Pakistan Policy Note—Enhancing labor market conditions for vulnerable groups 79571 Pakistan Policy Note 7 Cem Mete and Rehan R. Jamil 1 Enhancing Labor Market J une 2 0 1 3 Conditions for Vulnerable Groups For the current generation of working-age adults, public underemployment is high. Many workers work interventions to increase female employment will require part-time or contractually in the informal sec- policies that address the social, cultural, and security tor, which makes up 56–73 percent of nonag- constraints facing urban and rural women in entering ricultural employment (Cnobloch and Salam labor markets. In the long term, targeted investments 2012).1 The country’s labor force participation to improve female education and health can enable rates for women are among the world’s lowest, more women to secure higher paying jobs. Child labor, at 28  percent, compared with 82  percent for however, will require a more strategic approach, going men (World Bank 2013). Household survey beyond legislation to eliminating the worst forms of evidence reveals that many women and youths child labor and increasing school enrollment for poor from low-income households find temporary and vulnerable households through targeted cash trans- and low-paying employment in the informal fers. Long term, child labor can be eliminated only by sector, with minimal job security. investments in human capital; short term, by incentives and penalties for private and informal employers. Pakistan’s population is young—and will remain so in Because Pakistan has so many informal work- the short term. Total fertility rate, at 4.1, remained THE WORLD BANK GROUP SOUTH ASIA REGION ers, skills-enhancing policies are needed to allow well above the South Asia average of 2.8 in 2006– them to graduate into better paid formal jobs and to 07 (Figure  1). Although the total fertility rate ensure that they receive basic social protection. Pro- declined from 7 in 1970 to 3.9 in 2009, it is still grams providing market-oriented skills hold promise, high. The youth cohort (ages 15–24) and adoles- though they need to be evaluated carefully before being cent cohort (ages 10–14) are the most populous implemented on a large scale. Active labor market segments of the population (World Bank 2011a). programs such as short-term public works programs Investing in youth education and health will that hire seasonal laborers, might also be considered thus be crucial for the future competitiveness of for vulnerable workers, primarily postcrisis. labor markets and other social outcomes. With- out these investments in health and education, As Pakistan seeks high and sustainable economic especially for girls, the economy’s competitive- growth, increasing employment and productivity will ness will be seriously compromised. And instead be crucial. The labor force is poorly educated, of a “demographic dividend,� Pakistan will face a highly segmented by gender, and concentrated “demographic bust� (World Bank 2011a). in low value-added informal sector activities. It thus presents a serious constraint on economic Pakistan’s low human development indicators growth and poverty reduction. Although unem- undermine its labor force productivity and eco- ployment is relatively low at around 5 percent, nomic growth. In 2009, the country’s public Pakistan Policy Note—Enhancing labor market conditions for vulnerable groups Figure Selected South Asia fertility rates, 2006 – 07 1 5 Fertility rate, total (births per woman) 4 3 2 2 1 0 Pakistan Bangladesh India South Asia average Source: Data for Pakistan are from MEASURE DHS (2007); all other data are from the World Development Indicators database. spending on education as a share of GDP was third is high levels of child labor and uneasy 2.7  percent and on health 2.2  percent—plac- transition of youths from school to work. ing it among the world’s lowest spenders on these two sectors (World Bank 2012b). This Low formal female employment needs to change. Improvements in education and high underemployment and health enhance labor outcomes, such as employment rates and productivity, which stim- Most women entering the labor force work in agricul- ulate economic growth. Cnobloch, Jamil, and ture, often without pay. Of the more than 3 mil- Chowdhury (2012) use cross-country data to lion jobs for women created between 2001/02 document the links between selected human and 2005/06, 72 percent were in unpaid fam- development outcomes and GDP over the last ily work in agriculture, and only 19  percent two decades. They draw two main conclu- were paid nonfarm jobs (World Bank 2011a). sions: the projections of how human develop- Female unpaid family workers in agriculture ment indicators might evolve under different typically live in rural areas, work on family- economic growth scenarios in the short term owned farms, or work as sharecroppers. They indicate that even under the best-case scenario also have the traditional responsibility of tak- Pakistan would not catch up with the human ing care of household chores and childrearing, development outcomes of most developing beyond their agricultural work. countries; and if the short-term human devel- opment