescence, with
      a mixture of commitment, missteps, risk-taking, and optimism.
      Currently, most programs are not implementing linked strategies in
      an optimal fashion, often achieving only marginal effectiveness in
      meeting both the reproductive health and livelihoods needs of young
      people (Esim et al. 2001).


The ICRW report makes several useful recommendations for strengthening
programs that seek to address both young women’s economic and reproductive
health needs. Among these are: (1) Develop institutional and technical capacity of
implementing organizations, including staff training in new areas of intervention
and impact evaluation techniques; and (2) Integrate market assessment and
outreach as an essential component of livelihoods interventions. With respect to the
                                           33
second recommendation, the ICRW team makes the important observation that
translating livelihoods interventions into income-earning opportunities for youth
requires a thorough evaluation of and interaction with market needs, contacts, and
networks. As indicated by the evaluations reviewed above, many livelihood
programs have limited success with respect to employment outcomes due to lack of
information and knowledge about the skills and products that are most viable in the
local and national economies. Program personnel are also limited in their
understanding of how to best commercialize and market the training, skills and
products that youth develop. ICRW identifies the lack of a sufficient pool of
qualified experts with an overlap of programmatic experience and an understanding
of labor markets as a major constraint in incorporating such market assessments
and links in program design (Esim et al. 2001). In this aspect, livelihood programs
could look to the Latin American experiences for successful methods of linking
vocational training with private sector labor demand.


Taken together, these assessments of the livelihoods approach to adolescent
programming, in which employment and/or business skills are imparted as part of a
package of services addressing girls’ needs for social capital, mentoring, and access
to health information and services, suggest that while such programs are extremely
promising, there is much to be done in the way of strengthening the ability of NGOs
and other implementing organizations to effectively integrate and appropriately
sequence the diverse range of activities that these kinds of programs encompass.
This would imply that piloting of relatively modest projects that are flexible in
design and include technical assistance to allow providers to be as effective as
possible across the areas of intervention may be the best course of action at present.


      4.3. Minimalist Approach or Integrated Adolescent Services?


There is no straightforward answer to the question of whether the promotion of
young women’s employment and economic empowerment in developing countries is
better served by programs that limit themselves to high quality vocational training
                                          34
– such as the recent Latin American experiences reviewed above – or by programs
that simultaneously address multiple constraints limiting young women’s labor
market participation, as with the “livelihoods” approach. As discussed earlier, the
documented successes of the Latin American programs may be in part attributable
to the pre-existing institutional infrastructure (vocational training centers and
formal private sector) and a sociocultural environment in which women’s mobility
and access to education and employment is relatively equal, compared to other low-
and middle-income regions. That is to say, the ability of those “minimalist”
programs to dramatically improve young women’s prospects in the labor market
implies that – in the socioeconomic context in which they were implemented –
relevant skills acquisition, coupled with links to local employers, is enough to get
many of these youth into the workforce with decent earnings. It also implies that,
in the countries in which the vocational training were particularly effective for
young women, other potential barriers to female youth employment, such as social
isolation and domestic responsibilities, were not strong enough to prevent a
narrowly-focused set of activities from achieving their objectives – which were also
narrowly-focused on labor market outcomes for participants.


The choice between minimalist and integrated approaches also depends on the goals
of the intervention. Vocational training programs are usually motivated solely by
the objective of enhancing the employability of participants, as measured by ex post
employment and earnings. Livelihoods programs tend to be much broader in scope,
with “fuzzier” objectives like “gender role attitudes” and “self-efficacy” (Mensch et
al. 2004). Practically speaking, many livelihoods programs have been added onto
existing reproductive health interventions, and the economic component is seen
principally as a means to enhancing young women’s empowerment with respect to
demographic decisions such as marriage and childbearing. Conversely, when social
networking and/or reproductive health training are integrated into vocational or
microenterprise programs, these new components are often treated as instrumental
to achieving the economic objectives, as opposed to worthwhile interventions in and
of themselves.
                                          35
It is unlikely that a pure vocational training program – even one as well-designed
as the Latin American examples discussed in this paper – could be as effective at
addressing the constraints on young women’s economic activity, in most of the
countries under consideration for inclusion in the World Bank’s Adolescent Girls
Initiative. These are low-income and post-conflict societies with limited
institutional capacities, weak private sectors, challenging health environments, and
highly vulnerable youth populations. It is therefore likely that the constraints on
young women’s employment are much more complex than simply lack of relevant
vocational skills – although this is certainly one of the constraints that individual
country-level programs within the Initiative should address. The model proposed
for the pilot program in Liberia (Economic Empowerment of Adolescent Girls), in
which job skills, life skills, and entrepreneurship training with linkages to
microfinance are being combined with ancillary interventions addressing gender-
based violence and reproductive health, seems a reasonable starting place to assess
the effectiveness of a more integrated approach.


