Urban Safety Nets and Activation in Mozambique JUNE 2017 i Urban Safety Nets and Activation in Mozambique June 2017 Eric L. Zapatero, Zaineb Majoka, Jordi Gallego-Ayala, Michele D. Zini, Peter Beck & Mohamed Ihsan Ajwad. Human Development Department Social Protection Unit Africa Region i TABLE OF CONTENTS ACKNOWLEDGMENTS ......................................................................................................................................... v ABBREVIATIONS.................................................................................................................................................. vi EXECUTIVE SUMMARY......................................................................................................................................... 7 Introduction and Rationale ............................................................................................................................... 12 Chapter 1: Profile of Mozambique’s Poor Urban Population............................................................................ 15 1 Data ............................................................................................................................................................. 16 1.1 Household survey - IOF 2008-09 ............................................................................................................. 16 1.2 Qualitative Data ...................................................................................................................................... 17 1.3 Administrative – PASP Registry Data ...................................................................................................... 17 2 Socio-economic Profile of the Urban Poor in Mozambique ....................................................................... 18 2.1 Urban context ......................................................................................................................................... 18 2.2 Profile of urban poor............................................................................................................................... 18 2.2.1 Main determinants of Urban Poverty .............................................................................................. 25 3 The Productive Social Action Program (PASP) ............................................................................................ 34 3.1 Targeting performance of PASP .............................................................................................................. 35 3.2 Socio-economic Profile of Urban PASP Beneficiaries ............................................................................. 38 Chapter 2: Social Safety Nets and Activation Programs Assessment in Urban Mozambique........................... 44 4 Social Safety Nets and Activation Programs in Urban Mozambique .......................................................... 45 4.1 Scope and context................................................................................................................................... 45 4.2 Legal background .................................................................................................................................... 47 4.3 Institutional and fiscal background ......................................................................................................... 48 4.4 Mapping of urban safety nets and activation programs ........................................................................ 50 4.5 National Level Safety Net Programs ....................................................................................................... 50 4.5.1 The Basic Social Subsidy Program (PSSB) .................................................................................... 52 4.5.2 The Direct Social Action Program................................................................................................ 53 4.5.3 The Productive Social Action Program......................................................................................... 54 4.5.4 Social Services and In-care Placement Programs (PSSAS) ............................................................ 54 4.6 National level Activation Programs......................................................................................................... 54 4.6.1 The Strategic Program for Urban Poverty Reduction (PERPU) .................................................... 55 4.6.2 Skill development programs and access to employment programs offered by INEFP .................. 55 ii 4.6.3 PASP activation measures (ADGIR) ............................................................................................ 55 4.6.4 Other skills development programs............................................................................................ 56 4.7 Municipally Implemented National Programs ........................................................................................ 57 4.8 Municipal level Programs........................................................................................................................ 58 4.9 NGO and Private Sector Run Programs................................................................................................... 59 Chapter 3: Evaluating the Readiness of Safety Nets and Activation Initiatives in Urban Mozambique. Conclusions and Recommendations ........................................................................................................... 63 5 Conclusions ................................................................................................................................................. 64 6 Recommendations ...................................................................................................................................... 68 6.1 Recommendations at the Policy Level .................................................................................................... 68 6.2 Recommendations at the Administrative Level ...................................................................................... 71 6.3 Recommendations at the Program Level................................................................................................ 73 6.3.1 Recommendations for PASP ........................................................................................................ 73 References ......................................................................................................................................................... 76 Annexes Annex 1: Data collection tool for APS Single Registry of Beneficiaries ...................................................... 74 Annex 2: Data collection instruments for the qualitative analysis in Maputo peri-urban areas .............. 78 Annex 3: Skill Training Projects from Non State Agencies.......................................................................... 82 Annex 4: National Level Programs in a Life Cycle Approach ...................................................................... 84 Annex 5: Coverage and Costs of Cash Transfers in Mozambique .............................................................. 88 Figures Figure 1.1: Labor force participation rates ................................................................................................................. 17 Figure 1.2: Employment .............................................................................................................................................. 18 Figure 1.3: Share of temporary employment .............................................................................................................. 18 Figure 1.4: Type of employment for youth (a) and people 25 years and older (b ...................................................... 19 Figure 1.5: Sector of employment for youth (a) and people 25 years and older (b .................................................... 21 Figure 1.6: Educational Attainment among Urban Population in Mozambique ......................................................... 25 Figure 1.7: Average Household Size in Urban Mozambique ....................................................................................... 26 Figure 1.8: Percentage of Households with Children and Elderly ............................................................................... 26 Figure 1.9: Distribution of urban poverty by age cohort in Mozambique ................................................................... 26 Figure 1.10: Gender of Household Head in Mozambique ........................................................................................... 27 Figure 1.11: Marital Status of Urban Population in Mozambique ............................................................................... 27 Figure 1.12: Baseline maps based on 2007 census average PMT scores .................................................................... 