projections were to be realized, long- There is a wide urban–rural gap among women term economic growth would suffer. In other entering the workforce. In rural areas, nearly 1 in 3 words, without major government reforms and women work, against only 1 in 12 in urban areas investments in health and education (thus devi- (Figure 2). Although women in urban areas are ating from the trends of the last two decades), better educated than their rural counterparts, Pakistan will not only miss its Millennium their employment rates are much lower. Urban Development Goal targets but also hinder its women’s labor participation was just 9 percent long-term economic growth potential. in 2001/02, and by 2007/08 had declined to 7.6 percent. The low rate of high-skilled women Policy Issues entering the workforce is reflected in their min- imal representation in senior or managerial The key policy issues fall under three headings. First positions (only 6 percent; Nikitin 2011). is low formal female employment and high underemployment. Second is lack of basic social Although female labor participation rates are low protection and limited access to education and across South Asia, Pakistan’s rates are among training among informal sector workers. And the lowest (IFC 2007). Bangladesh, India, and Figure Labor force participation, ages 18 –59, 2001– 07 2 Rural men Urban men Rural women Urban women 100 75 Percent 50 3 25 0 2001 2003 2005 2006 2007 Source: Nikitin 2011. Pakistan, which together account for 95  per- Unemployment is low overall but far higher for urban cent of the region’s working-age population, women. Despite the decline from 12  percent also have the lowest rates of female participa- to barely 9  percent over 2001–07, unemploy- tion: 31  percent for Bangladesh, 30  percent ment among urban women remained three for India, and 22 percent for Pakistan (World times that of rural women and more than four Bank 2011b). By province, working-age female times that of urban men (Figure 3). Similarly, employment rates are much higher in rural unemployment among women ages 15–24, areas than in urban areas (Table 1). at 10.5  percent in 2008, was far higher than Table Working-age female employment, by province, 2007/08 1 (percent) Punjab Sindh Khyber Pakhtunkhwa Balochistan Urban, all 10 5 6 5 Urban, paid 8 4 5 3 Rural, all 31 27 18 12 Rural, paid 11 1 6 1 Source: World Bank staff estimates based on Pakistan Bureau of Statistics (2008). Figure Unemployment, ages 18 –59, 2001– 07 3 Rural men Urban men Rural women Urban women 15 10 Percent 5 0 2001 2003 2005 2006 2007 Source: Nikitin 2011. Pakistan Policy Note—Enhancing labor market conditions for vulnerable groups among young men, at 7.0  percent (United to them. In addition to house work, limited Nations Statistics Division n.d.). formal arrangements for child day care and elderly care have detrimental effects on female Low schooling levels for women limit their earn- employment opportunities. Married women ings potential. Because women’s educational are expected to look after both their children opportunities are lower than men’s (in both and their parents-in-law. The uneven division of urban and rural areas), women entering the household labor makes the opportunity for full- labor force tend to be less educated and lack time employment very difficult for many women, specialized skills. Women are much less likely particularly poor women who cannot afford the 4 than men to have secondary or higher educa- costs of day care even when it is available. Work- tion. For Pakistani men, education is positively ing women are often expected to maintain their associated with entering the labor force; for household responsibilities, which has a negative women, however, education at the primary and impact on their well-being (Kabeer 2008). lower secondary levels is associated with lower labor force participation. Only higher educa- Travelling long distances for work is not always fea- tion (at the college level or beyond) is positively sible for women. Despite wide regional and class correlated with female labor force participa- variation, the social tradition of veiling and tion (Nikitin 2011). segregating women from nonfamily men— pardah —restricts certain groups of women’s With minimal education, the majority of rural women access to public spaces in socially conservative entering the workforce typically work in low-paying or areas.2 In rural areas, the closeness of homes to unpaid jobs in agriculture, while urban women work agricultural areas where women work allows a in low-paying or part-time jobs in manufacturing or greater degree of freedom to work, while main- services. A small but growing number of edu- taining the social norms of adhering to pardah. cated urban women are entering the rapidly For many urban women, however, the lack of growing services sector. Overall, however, the easily accessible female-friendly workspaces number of educated professional women enter- severely reduces employment options. ing the workforce remains very small. The sectoral composition of employment and high High fertility rates often mean heavy child care duties rates of unpaid female labor hurt female