5. Conclusions and Recommendations


High rates of female youth joblessness across the developing world are partly
attributable to gender-specific barriers to young women’s labor force participation.
While demand-side constraints, such as employer discrimination, are a factor in
many labor markets, this paper focuses on programs which address the multiple
obstacles facing young women with respect to labor supply: inadequate skills, lack
of knowledge of job search techniques, unpaid domestic labor responsibilities, and,
in some places, social isolation and restrictions on mobility.


Two different programmatic approaches to addressing these constraints on young
women’s employment have been discussed in some detail: the “Latin American”
model, which focuses on decentralized, gender-equitable vocational training with
strong linkages to the local private sector, and the “livelihoods approach,” which
                                          36
places more emphasis on peer group formation and the integration of sexual and
reproductive health education with vocational and/or microenterprise training.
While the generally positive employment and earnings effects of the Latin American
programs are well-documented, evaluations of livelihood programs suggest more
modest labor market outcomes.


The paper argues that the impact of the narrow vocational training programs could
be even further enhanced by the use of gender-aware needs assessments, improved
targeting and outreach to girls, adaptation of training curricula to the specific needs
and skills of young women, provisions for girls’ “double duty” as students and
unpaid household workers, incorporation of post-training outplacement and support
services, and high quality, gender-disaggregated monitoring, evaluation, and impact
assessment. Livelihood programs, for their part, could benefit from adopting a
phased approach to participation, investing in the technical capacity of project staff
to be able to carry out multidisciplinary activities, and strengthening knowledge of
and links with local labor market conditions.


In practice, the choice between minimalist and integrated female youth employment
programs depends on the social, economic, and institutional context in which the
program is going to be implemented. The importance of field-based assessment of
labor market conditions, life circumstances of adolescent girls and young women,
employer attitudes towards youth and women, and the existing institutional
infrastructure for program implementation, cannot be overstated. It is only through
such an assessment that the actual barriers to young women’s labor force
participation can be identified, and the strengths and weaknesses of youth and
labor organizations evaluated. Based on analysis of this context-specific
information, program planners can then design an initial intervention somewhere
sensible on the minimalist – integrated services spectrum. Keeping in mind that
employment programs in developing countries targeting adolescent girls and young
women are still in their infancy, any program should remain highly flexible and
responsive to feedback and experience as the various components are carried out.
                                          37
References
Alissa, Sufyan (2007). The school-to-work transition of young people in Syria.
       Employment Policy Paper 2007/3. Geneva: International Labour
       Organization.


Alzua, Maria Laura, Paula Nahirnak, and Belissario Alvarez de Toledo (2007).
      Evaluation of entra 21 Using Qualitative and Quantitative Data. Q-squared
      Working Paper No. 41. University of Toronto Centre for International
      Studies.

Amin, Sajeda and Luciana Suran (2005). “Program Efforts to Delay Marriage
      through Improved Opportunities: Some Evidence from Rural Bangladesh.”
      Paper presented at the Annual Meeting of the Population Association of
      America, Philadelphia, 31 March – 2 April.

Amin, Sajeda (2007). Empowering Adolescent Girls in Rural Bangladesh: Kishori
      Abhijan. Transitions to Adulthood Brief No. 13. New York: The Population
      Council.

Amin, Sajeda (2008). Enhancing the Benefits of Girls’ Livelihood Initiatives.
      Transitions to Adulthood Brief No. 17. New York: The Population Council.

Amin, Sajeda, Ian Diamond, Ruchira T. Naved, and Margaret Newby (1998).
      “Transition to Adulthood of Female Garment-Factory Workers in
      Bangladesh.” Studies in Family Planning 29(2):185-200.

Arizpe, Lourdes and Josefina Aranda (1981). The "Comparative Advantages" of
      Women's Disadvantages: Women Workers in the Strawberry Export
      Agribusiness in Mexico. Signs 7(2): 453-473

Attanasio, Orazio, Adriana Kugler, and Costas Meghir (2008). Training
      Disadvantaged Youth in Latin America: Evidence from a Randomized Trial.
      NBER Working Paper 13931. Cambridge, MA: National Bureau of Economic
      Research.