30 Figure 1.13: Urban headcount rates by province ........................................................................................................ 30 Figure 1.14: Daily consumption, deflated .................................................................................................................... 30 Figure 1.15: PASP benefit incidence ............................................................................................................................ 34 Figure 1.16: Beneficiary Incidence of Public Works Programs in Urban Areas ........................................................... 34 Figure 1.17. Geographical distribution of PASP beneficiaries in Maputo .................................................................... 35 Figure 1.18: Proportion of population that is workable versus a dependent or disabled ........................................... 36 iii Figure 1.19: Educational Attainment among PASP Beneficiaries ................................................................................ 37 Figure 1.20: Age of Household Heads among PASP Beneficiaries ............................................................................... 37 Figure 1.21: Share of PASP beneficiary Households with dependent ......................................................................... 38 Figure 1.22: Gender of Household Heads of PASP Beneficiaries ................................................................................. 39 Figure 1.23: Marital Status of PASP Beneficiaries........................................................................................................ 39 Figure 2.1: Safety Nets and Activation Programs and Services Through a Life Cycle Lens .......................................... 43 Figure 2.2: Basic Social Security Budget 2011-2015, as a percent of GDP .................................................................. 46 Figure 2.3: Beneficiaries and Executed Budget of Main National Social Assistance Programs, 2012-2016 ................ 50 Figure 2.4: PASP activation measures (ADGIR) ............................................................................................................ 53 Tables Table 1.1: Safety nets beneficiary households ............................................................................................................ 22 Table 1.2: Factors Affecting Poverty in Urban Areas in Mozambique ......................................................................... 23 Table 1.3– Completed Level of Education ................................................................................................................... 24 Table 1.4: Asset ownership of urban households (%) ................................................................................................. 29 Table 1.5: Targeting of urban beneficiaries (avg. PMT scores and percent eligible) ................................................... 35 Table 1.6: Asset Ownership for PASP beneficiaries .................................................................................................... 40 Table 2.1: Targets and Actual Beneficiaries, Allocated and Executed Budgets of Main National Social Programs, 2012-2016............................................................................................................................................................. 48 Table 2.2: PERPU’s Expenditure .................................................................................................................................. 52 Table 2.3: ADGIR program targets in urban PASP region ............................................................................................ 53 Table 2.4: Urban Public and Private Training Provision, 2014 ..................................................................................... 54 Table 2.5: Sources of Foreign Government Support to Activation Sector ................................................................... 57 iv ACKNOWLEDGMENTS This report is co-authored by Eric L. Zapatero (Senior Social Protection Specialist, GSP07), Zaineb Majoka (Consultant), Jordi Gallego-Ayala (Social Protection Specialist, GSP01), Michele D. Zini (Senior Economist, GSP03), Peter Beck (Consultant) and Mohamed Ihsan Ajwad (Senior Economist, GSP03). The report was prepared under the technical guidance of Eric Zapatero. This report and its results were discussed intensively with representatives of the Ministry of Gender, Children and Social Action and the National Social Assistance Institute. The team benefit from comments and observation from technical workshops and consultations with Government of Mozambique institutions, development partners and local NGOs. Invaluable comments were received from colleagues at the World Bank and outside, including: Andrea Vermehren, Ian Walker, Cristina Panascos, Penny Williams, Rodrigo Madariaga, Ugo Gentilini and Mohamed Muderis. The team is also grateful from the support received from Isadora Nouel during the preparation and editing of the report. This report also benefits from the data facilitated by MGCAS, INAS, DFID, INEFP, ESSOR and LIGADA. The team would like to thanks for the funding and support to the Mozambique Country Management Unit (ASA; task ID P156601) and the direct support received from the leadership of Mark Lundell (Country Director AFCS2) and Dena Ringold (Practice Manager GSP01). v ABBREVIATIONS ASPIRE Atlas of Social Protection Indicators of Resilience and Equity CBT Community Based Targeting CCT Conditional cash Transfers CPF Vocational Training Center DFID Department of International Development - UK DINET National Directorate of Technical Education ENSSB Estratégia Nacional de Segurança Social Básica GIZ German Cooperation GoM Government of Mozambique INAS National Institute of Social Action INEFP National Professional Training Institute IOF Mozambique Household Budget Survey MGCAS Ministry of Gender, Children and Social Action MINEDH Ministry of Education NGO Non Government Organization PASD Direct Social Action Program PASP Productive Social Action Program PERPU Strategic Urban Poverty Reduction Program PMT Proxy Means Testing PSSAS Social Services and In-care Placement Programs PSSB Basic Social Subsidy Program KFW German Development Bank UN United Nations USAID United Stated Agency for International Development vi EXECUTIVE SUMMARY Recent economic and political in Mozambique events are resulting in an increased vulnerability in urban areas, which makes a stronger case for safety nets and activation activities to protect the poorest and foster opportunities. Around 3.4 million poor individuals live in urban areas in Mozambique and transient poverty is very significant with a large share of the urban population staying close to the poverty line, making these groups vulnerable to small variation in income. The latest political and economic developments have resulted in a depreciation of the metical by 42 percent against the dollar in 2016, an inflation reaching 20 percent in 2017 and rising food prices by almost 40 percent. This situation could be aggravated by forthcoming fiscal measures that the Government is planning to put in place to improve near and medium term macro-economic prospects. These measures include dismantling fuel and fuel subsidies and other cuts in social expenditure. In addition to the political and economic context, urbanization in Mozambique has been a steady process over the last years and this urbanization dynamic is likely to continue in the next coming years transforming Mozambique into one of the most urbanized countries in the region. This process will potentially exacerbate the unemployment situation in the urban settings, enhancing the vulnerability of the urban poor and specially the urban youth. Despite the high vulnerability context, coverage of safety nets programs in urban areas is very low. Poor urban households have very limited access to safety nets and activation programs, that only cover 92,017 households, which represented around 15 percent of the target population in 2015. Moreover, the way current programs are designed seems to exclude one of two of the main target groups in urban settings, poor families with children and the unemployed youth. Low coverage is the result of a combination of weak technical capacity to rapidly scale up programs in an efficient manner and low overall budget allocation to social assistance, that only represented 0.45 percent of GDP in 2016. In addition to the low coverage and low expenditure, operationalziation of existing safety nets programs is not efficient nor effective. Given the current vulnerability context, the absence of a pure poverty targeted cash transfer program, along the lines of those prevalent in many developing countries, is striking . Many countries are moving towards unconditional cash transfer programs, however, in Mozambique’s case, the absence of