employment for females. Pakistan presents much evidence outcomes. Most unpaid female workers are in that higher fertility is associated with lower the agricultural sector and live in rural areas: investments in human capital per child. Rural 78 percent of female workers are unpaid. Rank- areas have far higher fertility rates and lower ing a distant second is the wholesale and retail use of contraception than urban areas. The sector, where 44 percent of female workers are availability of labor-intensive jobs in agricul- unpaid (World Bank 2011a). Female unpaid ture makes large families economically viable. family workers tend to have little or no educa- In urban areas, however, the higher cost of tion (lower than grade 1): in 2005, two-thirds child rearing and the demand for education of working women in this education group in labor markets is slowly influencing parental were unpaid family workers, compared with choices on fertility rates, as is greater female lit- only 7 percent of working women with grade 11 eracy (World Bank 2011a). In urban and rural education or higher. Only 20 percent of urban areas, the burden of household work and child workers were unpaid, compared with 68  per- rearing falls on women, especially in poor cent in rural areas (World Bank 2011a). households where domestic help is unavailable. Working men and women display a wide earnings Extended family networks sometimes act as informal gap. A woman with similar qualifications to a social safety nets for working women. In the absence man earns far less. The services sector, often of facilities for these women, such as day care, a large source of employment for skilled extended family support systems often fill the women in developing countries—though rap- void, though not all working women have access ­ rowing—has been unable to attract high idly g numbers of skilled female professionals. A lack policy uncertainty, and courts (in order of impor- of human capital investment in women has tance; World Bank 2012a). These constraints resulted in many women from poor and low- are important both for the formal firms sur- income households taking jobs in the informal veyed and because they suggest reasons why sector, which pay little and do not require high 63–72  percent of workers are employed by levels of skills or education (Nikitin 2011). Hav- informal enterprises (Box 1). ing said that, Pakistanis—both women and men—enjoy respectable returns to education, Annual growth in Pakistan’s total factor productiv- with a 7–12 percent increase in earnings for an ity lags well behind South Asia’s better-performing 5 additional year of schooling (Figures 4 and 5). countries (Figure 6). This outcome is unsurpris- ing given how this highly informal economy Lack of basic social protection and relies on workers with little education and not limited access to education and training always in good health who are vulnerable to among informal sector workers various shocks, constrained by poor infrastruc- ture, and heavily concentrated in agriculture. The top five constraints reported by urban for- mal sector firms3 in Pakistan are tax administra- Many of the large numbers of young people entering tion, electricity, political instability, government the workforce lack the entrepreneurial and technical Figure Inflation-adjusted wage levels, 2001– 07 4 Men (nominal) Men (in ation adjusted) Women (nominal) Women (in ation adjusted) 30 Percent (base year, 2001) 20 10 0 2001 2003 2005 2006 2007 Source: Nikitin 2011. Figure Returns to education, change in earnings for an additional year of schooling, 2001– 07 5 Rural men Urban men Rural women Urban women 15 10 Percent 5 0 2001 2003 2005 2006 2007 Source: Nikitin 2011. Pakistan Policy Note—Enhancing labor market conditions for vulnerable groups Box The informal sector and labor markets 1 To understand Pakistan’s labor markets, it is important to look beyond unemployment rates and consider the quality of employment. Many employed individuals find it difficult to find sufficient or secure employment. Cnobloch and Salam (2012) estimate that 72 per- cent of all workers were in informal enterprises in 2007/08. Using an alternative definition based on type of employment, in which casual, piecemeal, own-account, and unpaid family workers in the nonagricultural sector are in informal employment, informality is an estimated 63 percent, but because informal enterprises tend to be small and concentrated in low value-added activities, their share in national output is only 35–40 percent. Individuals in the informal sector come from the full spectrum of economic backgrounds, but most are uneducated and live in 6 rural areas. Punjab, the most populous province, has the highest number of informal workers. Some workers switch back and forth between the formal and informal sectors. As individuals age, they can transition to the formal sector, so some aspects of informality are compatible with economic mobility. The informal labor market is segmented, especially along the urban–rural split. In urban areas, nearly 40 percent of informal workers come from the top 20 percent of the welfare distribution; in rural areas, only 8 percent do. With