Betcherman, Gordon, Martin Godfrey, Susana Puerto, Friederike Rother, and
      Antoneta Stavreska (2007). A Review of Interventions to Support Young
      Workers: Findings of the Youth Employment Inventory. Social Protection
      Discussion Paper No. 0715. Washington, DC: The World Bank.

Braunstein, Elissa (2000). Engendering Foreign Direct Investment: Family
     Structure, Labor Markets and International Capital Mobility. World
     Development 28(7): 1157-1172.


                                         38
Brewer, Laura (2004). Youth at Risk: The Role of Skills Development in Facilitating
     the Transition to Work. Skills Working Paper No. 19. Geneva: International
     Labour Office.

Buvinic, Mayra (1998). Women in Poverty: A New Global Underclass. N° WID-101.
      Washington, D.C.: Inter-American Development Bank.

Calves, Anne-Emmanuele and Bruno Schoumaker (2004). Deteriorating Economic
      Context and Changing Patterns of Youth Employment in Urban Burkina
      Faso: 1980-2000. World Development 32(8): 1341-1354.

Corbanese, Valli’ and Gianni Rosas (2007). Young people’s transition to decent
     work: Evidence from Kosovo. Employment Policy Paper 2007/4. Geneva:
     International Labour Organization.

Diaz, Juan Jose and Miguel Jaramillo (2006). An Evaluation of the Peruvian “Youth
       Training Program” – PROJOVEN. Office of Evaluation and Oversight
       Working Paper OVE/WP-10/06. Washington, DC: Inter-American
       Development Bank.

El Zanaty and Associates (2007). School-to-work transition: Evidence from Egypt.
      Employment Policy Paper 2007/02. Geneva: International Labour
      Organization.

Erulkar, Annabel, and Erica Chong (2005). Evaluation of a Savings and Micro-
      Credit Program for Vulnerable Young Women in Nairobi. New York: The
      Population Council.

Erulkar, Annabel, Judith Bruce, Erica Chong, Aleke Dondo, Jennifer Sebstad,
      James Matheka, Arjmand Banu Khan, and Anne Gathuku (2006). Tap and
      Reposition Youth (TRY): Providing Social Support, Savings, and Microcredit
      Opportunities for Young Women in Areas with High HIV Prevalence.
      Transitions to Adulthood Brief No. 15. New York: The Population Council.

Esim, Simel, Anju Malhotra, Sanyukta Mathur, Guadalupe Duron, and Charlotte
      Johnson-Welch (2001). Making It Work: Linking Youth Reproductive Health
      and Livelihoods. Washington, DC: International Center for Research on
      Women.

Fares, Jean, Claudio E. Montenegro, and Peter F. Orazem (2006). How Are Youth
      Faring in Labor Markets? Evidence from Around the World. World Bank
      Policy Research Working Paper 4071.

Fawcett, Caroline S. and Sarah Howden (1998). Gender Issues in Technical
     Training and Vocational Education Programs. WID-103. Washington, DC:
     Inter-American Development Bank.
                                        39
Glick, Peter and François Roubaud. (2006). Export Processing Zone Expansion in
       Madagascar: What are the Labour Market and Gender Impacts? Journal of
       African Economies 15(4):722-756.

Godfrey, Martin (2003). Youth Employment Policy in Developing and Transition
      Countries – Prevention as well as Cure. Social Protection Discussion Paper
      No. 0320. Washington, DC: The World Bank.

Griffith, Jean (2002). To Handle Life's Challenges: A Tracer Study of Servol's
       Adolescent Development Programme in Trinidad. The Hague: Bernard van
       Leer Foundation.

Hardoy, Ines (2005). Impact of Multiple Labour Market Programmes on Multiple
     Outcomes: The Case of Norwegian Youth Programmes. Labour 19(3): 425-
     467.

Hettige, S.T., Markus Mayer, and Maleeka Smith, eds. (2004). School-to-work
      transition of youth in Sri Lanka. Employment Strategy Paper 2004/19.
      Geneva: International Labour Organization.

Ibarraran, Pablo, David Rosas and Yuri Soares (2006). Impact Evaluation of a
      Youth Training Program in the Dominican Republic: Ex-Post Project
      Evaluation Report of the Labor Training and Modernization Project
      (DR0134). Office of Evaluation and Oversight OVE/EPPER-03/06.
      Washington, DC: Inter-American Development Bank.