a conditional poverty targeted cash transfer to protect households from the risk of falling into poverty is noteworthy. This is especially so if we take into account the country’s low ranking in the Human Development Index. Mozambique seems to be missing a good opportunity to invest in the activation of the next generation. The activation sector is still a work in progress in Mozambique lacking a clear policy and institutional framework and with many small and fragmented initiatives. First, activation initiatives in Mozambique are both fragmented and do not work in an integrated manner with safety nets programs. The Ministry of Gender, Children and Social Action (MGCAS) is in charge of designing and implementing safety nets programs, while the Ministry of Labor and the Ministry of Education are in charge of activation initiatives. Even though coordination bodies such as the National Council for Social Action (CNAS) have been created to promote the coordination between these three Ministries, no actual referral system exists for safety nets beneficiaries to be referred to activation initiatives that could potentially take them out of poverty in a sustainable manner. Activation policies are not particularly developed at the national or local level and the only program directly linked to a safety net and favoring the inclusion of a targeted vulnerable population into the labor market is the activation component of the Productive Social Action Program (PASP), the 7 implementation of which is delayed and still in its initial stages. Professional trainings offered by INEFP’s 16 urban training centers, and the training programs offered by the Ministry of Education are the only sizeable and central level activation measures offered throughout the country. The private professional training sector, which recently has attracted the attention of donors in urban areas, remains small and its ability to absorb and train young and vulnerable school leavers is still a question mark. In addition to being small and fragmented, safety nets and activation programs in urban settings seem not to consider the specific vulnerabilities of poor urban households and the significant differences in labor participation patterns among different age-groups and geographical locations. Urban residents face risks such as protracted underemployment, low income, food price increases, and other financial shocks less commonly felt in rural settings. Despite the high labor force participation, the productivity of poor urban households is very low as the majority of the urban poor are engaged in non-wage employment related to household entreprises, most of these in the agriculture sector and overall with very low levels of productivity. Their level of education is very low with only 31 percent of them having completed secondary school or more, which makes them not elegible for most of the existing activation programs that require at least completion of sixth grade. Urban children are more likely to be poor than the other age cohorts, and there exists a very large number of unemployed urban youth that have a great potential for activation. These two groups are currently excluded from the core safety nets and activation programs in the country. For instance, only 25 percent of PASP beneficiaries in urban settings are between the age of 15 and 24. To effectively mitigate vulnerability and foster opportunity in urban Mozambique, safety nets and activation programs need to be able to work together as a portfolio of complementary initiatives, under a systems approach. Systems are “portfolios� of coherent programs that can communicate with each other, share common administrative subsystems, and work together to respond to risks and to deliver resilience, equity, and opportunity to the population.. A systems approach involves three levels of engagement: the policy level, the administration level and the program level. At policy level, the government needs to ensure policy coherence attuning the entire portfolio of safety nets and activation programs to act coherently in delivering national goals and social contracts. To do that, there is a need to strengthen the role of the national council of Social Action (CNAS) to coordinate social assistance and activation programs. This coordination would be strengthened by promoting additional synergies and complementarities at administration and program level. In fact, the CNAS should act as a more explicit mechanism to promote coordination not only at central level, but with the municipal governments as well, considering that municipalities are a key role players in the implementation of social programs. Government should reduce the fragmentation of safety nets programs and invest in the next generations through targeted cash transfers. Targeted cash transfers would have the objective of protecting households from the risk of falling into poverty and should include target groups that are currently excluded from safety nets programs such as poor families with children and the unemployed youth. A growing body of evidence shows that imposing co-responsibilities on beneficiaries in exchange for receiving cash transfers can help promote specific positive behaviors and generate significant outcomes on education and health including better school enrollment, attendance, and completion, improved nutrition and increased productivity at households level that would break the cycle of poverty by “activating� the next generations. It is estimated that with a similar budget to what was spent in subsidies in 2014 (1.1 percent of GDP), 98 8 percent of the poor in Mozambique could be supported by a cash transfer equivalent to 50 percent of their consumption.1 As for activation activities, there is a need to further improve the policy framework and further develop activation programs at the national level. Professional trainings offered by INEFP’s 16 urban training centers, and the training programs offered by the Ministry of Education are the only sizeable and central level activation measures offered throughout the country. The private professional training sector, which recently has attracted the attention of donors in urban areas, remains small and its ability to absorb and train young and vulnerable school leavers is still a question mark. The biggest challenge consists in expanding financially affordable professional training opportunities to all provinces, districts and localities and adapt the curriculum to cater to the uneducated youth. In line with this, INEFP should adapt the skills development programs targeting young population that was unable to complete primary school. The so-called Formação em Aprendizagem model might be an option to take into consideration. Activation initiatives should broaden their scope and consider promoting interventions like productive inclusion or internships in public sector services. The household enterprises segment is rarely recognized by governments as a source of productive employment, but data show that a large proportion of the urban poor are engaged in this activity and, therefore looking into options to increase their productivity should be a priority. Productive inclusion activities, combining technical assistance, skills transfer, asset or cash transfers and extension services for the improvement of household enterprises have shown very positive results in countries like Brazil or Peru, such as increased income of beneficiary households, increased access to productive assets, a higher participation of women in productive activities. Another way of “activating� poor households that has shown positive preliminary results in Mozambique is using the public sector to satisfy the demand for subsidized internships to increase the employability of poor households. The public sector has an important capacity to generate and absorb temporary low-skilled employment. Several successful initiatives have been put in place in Mozambique in this regard including PASP beneficiaries graduating into groups offering services to municipalities to cover gaps on street cleaning and gardening and providing social and community services; or safety nets or activation beneficiaries being trained as assistants for public sector services such as teachers’ assistants to reduce the very high teacher-student ratio. For these initiatives to be more efficient and effective, deconcentration and decentralization processes should be promoted to give municipalities a higher financial and technical autonomy to implement safety nets and activation interventions. Municipal departments of social assistance should play a more prominent role in the implementation of national programs, by receiving specific budget allocations based on performance. This step forward in the decentralization would need a proper articulation between INAS and local governments, as well as proper resource planning and financial autonomy for