fewer opportunities in the formal sector, informality is the next best option for younger people, especially those in urban areas and with less schooling, despite the lower wages. The informal sector also attracts a greater share of poorer women, while men come from a variety of income levels. Source: Nikitin 2011; Cnobloch and Salam 2012. Annual growth in total factor productivity in Pakistan, Thailand, India, and China, Figure various periods 6 Reallocation Services Industry Agriculture 4 Total factor productivity 3 2 1 0 Pakistan Thailand India China China (1980–2008) (1977–96) (1980–2008) (1978–93) (1993–2004) Source: World Bank 2012a. skills that formal firms need, particularly the larger, Formal firms often prefer short-term contractual more specialized firms. The average worker has arrangements with workers. Short-term hiring is a only three or four years of schooling. The prob- common trend in Pakistani labor markets, with lem is especially acute in manufacturing, where an average temporary contract of around five two-thirds of employees have fewer than three months. Of manufacturing employees, 30 per- years of schooling. In services, half the workers cent have short-term employment arrangements, are educated above grade 6 (Nikitin 2011). or more than double India’s rate (13 percent) and about Sri Lanka’s (28 percent; Nikitin 2011). An International Finance Corporation survey of small and medium-size firms in Pakistan found that High levels of child labor and uneasy less than 6 percent of the youths surveyed had any transition of youths from school to work technical or vocational skills. Only 2.5  percent reported on-the-job-training, and most of the Child labor remains widespread in Pakistan, despite new low-skilled labor is absorbed by the infor- some success in reducing it in certain industries. In mal sector (IFC 2007). 1998, 16.3 percent of children ages 10–14 were working (ILO 1998). In this age group, 73 per- period show that poor households responded cent (2.4 million) of child laborers were boys by cutting spending and, where possible, while just 27 percent (0.8 million) were girls. expanding adult and child labor participation Children in rural areas were eight times more (Nikitin 2011). Poor households that reported likely to be economically active than those in shocks in 2009/10 saw higher student drop- urban areas. About 71 percent of working chil- out rates—either to enter labor markets or to dren were involved in unskilled occupations reduce household expenditures. The drop in in agriculture and manufacturing, and about spending on education was far higher for girls 70  percent of working children were unpaid than boys. Thus, even when returns to educa- 7 family helpers (75 percent in rural areas and tion are high, poor households may not invest less than 33  percent in urban areas). More enough in schooling their children (World recent work confirms the high prevalence of Bank 2011c). child labor, estimated at 13.7 percent for ages 5–14 in 2004 (ILO 2006). U.S. Department of A girl is much less likely to enroll in school if her Labor (2002) statistics further show that 13 per- mother had not attended school or if her father works cent of the children worked 56 or more hours in agriculture (Lloyd, Mete, and Sattar 2005). For a week, and about 7 percent suffered from fre- both sexes, the levels of community develop- quent illnesses or injuries. ment and opportunities have a positive impact on primary-school enrollment: in communities Despite government commitment to outlaw child that have many schools, parents are more likely labor, implementation lags far behind. In its to invest in education. This makes investment national strateg y to combat child labor, in schools in remote areas with few public or the government has committed to outlaw private schools even more critical. child labor by focusing on eliminating the worst forms of child labor prevalent in cer- Pakistan’s youth employment challenge would tain industries and by expanding education already be addressed to a large extent if the coun- opportunities at the primary and secondary try made progress in tackling the female employ- levels, to keep children out of the workforce. ment and informal sector employment. Women Yet a large informal economy enables employ- between ages 15 and 24 also explain a large ers to easily employ children for minimal share of those who are not in education pay and without record of their employment employment or training (ILO 2012). The or need to adhere to government regula- share of women not in education employ- tions. From a demand perspective, children ment or training in Pakistan was 66.4 percent remain a cheap supply of labor, working for in 2005, the second highest among develop- paltry amounts with few benefits. From a sup- ing countries, compared with 10.9 percent of ply perspective, poverty remains the primary men, better than other countries. The qual- reason why poor households have children ity of employment also matters. For informal work instead of attending school. Deficien- sector employment, large numbers of youths cies in enforcing regulations and high infor- entering the workforce lack