ILO Gender Promotion Programme (n.d.). Report of Survey on the School-to-Work
     Transition of Young Women and Men in Vietnam. GENPROM Series on
     Gender in the Lifecycle. Geneva: International Labour Office.

International Labour Office (2006). Global Employment Trends for Youth. Geneva:
      ILO.

International Labour Office (2007). Key Indicators of the Labor Market Fifth
      Edition. Geneva: ILO.

International Labour Organization (2008). Youth Employment: Breaking Gender
      Barriers for Young Women and Men. Geneva: ILO.

Kalyanwala, Shveta (2007). Influencing Girls’ Lives: Acceptability and Effectiveness
     of a Livelihoods Skill Building Intervention in Gujarat. Transitions to
     Adulthood Brief No. 18. New York: The Population Council.

Levine, Ruth, Cynthia Lloyd, Margaret Greene, and Caren Grown (2008). Girls
      Count: A Global Investment and Action Agenda. Washington, D.C.: Center
      for Global Development.
                                        40
Mensch, Barbara S., Monica J. Grant, Mary P. Sebastian, Paul C. Hewitt, and Dale
     Huntington (2004). The Effect of a Livelihoods Intervention in an Urban
     Slum in India: Do Vocational Counseling and Training Alter the Attitudes
     and Behavior of Adolescent Girls? Policy Research Division Working Paper
     No. 194. New York: The Population Council.

Newman, Constance (2002). Gender, time use, and change: The impact of the cut
    flower industry in Ecuador. The World Bank Economic Review 16(3): 375-
    396.

Ñopo, Hugo, Miguel Robles, and Jaime Saavedra (2007). Occupational Training to
      Reduce Gender Segregation: The Impacts of ProJoven. Research Department
      Working Paper #623. Washington, DC: Inter-American Development Bank.

Okojie, Christiana E.E. (2003). Employment Creation for Youth in Africa: The
      Gender Dimension Paper prepared for the Expert Group Meeting on Jobs for
      Youth: National Strategies for Employment Promotion, Geneva, Switzerland

Parrado, Emilio A. (2002). Socioeconomic Context, Family Regimes, and Women’s
      Early Labor Market Experience: The Case of Colombia and Venezuela. World
      Development 30(5): 799-816.

Ritchie, Amanda, Cynthia B. Lloyd, and Monica Grant (2004). Gender Differences
      in Time Use Among Adolescents in Developing Countries: Implications of
      Rising School Enrollment Rates. Policy Research Division Working Paper
      No. 193. New York: The Population Council.

Ruiz Abril, Maria Elena (2007). Girls’ Vulnerability Assessment. Background
      study for “Economic Empowerment of Adolescent Girls in Liberia.”
      Washington, D.C.: The World Bank.

Sziraczki, Gyorgy and Annemarie Reerink (n.d.) Report of Survey on the School-to-
      Work Transition of Young Women and Men in Indonesia. GENPROM
      Working Paper No. 14 Series on Gender in the Lifecycle. Geneva:
      International Labour Office.

Weller, Jurgen (2007). Youth Employment: Characteristics, Tensions and
      Challenges. CEPAL Review 92: 61-82.

World Bank (2007a). World Development Report 2007: Development and the Next
      Generation. Washington, D.C.: The World Bank.

World Bank (2007b). Improving Opportunities for Sustainable Youth Employment
      in Sierra Leone. Environmentally and Socially Sustainable Development
      Unit, West Africa. Washington, DC: The World Bank.
                                        41
World Bank (2007c). Economic Empowerment of Adolescent Girls. Project
      Identification Mission Aide-Mémoire.




                                      42
Endnotes
1
  The International Labor Organization defines youth as the 15 to 24 age group, as this is the widely accepted
statistical convention. Differences continue to exist, however, in the way many national statistics programs define
and measure youth (ILO 2006).
2
  Fares et al. (2006; Appendix Table II) calculate a statistically significant correlation coefficient in a cross-country
sample of .81 for female home status and jobless rates; for young men, the comparable correlation coefficient is .63.
3
  Baseline values for some characteristics, such as parents’ literacy, own age and education, and household assets,
are comparable across the control and experimental sites. Because of residential segregation, differences are evident
in the religious make up of the samples (the experimental sites having a higher percentage of Muslims and the
control sites a larger percentage of Hindus) and caste status of households (a higher percentage of schedule caste
households in the control sites) (Mensch et al. 2004, Table 3).
4
  It is unclear if the definition of “working for pay” includes self-employment.




                                                           43