municipalities which currently almost does not exist. At the administration level, the focus in on developing the “nuts and bolts� subsystems that facilitate the core business processes of safety nets and activation programs. These include, for instance, beneficiary identification systems and registries, targeting schemes, monitoring and evaluation arrangements, etc. It is fundamental that Government endorses and develops the basic subsystems to operationalize social programs including safety nets and activation. The main priorities should be the development of a Single Registry of Beneficiaries (SRB) and the registration of all current and future social programs 1 World Bank, 2015. From Gas to Cash: Policy Options for Transferring Resource Revenues to Citizens in Mozambique 9 beneficiaries, the adapation of the targeting methodology for the urban areas and the endorsement and outsourcing of a formal payment system for cash transfers and other program payments. The SRB would help the government operationalize safety nets and activation initiatives as a system and can improve the efficiency of the programs by avoiding duplications, minimizing inclusion errors and facilitating referral to different programs. The improved common targeting system should take into consideration the specificities of the urban poor individuals. For instance, the development of accurate poverty maps would be an important tool, as poor households tend to concentrate in specific areas of the cities, where shelter, water and electricity conditions are worse. The information from Chapter 1 of this report can help the government determine what are the most effective variables for categorical targeting. For activation programs, more than a targeting methodology, referal systems should be developed so that targeted safety nets beneficiaries could be refered to existing activation initiatives, filtering specific profiles that would probably consider variables related to the level of education. The payment system would guarantee that beneficiaries receive their entitlements in a secure, predictable, and timely manner, minimizing the administrative and transaction costs for both beneficiaries and the program. These elements are key for beneficiary households to be able to plan their productive invetsments and therefore graduate out of poverty in a sustainable manner. Finding a payments service provider for the urban settings in Mozambique should not be a problem as the market of potential service providers includes several options like e-payments through mobile phones, payments through banks, payments through security firms, etc. At the program level, the programs with higher potential activation and graduation out of poverty in urban Mozambique are the PASP and the activation activities implemented by INEFP. The main recommendations of the report focuss on design and implementation issues for these initiatives. As previously stated, the safety nets and activation sector in Mozambique is still a work in progress and the country’s capacity to operatioanlize and scale up these is still very weak. Related to PASP, several design features need to be reconsidered, including the generosity of the program, its targeting, its seasonality and the design of the labor-intensive public works components. The generosity of the program is very low and beneficiaries rightly complain that PASP benefit amounts only cover for some basic expenses, do not allow for savings and have no potential for productive investments. In fact, it is estimated that it would cost about MZM 2,570 to purchase a 2,000 calorie diet in Maputo, and around MZM 12,850 for an average 5-members family, compared to the MZM 650 that a beneficiary household participating in the program currently receives. The type of work that beneficiaries are assigned to, should play a significant role in their engagement with the program and skills development. The type of works organized by the program, including cleaning of streets, roads, sewage canals, beachfront and gardens has no scope for any capacity building or skills transfer element. Public works to be organized under the program should enable beneficiaries to develop skills that would enhance their employability. Moreover PASP in urban settings is implemented for only 6 months every year. The logic for this seasonality comes from the experience in rural areas, where households are engaged in agriculture for longer periods of the year and labor-intensive public works are only implemented during the lean season in order not to overlap with households’ productive activities. However, this seasonality does not exist in urban areas. Given the employment type of the majority of the urban poor, actually being engaged in household enterprises, it would make more sense to use a productive inclusion approach. The productive inclusion approach would include providing skills and grants to poorer population groups, to promote self- 10 employment, entrepreneurship and increase local productivity and enhancing linkages with other existing Government initiatives. Productive inclusion initiatives could combine three core pillars (i) a mandatory training/capacity building package to make sure that all productive inclusion participants receive training on soft skills, basic financial management, financial literacy (including savings) and entrepreneurship, including how to prepare a simplified business plan; (ii) a cash transfer based on performance that would be used as capital investment for their productive activities; and (iii) technical assistance and extension services. Similar experiences in Brazil or Peru, combining technical assistance, skills transfer, cash transfers and extension services have shown very positive results such as increased income of beneficiary households, increased access to productive assets, a higher participation of women in productive activities and an important diversification in the productive activities of beneficiary households. In sum, the safety nets and activation sector is a work in progress and the supply side is far from being able to satisfy the increasing demand in urban areas. Urban vulnerability is very likely to increase substantially over the next months due to the current economic and fiscal scenario and the forthcoming fiscal measures that the Government plans to put in place. Even if the budget allocation to the sector could be increased, it is very unlikely that Mozambique would have the operational capacity to scale up cash transfers and activation programs substantially in the short term. However, to maximize its current capacity, it is fundamental to consolidate and improve the existing programs by developing the above-mentioned delivery systems, modifying some design features and as well by outsourcing national programs and processes to private sector and civil society. The scale up of activation programs cannot only rely on Government structures. Instead, INEFP and INAS should rely on external support from third parties (NGOs, private sector, etc.), which have demonstrated experience in implementing this type of programs in an efficient manner and effectively reaching the poorest. 11 INTRODUCTION AND RATIONALE 38 percent of the urban population in Mozambique is poor and the latest political and economic developments are likely to increase their vulnerability. Moreover, a large share of the urban population that is close to the poverty line and that could worsen their poverty status through small variations in income. The depreciation of the metical by 42 percent against the dollar in 20162, the inflation reaching 20 percent in 2017 and the rising food costs as a result of the El Niño phenomenon have already amplified prices by almost 40 percent. This situation could be aggravated by forthcoming fiscal measures including dismantling fuel and fuel subsidies. Urban households are more susceptible to these economic shocks and increases in food prices and, therefore, mitigation measures to protect them from this economic shocks and foster productive opportunities are necessary. In addition to the political and economic context, urbanization in Mozambique has been a steady process over the last years and this urbanization dynamic is likely to continue in the next coming years transforming Mozambique into one of the most urbanized countries in the region. This process will potentially exacerbate the unemployment situation in the urban settings, enhancing the vulnerability of the urban poor and specially the urban youth. Motivated by the increasing urban vulnerability, this report is one of the first attempts to review programs, practices and expenditure on safety nets and activation interventions in urban settings in Mozambique. In particular, the report aims at enhancing Government knowledge and understanding about existing safety nets and activation programs in urban areas in Mozambique, assess their efficiency and effectiveness, and identify emerging lessons and challenges. The report aims at informing Government on how to enhance the implementation of safety nets and activation programs is response to the increased vulnerability in urban areas. It does so by reviewing existing programs, published literature, and analysing needs and target populations with household survey and administrative data and qualitative information collected through field surveys. The analysis tries answer to the following questions:  What is the profile of the potential safety nets and activation beneficiaries in urban areas (e.g., in terms of vulnerabilities, risks, employment, education, and asset ownership) and how does this compare to the national distribution?  