the entrepre- mal employment thus limit the effectiveness neurial and technical skills that formal firms of policies to reduce child labor.4 need. Credit constraints and an uncertain future eco- The main reasons for not working for youths also nomic environment serve as disincentives for poor differ significantly between men and women. For households to invest in their children’s schooling, men, even if one focuses only those between which can make child labor a more attractive option. ages 20 to 24, 70 percent do not work because In 2008–10, households experienced several they are students and not willing to work. But severe external shocks, such as spikes in food for women, only 10 percent do not work for prices in 2008 and floods in 2010 (World Bank these reasons. Instead, more than 88 percent 2010). The Pakistan Social and Living Stan- of women indicated housekeeping as the main dards Measurement Survey data from this reason for not working. Pakistan Policy Note—Enhancing labor market conditions for vulnerable groups Policy Recommendations Pakistan’s flagship social safety net program, has taken the lead by targeting cash transfers toward The policy recommendations for improving the labor poor women. The program is piloting poverty market outcomes of the most vulnerable groups are graduation programs, including skills building summarized under three subheadings. Many of and access to microfinance programs for ben- them are complementary and will require an eficiary households. Evidence from other coun- integrated approach that specifically targets tries suggests that skills programs that include the employment and social protection needs of an employment-matching component can help poor and vulnerable workers, a large portion of graduates find jobs. In Latin America, for 8 them women and children. example, a number of youth employment pro- grams, such as Peru Proven (Peru) and Jóvenes Improve conditions for women’s employment en Acción (Colombia), contract with decentral- ized training entities to offer vocational courses Public provision of, or subsidies for, child care can in which beneficiaries can enroll. The courses, reduce the costs that poor women incur at home when offering classroom and on-the-job training, they enter the labor market, particularly in urban match local firms’ needs with the content of the areas. Urban and rural women report house- training curricula. These two programs are suc- hold responsibilities as one the primary reasons cessful because they are well targeted, demand for being unable to seek employment. Although driven, and linked to private labor demand. Not extended families play an important role in all skills programs work as intended, however, providing child care for working women, the and such interventions should first be piloted absence of high-quality public or private day and then evaluated in Pakistan. care facilities in major cities is a large barrier for women’s access to labor markets. Publicly Investments in infrastructure help connect women provided or subsidized day care can be found and vulnerable youths to markets and reduce the time in many Latin American countries, including spent doing household chores. Evidence from the Argentina, Brazil, and Colombia. Colombia’s developing world suggests that improvements in Hogares Comunitarios program provides sub- basic infrastructure services—especially water sidies to designated homes turned into com- and electricity—can help free women’s and chil- munity child care centers, and its provision of dren’s time and increase female labor force par- affordable child care in low-income neighbor- ticipation (World Bank 2012b). Access to water hoods has greatly increased mothers’ participa- closer to households frees up women’s time, tion in the labor market as well as their hours especially in rural areas where women often worked (World Bank 2012b). spend much time collecting water. A study from Pakistan showed that water sources closer to Legal reform to reduce gender discrimination in the home were associated with increased time allo- workplace is necessary to increase female employ- cated to labor market work (Ilahi and Grimard ment rates—but not sufficient. Affirmative action 2000). Similarly in Bangladesh, upgrading and programs to increase the entry of women into expansion of rural roads led to increased labor wage employment and their advancement on supply for men and women, while raising house- the job once they are employed is one way to hold incomes (Khandker, Bakht, and Koolwal increase the hiring of women. Women’s pres- 2006). Finally, transport facilities targeted at ence in public sector jobs remains low, and the working women, such as buses to take them public sector can take the lead in affirmative directly to and from their workplaces, can foster action—with incentives such as tax breaks for a safe and convenient environment to encour- private firms that follow this example. age women to find employment. Skills development and microfinance focusing on Implement measures to support urban female workers and urban youths to increase informal workers their assets might be scaled up after due documenta- tion of their