Are safety nets and activation programs in urban settings reaching the poorest?  Are safety nets and activation programs in Mozambique adapted to the specific vulnerabilities of the urban poor?  What would be the main policy, programmatic and operational options to increase efficiency and effectiveness of safety nets and activation programs in urban Mozambique? Safety nets and activation programs are two set of instruments under the broader umbrella of Social Protection and Labor interventions. In this report, ‘safety nets’ will refer to non-contributory transfers in cash or in-kind, including programs such as unconditional and conditional cash transfers (CCTs), school feeding, and public works targeted to the poor. ‘Activation programs’ will refer to programs favoring the inclusion of the vulnerable population into the labor market. These programs can include skills development, 2 World Bank. Mozambique Economic Update. December 2016. 12 entrepreneurship, livelihood support programs, employment services, etc. Other key aspects of social protection such as pensions or labor regulations, are not addressed in this report. The report focusses only on the core urban safety net and activation initiatives. In the last years, there has been a proliferation of these programs in Mozambique as a result of an increase in budget support by Government and donors to the sector. Specifically for the activation agenda, this has also been due to the great expectations created by a potential increase in labor demand from the gas and mining sectors. For instance, DFID (US$22.1 million), the Canadian Cooperation (US$13.6 million) or the GIZ (US$5.2 million) have recently approved funding for programs aiming at supporting the private and public sectors as service providers for vocational and professional training activities and employment services. As for safety nets, the Public Expeniture Review carried out by the World Bank in 20123 already showed the level of fragmentation of the system, with multiple NGOs and donors involved in financing and implementing more than 20 different cash and near cash transfer programs. The urban safety nets and activation programs assessment looks at national programs with urban coverage and at programs directly implemented at municipal level. At national level, but specifically for the urban areas, the report assesses the main public and private interventions framed under the National Social Protection Strategy (Estratégia Nacional de Segurança Social Básica, ENSSB), the Strategic Urban Poverty Reduction Program (Programa Estratégico de Reducção da Pobreza Urbana, PERPU) and programs from the Ministry of Labor, including INEFP’s interventions. At municipal level, the report looks at existing Government and donor funded programs. The report is divided into three main chapters: Chapter 1: Profile of Mozambique’s poor urban population and urban safety nets beneficiaries The report builds a profile of the urban poor in Mozambique and of the beneficiaries of the Productive Social Action Program (PASP) in order to better understand specific vulnerabilities in urban settings of Mozambique and assess to which extent safety nets and activation programs in urban settings are targeting the appropriate beneficiaries for this type of programs. Chapter 2: Safety nets and activation programs assessment in urban Mozambique The report carries out a safety nets and activation programs assessment in urban areas, considering national programs implemented in urban municipalities and decentralized municipal programs implementing by local authorities, NGOs or private sector. The urban safety nets and activation programs assessment looks at national programs with urban coverage and at programs directly implemented at municipal level. At national level, but specifically for the urban areas, the report assesses the main public and private interventions framed under the National Social Protection Strategy (Estratégia Nacional de Segurança Social Básica, ENSSB), the Strategic Urban Poverty Reduction Program (Programa Estratégico de Reducção da Pobreza Urbana, PERPU) and programs from the Ministry of Labor, including INEFP’s interventions. At municipal level, the report looks at existing Government and donor funded programs. Chapter 3: Evaluating the readiness of Safety Nets and Activation initiatives in urban Mozambique. Conclusions and recommendations 3World Bank. Mozambique: Social Protection Assessment. Review of Social Assistance Programs and Social Protection Expenditures. 2012 13 Drawing on the previous two chapters and relevant international experience on urban safety nets and activation programs, the report assesses the readiness of the safety nets and activation sector in urban Mozambique and its capacity to be scaled-up and provides a series of recommendation at policy, administrative and programmatic level. 14 CHAPTER 1: PROFILE OF MOZAMBIQUE’S POOR URBAN POPULATION 15 1 DATA This chapter presents a profile of the potential clients of safety nets and activation measures in urban Mozambique. Safety nets or social assistance is used to refer to noncontributory transfer programs targeted in some manner to the poor or vulnerable to help protect families from the impact of economic shocks, natural disasters, and other crises. Activation is a policy ‘package’ that includes incentives, programs, and financial support to favor the inclusion of the vulnerable population into the labor market as a means of sustainable poverty reduction. While safety nets focus on all vulnerable populations, activation policies focus only on work-able individuals. This chapter provides a profile of the urban poor comparing them to the non-poor and the overall urban population and disaggregating the analysis by age cohorts. By defining a profile of poor households in urban settings, the chapter tries to identify the main set of factors that could partially explain why poor households and, more specifically, social safety net beneficiaries have limitations to participate in the formal and informal labor market. The analysis looks at factors such as employability, which includes elements that make an individual suitable for a job. The chapter also explores the reasons of low access of poor households to safety nets and activation initiatives in urban settings. One of the main concerns of the Government of Mozambique to scale up safety net programs is the risk of dependency for able-bodied, working age beneficiaries and the potential disincentives that cash transfers could create for their participation in other productive activities. This chapter provides as well an in-depth analysis of the specific characteristics of the work-able poor population, including beneficiaries of safety net programs that could be relevant to their capacity to participate in the labor market or engage in productive activities. By using registry and household survey data, this chapter also describes how the profile of current Productive Social Action Program (PASP) beneficiaries in urban settings compare to the national distribution of urban poor. This profile intends to provide information on how to better adapt the Program and other activation initiatives in order to make these more effective. Three sources of data form the foundation of this chapter: (i) the Mozambique Household Budget Survey (IOF 2008/09); (ii) qualitative data collected from Productive Social Action Program (PASP) beneficiaries and non-beneficiaries in Maputo; and (iii) PASP registration data collected in 2014 (administrative). 1.1 HOUSEHOLD SURVEY - IOF 2008-09 IOF 2008-09 is Mozambique’s Household Budget Survey. The survey contains information on demographic characteristics, household composition and household consumption. However, several adjustments were made to calibrate the data and correct consumption underreporting4. The most recent IOF 4 About 30 percent of households reported consuming an average of less than 1,000 k-calories per capita per day, which is below the World Health Organization’s minimum standard for survival. The calorie intake is below even this basic threshold in some Southern regions of Mozambique. An OLS regression between per-capita consumption expenditure as the dependent variable and poverty correlates as independent variables showed that for the households reporting less than 1,000 calories per capita per day, the predicted 16 was conducted in 2014 but it was not used for the current analysis as the consumption aggregates were not validated at the time of draftting this report. Hence, it could not be used for any kind of consumption based poverty analysis.5 1.2 QUALITATIVE DATA Qualitative data was collected for the PASP in Maputo particularly focusing on main barriers for poor households to access safety nets. The data was gathered though focus group discussions and individual interviews of beneficiaries and non-beneficiaries. It also includes key informant interviews with local officials in each of the seven selected bairros6 (neighborhoods) of Maputo.7 During the interviews, questions were asked on a number of topics related to PASP. Questions were asked regarding contents and conditions of work in the PASP; availability of information on PASP; effectiveness of communication strategy used by the government; process for selecting PASP participants; reasons for not participating and other constraints; complaint mechanism; management and organization of work; and suggestions for program improvement (see annex 1). 1.3 ADMINISTRATIVE – PASP REGISTRY DATA PASP registry data has information on all potential beneficiaries that were shortlisted after undergoing community based targeting for six provinces – Cabo Delgado, Manica, Nampula, Tete, Zambezia and Maputo. It includes information for the household head as well as for the other household members from 33,084 households accounting for 146,943 observations. In this Chapter, the data is restricted to households in Maputo (1,690 households with 7,530 individuals in total) and Tete (2,709 households and 11,554 individuals in total). The reason behind disaggregating the analysis for Maputo and Tete is that given the differences between the two municipalities, even though both are urban, it is likely that they will exhibit different trends. Maputo, as compared to other urban municipalities, is considered to be more developed with a diverse labor market. Whereas Tete could be used a proxy for other urban districts that are likely to be less developed. The registration data has information on demographic characteristics; household composition; shelter and possession of productive and household assets (see annex 1). expenditure values were higher than the original expenditure values as reported in the survey. This coupled with the finding that Engel’s law does not hold in the rural areas where the expected negative correlation between share of food and other expenditure levels is not observed, leads to the conclusion that there is underreporting of consumption in IOF 2008-09. To account for the underestimation, the unreliable observations were identified, which were either because of underreporting of consumption or overestimation of prices for household production. These observations were re- estimated using sample estimates based on 1996/97 survey, which was used because 1996/97 estimates are less affected by measurement error and hence are less biased than the figures in2008/09 figures. After making adjustments, the median consumption in Southern provinces increases whereas consumption estimates in Northern and Central provinces and rural increases decreases. One final adjustment was made using CPI estimates to make these consumption estimates usable for 2014 data. 5 Another available data source is INCAF 2013 but its variables have also not been validated yet. 6 A subdivision of an urban district Maputo is divided in six urban districts, and each of them is divided into several bairros, or neighbourhoods, with distinct limits and administrative structures. 7 Polana Caniço A, Polana Caniço B, Maxaquene B, Chamanculo D, Laulane, Costa do Sol and Mavalane B. 17 2 SOCIO-ECONOMIC PROFILE OF THE URBAN POOR IN MOZAMBIQUE 2.1 URBAN CONTEXT Mozambique’s urban settings represent a significant developmental challenge for the government. Cities and towns are increasingly being seen as key for both economic growth and poverty reduction, through a combination of improved urban living conditions, a more competitive platform for investment by the private sector and increased productivity of both people and assets. According to the UN, Mozambique’s urban population is estimated at 9 million out of a total population of 27.98 million.8 Despite poverty being predominantly rural in Mozambique, 21.1 percent of poor population live in the urban areas, representing around 3.4 million poor individuals. It is estimated that 38 percent of the population in urban areas live below the poverty line. Urban population is growing faster than the country as a whole, and is expected to double by 2030. Urbanization in Mozambique has been a steady process over the last years. In 2005, Mozambique was among the least urbanized countries in Southern Africa. By 2025, it is projected to be the fourth most urbanized country in the region, after Botswana, South Africa and Angola9. Currently the urban population represents 32.3 percent10 of the total population in Mozambique with an annual urban population growth of 3.3 percent11. The capital city, Maputo, despite being the main economic pole in the country, has an urbanization rate lower than other cities such as Nampula or Tete. Among urban residents, about 38.2 percent obtain their primary income from non-agricultural activities, face risks such as protracted underemployment, low income, food price increases, and other financial shocks less commonly felt in rural settings. For instance, data from 2009 indicate that in Mozambique, it was urban households who actually bore the brunt of the food price and financial crises, partially because they rely more heavily on food and fuel subsidies and thus face higher financial burdens when they are reduced or eliminated than the rural poor12. Estimates suggest these shocks led to a more than 11 percent decrease in consumption for the poorest 60 percent of the population in urban areas compared with less than 4 percent in rural areas13. 2.2 PROFILE OF URBAN POOR Survey data suggest that 37.3 percent of the population in urban Mozambique could be presumed to be able to work, that is, they are not in school, are able-bodied (not disabled) and are of working 8 Mozambique’s total population figure is from WDI database for 2015 9 Cities Alliance, World Bank. 2014 10 UN Data 2015 11 ASPIRE. 2014. 12 World Bank. Higher Fuel and Food Prices: Impact and Responses for Mozambique. World Bank and Ministry of Planning and Development of Mozambique. 2009. 13 Ardnt et. al. Higher fuel and food prices: impacts and responses for Mozambique in Agricultural Economics, 39. 2008 18 age. From the 3.4 million people living under the poverty line in the urban areas, 36.5 percent (2.1 million) could be able to work. Labor force participation rates are exceptionally high in urban settings of Mozambique.14 Labor force participation rates are generally high across age groups and urban areas, but youth (15-24) have lower labor force participation rates than older adults (25-64). Non-poor labor force participation rates are slightly higher than rates among poor people for both youth and older adults. Figure 1.1: Labor force participation rates 100% 80% 60% 40% 20% 0% Poor Poor Poor NonPoor Non Poor Non Poor Maputo Tete Urban Youth 15-24 25 and over Working Age Source: IOF 2008/2009 Unemployment among the youth is a large concern in Mozambique. At the national level, only 26 percent of the urban 15-24-year-olds state that they worked during the past 7 days with an additional 7 percentage points noting that they are employed although they didn’t work the past week. Adults are far more likely to have worked the past week, with 60 percent confirming this and additionally 14 percentage points saying that they are employed leading to an employment rate of 74 percent. 14 Labor force participation rate is defined as the percentage of working age population which is either employed or looking for employment. 