impacts on employment outcomes. The Pilot skills programs that target informal work- Benazir Income Support Programme (BISP), ers need to be evaluated. Public skills training programs in Pakistan, as across South Asia, countries. In some cases, the poor hold assets, generally favor formal workers. Such programs but these assets represent “dead capital� have had limited success, in part because they because they are held in forms not recognized reach very few workers and cannot match the by most legal systems and cannot be used to skills demanded by job markets. Successful generate productive capital that the poor could skills training programs are integrated pro- then use for formal economic activities (de Soto grams that include both on-the-job and class- 2003). Legal reforms to grant property rights room components (World Bank 2013). As many to informal property owners and business ven- poor and informal workers have little educa- dors must be backed up by other services, such 9 tion, courses will need to be tailored to their as access to microfinance and streamlined and needs. accessible government services for registering property and obtaining national identity cards. In addition, skills or vocational training provid- ers need incentives to ensure that the training they A targeted social insurance scheme should be consid- offer is relevant for employers. Publicly subsidized ered for informal workers. Most workers in Paki- training agencies are often out of touch with stan, particularly in the informal sector, do not the changing demands and preferences of both participate in a social insurance program. Pen- firms and job seekers. An alternative approach sions are largely restricted to a small number of is for public training funds to be directed to government employees and full-time employees private and nonprofit providers on a competi- in the formal private sector. Given the fiscal tive basis. Performance-based tendering can constraints, piloting a program targeting poor create incentives for more relevant training and vulnerable workers who receive no social courses, while contracting can be designed so insurance would be an important step toward that the toughest-to-reach groups do not lose an integrated social protection system. And with out (World Bank 2013). high costs of health care and medicine, along with the detrimental impacts of health shocks Active labor market programs, such as public works on poor households’ incomes, health coverage programs, can help provide employment for poor should be an important part of any compre- informal workers, such as laborers, particularly in hensive social insurance program. The poverty the offseason. The cost effectiveness of existing database developed in 2011 by the National large-scale public works programs in South Database and Registration Authority and BISP Asia is debatable, but such programs that are provides a readily available information source well designed and implemented can mitigate for the design of a well-targeted program, which the negative impacts of economic downturns could be an extension of the current programs and fill the gap when employers or workers being piloted by BISP. Finally, although pub- themselves underinvest in training. Public licly financed noncontributory programs are works programs can also promote skills build- the most common social insurance programs ing, if vocational training is built into their in developing countries, subsidized social design. Such programs can also facilitate com- insurance programs (including government munity participation, particularly in postcrisis matching programs that double workers’ contri- areas with high unemployment but requiring butions) can help ensure the financial viability labor for reconstruction and infrastructure. of the noncontributory ones (Box 2). Geographically targeted districts in the Feder- ally Administered Tribal Areas, Khyber Pakh- Address high levels of child labor and facilitate tunkhwa, and Balochistan may be suitable for youths’ transition from school to work well-designed small public works programs to help with reconstruction. An important first step toward reducing child labor is eliminating the worst forms of employment hazardous Pakistan needs laws to recognize the rights and to children. The International Labour Organiza- assets of informal workers. Laws governing prop- tion defines the “worst forms of child labor� as erty rights and the titling of assets of the poor any work that jeopardizes the health, safety, or are a cause of informality in many developing morals of a child. Such work is determined to be Pakistan Policy Note—Enhancing labor market conditions for vulnerable groups Box The Self-Employed Women’s Association’s Integrated Insurance Scheme, India 2 The Self-Employed Women’s Association (SEWA) runs the largest comprehensive contributory social protection scheme for informal workers in India. One-third of the premium is financed through interest paid on a grant provided by the Deutsche Gesellschaft für Technische Zusammenarbeit, one-third through direct contributions by women workers, and one-third through a subsidized package provided by the Life