19 Figure 1.2: Employment Figure 1.3: Share of temporary employment Source: IOF 2008/2009 Among the employed, urban poor have a higher propensity to work in temporary positions (defined as either seasonal or occasional). Especially the young are prone to these types of jobs as 28 percent of workers in this age group hold temporary positions. The differences between poor and non-poor in terms of employment status seem to be larger in Maputo than in the country as a whole, both for the youth and for adults. Formal sector in the urban areas, as represented by wage employment, is very small. Figure 1.4 shows that only 16.2 percent of the urban poor and 11 percent of poor urban youth has wage or salaried jobs. The percentage increases for the urban poor over 25, which suggests that youth struggle more in accessing jobs than the older age cohort. For the non-poor, access to wage employment is more frequent with almost 25 percent of the non-poor urban employable population having access to wage employment. Even though labor force participation and employment rates are high in Mozambique, the majority of the labor force works either for themselves or their families, as steady wage employment is scarce even in urban areas. Recent research work on jobs in Mozambique shows that most urban residents have been struggling to find sustainable livelihood opportunities despite working long hours. The situation is made worse by lack of assets and the burden of household chores in case of female headed households.15 There are significant differences in the type of employment between poor and non-poor individuals and across urban regions. The main types of employment for the urban poor are: self-employment (26.9 percent); family worker (54.7 percent); and wage employment (15.9 percent). The overall urban (poor and non-poor) are more likely to be engaged in waged employment (29 percent). Urban poor in Maputo for instance, are more likely to have access to wage and salaried work (36 percent) than poor households in Tete (15 percent), where poor households tend to be more engaged in urban and periurban agriculture (see Figure 1.4). 15 World Bank (2016) Mozambique: Jobs for Development 20 When comparing age cohorts, differences in the type of employment do also exist . 18 percent of poor 25 and older individuals are engaged in wage and salaried work, compared to 11 percent for 15-24 individuals. The urban youth is more likely to be engaged in a household enterprise (87 percent) than the older cohort (80 percent). Figure 1.4: Type of employment for youth (a) and people 25 years and older (b) (a) (b) Youth: 15-24 25 and over 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% Poor Poor Poor Poor Poor Poor NonPoor Non Poor NonPoor Non Poor Non Poor Non Poor Maputo Tete Urban Maputo Tete Urban Wage & Salaried Household Enterprise Wage & Salaried Household Enterprise Domestic Employee Volunteer Domestic Employee Volunteer Working Age 100% 80% 60% 40% 20% 0% NonPoor Poor Non Poor Non Poor Poor Poor Maputo Tete Urban Wage & Salaried Household Enterprise Domestic Employee Volunteer Source: IOF 2008/09 The informal sector is large and, overall, levels of productivity are fairly low16. The small proportion poor individuals in urban areas engaged in wage employment (only 15.9 percent) shows that the economy has not created good formal sector wage jobs and the poor have not had access to the few created (Figure ). In 16Labor productivity is defined as output per unit of labor input. GDP per person engaged in 1990 US$. Source: Key Indicators of Labor Market (KILM), International Labour Organization (ILO) 21 particular for the urban poor, the primary sector which is dominated by agriculture, and which has the most share of jobs, has the lowest labor productivity as compared to other sectors of the economy. Not only is labor productivity very low but it also has remained fairly stagnant in the past decade because of rudimentary technologies and lack of access to credit, information, infrastructure and markets17. There are significant differences in the sector of employment between poor and non-poor households, across age groups and across urban regions. Despite living in urban areas, poor households are more likely to be employed in the agriculture and fisheries sector (70 percent) than in other sectors. When compared to non-poor households, the urban poor are more likely to work in agriculture than the non-poor (43 percent). Across age cohorts important differences in type of employment are also found, because 78 percent of the urban poor youth work in agriculture compared to 67 percent of those being 25 and over. In contrast, 25 and older poor individuals living in urban areas are more likely to work in commerce (13 percent) and manufacturing (5 percent) as compared to the poor urban youth (11 percent and 3 percent respectively). Across urban regions too, there are significant differences in the sector of employment with more poor working age individuals engaged in agriculture in smaller cities like Tete compared to Maputo where many more opportunities seem to exist in commerce and services. For example, in Maputo, only 24 percent of working age poor individuals are engaged in the agriculture sector (64 percent in Tete), while only 14 percent of non-poor working age individuals are in the agriculture sector (55 percent in Tete). This trend of high employment in agriculture and commerce is consistent with other countries in the region. Here commerce refers to household enterprise activities that revolve around retail trading such as operating kiosks, market stalls or roadside locations selling fruits, vegetables and other processed products, houseware, second hand clothes, electronics, soaps and other personal care products. Following commerce and services, individuals are engaged in manufacturing sector which is usually natural resource processing like brewing beer, making charcoal, processing food and brickmaking.18 17 Jones and Tarp. Jobs and Welfare in Mozambique, 2013. Background paper to World Development Report (WDR) 18 World Bank (2016) Mozambique: Jobs for Development 22 Figure 1.5: Sector of employment for youth (a) and people 25 years and older (b) Youth 15-24 Poor Urban Non Poor Poor Maputo Tete Non Poor Poor Non Poor 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Agriculture & fishing Mining Manufacturing Electricity & utilities Construction Commerce Transport, storage & communications Financial, insurance & real estate Services, public administration Other services 25 and Over Poor Urban Non Poor Poor Maputo Tete Non Poor Poor Non Poor 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Agriculture & fishing Mining Manufacturing Electricity & utilities Construction Commerce Transport, storage & communications Financial, insurance & real estate Services, public administration Other services 23 Working Age Poor Urban Non Poor Poor Maputo Tete Non Poor Poor Non Poor 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Agriculture & fishing Mining Manufacturing Electricity & utilities Construction Commerce Transport, storage & communications Financial, insurance & real estate Services, public administration Other services These employment sector data are to be looked at carefully for the case of Tete. The data describe the situation in 2009 (when the household survey was carried out) and since then the mining sector has grown significantly in Tete and it is likely that currently, a larger number of households are engaged in mining there. Due to the high levels of informality, many households are not covered by the social security system. The small proportion of wage employment in urban settings shows that in spite of economic growth over time, the economy has not created good formal sector wage jobs and, therefore, a large share of households do not contribute to the social security system, relying therefore for insurance on non-contributory social protection initiatives. Poor urban households have very limited access to safety nets and activation programs. Despite an increase in coverage of 6.5 percent from 2015 to 2016, coverage of safety net programs in urban settings remains very low (only 15 percent of the target population in 201619). Moreover, the way current programs are designed seems to exclude one of the main target groups in urban settings, the unemployed youth. Less than 1 percent of the target group of the main activation program in the country, the Productive Social Action Program, is currently participating in the program. Table 1.1: Safety nets beneficiary households Safety nets beneficiaries (households) 2015 2016 Total 381,843 467,970 Urban 92,017 98,028 Source: INAS, 2016 Qualitative analysis carried out in Maputo suggests that the main barriers for poor households to access to safety nets relate to operational challenges faced by the National Institute of Social Action (INAS), which is the institution in charge of implementing safety nets programs in Mozambique. For instance, according to the plans and the budget allocated to the Productive Social Action Program, the Program should have enrolled around 30,000 households (150,000 individuals) in the five biggest cities of the 19 INAS. Balanço anual dos programas de Segurança Social Básica. 2016 24 country by 2017, however, only around 7,000 households have been enrolled so far20. Outreach campaigns seem to have been efficient with frequent communications by radio and TV and beneficiaries and non- beneficiaries also informed that the administrative burden for participating in the programs is not very heavy. 