Insurance Corporation of India and the United India Insurance Company. SEWA members can choose whether to participate in the scheme, which covers insurance for health (including a maternity benefits component), life (death and disability), and assets (loss or damage to housing unit or work equipment). SEWA has also designed the payment of premiums to suit different income groups among the very poor. 10 Source: Lund and Srinivas 2000. Sialkot soccer ball industry: a model for public–private partnerships to reduce child Box labor 3 The city of Sialkot in Punjab has traditionally been one of the world’s major suppliers of hand-stitched soccer balls, and for many years child labor was endemic. Under an agreement between the International Labour Organization and the soccer ball manufacturers in 1997, children younger than 14 years were withdrawn from workplaces and, with their younger siblings, directed toward educa- tional alternatives. Working with local communities, partners with the International Programme on the Elimination of Child Labour set up nonformal primary schools and vocational training centers for older children while mainstreaming others into formal school systems where possible. They also provided the children with health and social protection services and supported their families through income-generating activities. Source: IPEC 2009. hazardous depending on its specific nature, the average child receiving the transfer was demands on children in particular industries, 10 percentage points less likely to work for pay and the general working conditions (ILO 2007). (Filmer and Schady 2006). Industries such as bricklaying, recycling, and car- CCT programs must be supported by rigorous moni- pet weaving are the most dangerous for children. toring and evaluation to ensure that children stay in Enforcing strict regulation of these industries, school and do not return to work. A pilot CCT pro- including fines, is one way to ensure underage gram implemented by Pakistan Bait-ul-Maal children are not hired. Partnerships with pri- over 2008–2010 and evaluated by the World vate employers to help provide safety nets and Bank demonstrated promising results, leading education facilities for laid-off children is criti- to a 12  percentage point increase in primary cal to ensure that households can gain alterna- school enrollment for girls and a 9 percentage tive incomes without sending their children point increase for boys (Scott and others 2012). to work (Box 3). Public policies that promote BISP, having learned from the pilot, is now part-time and informal methods of teaching, rolling out a national CCT program (at the such as evening classes, can help older teenag- primary school level) as a top-off to the main ers from poor households engaged in part-time unconditional cash transfer. Once the CCT work complete their secondary education. program has been successfully implemented at primary schools nationally, current second- Conditional cash transfers (CCTs) might in the long ary-school provincial interventions can be run be the most effective way to reduce child labor, by supplemented by additional incentives to poor creating incentives and safety nets that allow former children (as defined by the BISP database) for and would-be child workers to go to school. Impact attending school. evaluations of programs that offer CCTs to combat child labor, such as those in Brazil, Pakistan’s youth employment challenge is much Colombia, Ecuador, and Mexico, have led to broader than the unemployment challenge of higher substantial decreases in child labor. In Cam- educated youths. Indeed, even though higher bodia—an example of a large reduction—the educated youths are most visible to policy makers and tend to voice their concerns more child labor are widespread and which than other youths, high unemployment rates of attracts some of the country’s poorest com- educated youths are due partly to (high) reser- munities (Collective for Social Sciences vation wages that they maintain before accept- Research 2004). ing jobs. Rates of returns to education remain respectable in Pakistan and thus the educated References youth tend to benefit from schooling in the long Cnobloch, Rodica, Rehan R. Jamil, and Afra run. R. Chowdhury. 2012. “How Are the MDGs Likely to Evolve in the Short Term and Why 11 The policy interventions that will improve youth Status Quo Is a Bad Outcome for Pakistan.� employment trends in Pakistan will be those that focus Unpublished manuscript. World Bank, on women and on less educated youths who often have Washington, DC. to work in the informal sector, simply because these Cnobloch, Rodica, and Naveeda Salam. 2012. groups overwhelmingly make up those who are either “Informal Employment: The Who, and How outside the labor force or occupied in low-productivity They Are Spending Their Time.� Unpub- and low-pay jobs. As a result, in addition to edu- lished manuscript. World Bank, Washington, cation policies covered in accompanying policy DC. notes, the recommended policies to increase Collective for Social Science Research. 2004. women’s employment and to support infor- “A Rapid Assessment of Bonded Labour in mal sector workers emerge as key priorities to Hazardous Industries in Pakistan: Glass Ban- address Pakistan’s youth employment challenge. gle-Making, Tanneries and Construction.� Working Paper 21. Karachi. Notes de Soto, Hernando. 2003. The Mystery of Capital. 1. Sectors other than agriculture make up New York: Basic Books. 58 percent of total employment. Filmer, Deon, and Norbert Schady. 2006. “Get- 2. Pardah — often translated as female ting Girls into School: Evidence from a ­ seclusion—is a contested term in Pakistan, Scholarship Program in Cambodia.� Policy as it can imply a variety of forms of veiling Research Working Paper 3910, World Bank, practiced in Pakistan—ranging from wear- Washington DC. ing a dupatta, chadar, or burqa to segregat- IFC (International Finance Corporation). 2007. ing women. “Enterprise Survey: Pakistan Country Pro- 3. Firms that keep accounts, pay taxes, and file.� Washington DC. use courts. Ilahi, Nadeem, and Franqui Grimard. 2000. 4. The National Assembly passed several “Public Infrastructure and Private Costs: important laws in the 1990s banning Water Supply and Time Allocation of or restricting child labor. The Employ- Women in Rural Pakistan.� Economic Develop- ment of Children Act of 1991 prohibited ment and Cultural Change 49 (1): 45–75. employment of children younger than ILO (International Labour Organization). 14 years in factories and mines or in any 1998. Child Labour Survey in Pakistan. hazardous environment (ILO 2009). The Islamabad. Bonded Labor Act of 1992 in theory abol- ———. 2006. “The End of Child Labour: ished indentured servitude and the tradi- Within Reach.� Global Report under the tional  peshgi  system. Peshgi is a form of Follow-Up to the ILO Declaration on Funda- cash advances to employees, which in its mental Principles and Rights at Work. Inter- worst form is bonded labor, as it allows national Labour Conference, 95th Session, employers to retain workers if they can- Geneva. not repay a cash advance. Despite being ———. 2007. “Child Labor and Conditional outlawed, it is still common in the infor- Cash Transfer Programmes in Latin Amer- mal sector. Peshgi is practiced routinely in ica.� Geneva. the brick-kiln, bangle-making, and leather ———. 2009. “National Legislation and Policies industries, in which labor violations and against Child Labour in Pakistan.� Geneva. Pakistan Policy Note—Enhancing labor market conditions for vulnerable groups ———. 2012. Global Employment Trends for Youths. Pakistan Bureau of Statistics. 2008. Labour Geneva. Force Survey 2007/08. Government of Paki- IPEC (International Programme on the Elimi- stan, Islamabad. nation of Child Labour). 2009. “Football for Scott, Kinnon, Rodica Cnobloch, Xiaohui Hope in Sialkot: A Football Lifeline for Chil- Hou, and Cem Mete. 2012. “Impact Evalua- dren from Football Stitching Communities.� tion Results of a Pilot CCT Program in Paki- IPEC News Volume 1, March. stan.� Unpublished manuscript. World Bank, Kabeer, Naila. 2008. “Gender, Labour Markets Washington DC. and Poverty: An Overview.� Poverty in Focus: United Nations Statistics Division. n.d. Millen- 12 Special Issue on Gender Equality (13): 3–5. nium Development Goals database. Updated Khandker, Shahidur R., Zaid Bakht, and Gayatri September 12, 2012, http://mdgs.un.org/ Koolwal. 2006. “The Poverty Impact of Rural unsd/mdg/Default.aspx. Roads: Evidence from Bangladesh.� Policy U.S. Department of Labor. 2002. “2001 Find- Research Working Paper 3875, World Bank, ings on the Worst Forms of Child Labor— Washington, DC. Pakistan.� Washington, DC. Lloyd, Cynthia, Cem Mete, and Zeba Sattar. 2005. World Bank. 2010. “Pakistan Floods 2010: Pre- “The Effect of Gender Differences in Primary liminary Damage and Needs Assessment.� School Access, Type, and Quality on the Deci- Washington, DC. sion to Enroll in Rural Pakistan.� Economic Devel- ———. 2011a. “Pakistan: Poverty Assessment.� opment and Cultural Change 53 (3): 685–710. Washington, DC. Lund, Francie Jane, and Smita Srinivas. 2000. ———. 2011b. South Asian Development Matters. Learning from Experience: A Gendered Approach Series. Washington, DC. to Social Protection for Workers in the Informal ———. 2011c. “Social Safety Nets in Pakistan.� Economy. Geneva: International Labour Washington, DC. Organization. ———. 2012a. “More and Better Jobs in South MEASURE DHS. 2007. Pakistan Demographic Asia.� Washington, DC. and Health Survey 2006/07. ICF Interna- ———. 2012b. World Development Report 2012: tional, Fairfax, VA. Gender Equality and Development. Washington, Nikitin, Denis. 2011. “Pakistan: Labor Market DC. Trends.� Unpublished manuscript. World ———. 2013. World Development Report 2013: Bank, Washington, DC. Jobs. Washington, DC. © 2013 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street NW Washington, DC 20433 USA All rights reserved This report was prepared by the staff of the South Asia Region. The findings, interpretations, and conclusions expressed herein are those of the authors and do not necessarily reflect the views of the World Bank’s Board of Executive Directors or the countries they represent. The report was designed, edited, and typeset by Communications Development Incorporated, Washington, DC.