2.2.1 Main determinants of Urban Poverty To identify the main determinants of poverty in the urban settings of Mozambique, a Proxy Means Test was developed. Proxy Means Tests generate a vulnerability index for households based on fairly easy- to-observe and not easily modifiable household characteristics that have been shown statistically to be highly correlated with consumption poverty and that usually are grouped into two categories: (i) information about household members, and (ii) information about housing characteristics and assets. The indicators used to calculate this score and their weights were derived from a regression analysis of data from 2008/2009 household survey. The list of indicators included in the urban PMT score and their respective weights is presented in Table 1.2 below. While Proxy Means Testing (PMT) can be a useful method for identifying poor, formulas should account for differences between urban and rural areas. This is needed to account for the differing characteristics and correlates of the poor and non-poor in city and rural zones. These include the cost of living, labor market conditions and level of monetization of payments, differing costs such as housing, and consumer preferences. Rural PMT formula often focus more on variables such as housing conditions and agriculture, while in urban areas there are generally higher prominence for durable goods. Since the dependent variable used in PMT score calculations is the aggregate consumption so a similar analysis can be conducted to determine factors that determine the likelihood of being poor in urban regions. Table 1.2 shows the list of variables and their marginal effects on probability of being poor for urban areas in Mozambique. Table 1.2: Factors Affecting Poverty in Urban Areas in Mozambique Dependent variable: Individual’s aggregate consumption falling below the National Poverty Line (yes=1) Group Indicator Margins Robust Standard Errors. Household members Household size 0.017*** 0.001 Female Headed Household 0.034*** 0.008 Children under 5 0.058** 0.008 Dependency ratio 0.031** 0.004 No Education Ref Ref Primary education -0.023 0.014 Secondary education -0.064*** 0.015 Post-Secondary education -0.109*** 0.022 College & post graduate -0.200*** 0.030 Housing/assets Rooms per HH member -0.056*** 0.010 Piped or protected water -0.054*** 0.009 Good walls -0.016 0.010 Good floor -0.148*** 0.011 Good roof -0.139 0.030 Toilet connected to sewer or septic tank -0.056 0.013 20 INAS, 2016 25 Electricity -0.071 0.009 TV -0.136*** 0.009 Radio -0.025*** 0.007 Computer -0.205*** 0.031 Vehicle -0.159*** 0.013 Bicycle -0.048*** 0.009 Note: Source: IOF 08/09. n = 17,984, Pseudo R-squared = 0.324. Differences among the urban districts were controlled by fixed effects at the district level. All variables except Household size, Dependency ratio and Rooms per HH member are dummies. The regression results in Table 1.2 show that the education level, the household composition and the possession of household and productive assets are important indicators determining poverty in urban settings of Mozambique. The following paragraphs will analyze the patterns of these indicators among age cohorts, regional areas and poor and non-poor households. Surprisingly, in urban Mozambique (and overall as well), there’s a negative correlation between employment and years of education. It means that with more years of education, chances of getting employment are low. However, while disaggregating the data by type of employment, chances of getting wage or salaried employment increase with more years of education whereas self-employment and family worker decrease with years of education. This suggests that individuals with higher education prefer to queue and wait for a good job, instead of accepting low-wage opportunities. In general, poor urban households are less educated than non-poor urban households. Non-poor urban households have on average completed 6.2 years of education and 45 percent of them have completed secondary, post-secondary, college or post-graduate whereas among the poor urban households they have on average completed 4.1 years of education and only 31 percent have completed secondary, post-secondary, college or post-graduate. The youth cohort in urban areas has slightly higher level of education than its older counterparts with an average of 7.2 years of education completed as compared to 6.4 years. This same pattern is observed when considering only poor urban households as the poor urban youth has higher education than the older cohort. Among the unemployed, the urban youth’s educational attainment (6.9 years) is only slightly below that of the older cohort (7.1 years), suggesting that it is not educational attainment that is behind the observed urban youth unemployment relative to that of the older cohort. It may be perhaps differentials in labor experience or even in education quality and skills across the two age cohorts. Table 1.3– Completed Level of Education Age cohorts Urban Urban Poor Employment Status 15 - 24 25+ 15+ 15 - 24 25+ 15+ Employed 6.7 6.3 6.6 5.6 4.5 5.1 Unemployed 8.9 8.4 8.8 7.1 6.1 6.7 Source: Author’s calculations using IOF 2008-09 data 26 There exist no regional disparities on the level of education for poor urban households. On average, poor individuals in Maputo have similar educational attainment than poor individuals in other urban areas. For example, in Maputo, 31 percent of the poor have attained secondary education, compared to 29.5 percent in Tete and 29 percent in all urban areas. Figure 1.6: Educational Attainment among Urban Population in Mozambique 100% 80% 60% 40% 20% 0% Overall Urban Non Poor Poor Urban Maputo Poor Tete Poor Urban No education Primary Secondary Post secondary College and Post Graduate Source: IOF 2008-09 Another determinant of poverty in the urban settings is the household size and the dependency ratio. There are differences in household size across poor and non-poor households and across households in different urban settings. Poor urban households have an average size of 6.5 members compared to 6.2 members for non-poor households. In terms of the composition of the households, almost 21 percent of poor households in urban areas have an elderly member (60 years and over) compared to 17 percent of non- poor households and about 75 percent of poor households in urban areas have a child under 5, compared to 58 percent of non-poor households. This shows that targeting families with children under five and with elderly members is a good proxy for poverty targeted safety nets programs. In Maputo, the size of poor households is surprisingly larger (7.5 members) than in overall urban areas where it is 6.3. Poor households in Tete have around 6.6 household members, which is similar to the household size in poor households in all urban areas of Mozambique. Surprisingly, poor households in Maputo tend to have less children under 5 (71 pernect) than the average (75 percent) and much less than poor household sin cities like Tete (85 percent). Dependency ratio calculated as the share of dependent members (children and elderly) divided by household 27 size among poor households is 0.20 against 0.15 for non-poor households. Again, this reflects a larger number of children in poorer households. Figure 1.7: Average Household Size in Urban Figure 1.8: Percentage of Households with Children and Elderly Mozambique 7.5 90% 80% 6.6 6.5 6.3 6.2 70% 60% 50% 40% 30% 20% 10% 0% Overall Non Poor Poor Urban Maputo Tete Poor OVERALL NON POOR MAPUTO TETE Urban Urban Poor URBAN POOR URBAN POOR POOR URBAN HH with Elderly HH with Children Source: IOF 2008-09 When looking at the distribution of poverty by age cohort, in urban Mozambique, poverty headcount rates are higher among people 60 years and over, relative to people 15-24 or people 24-60. About 42 percent of the population 60 and over living in urban Mozambique are poor. This poverty headcount rate is higher than the rate observed among people 15-24 (32 percent) and among people 24-60 (34 percent) living in urban areas. Surprisingly, the poorest age group in urban areas are the under 15 with 43.4 percent living in poverty, which supports again the idea that targeting households with children would be an effective categorical targeting for safety nets programs. However, as can be observed from Figure 1.9, when considering absolute numbers, the largest number of poor individuals are these in the ages from 35 to 44, 25 to 34 and 45 to 54. The youth and the elderly, despite a large share of them being poor, they only represent around 6 and 11 percent of the total urban poor respectively. Figure 1.9: Distribution of urban poverty by age cohort in Mozambique 28 35 30 25 20 15 10 5 0 Between 15 Between 25 Between 35 Between 45 Between 55 Above 60 & 24 and 34 and 44 and 54 and 59 Maputo Poor Other Urban Poor Overall Urban Tete Poor Source: IOF 2008/09 While most households are headed by males in urban areas of Mozambique (72 percent), poverty rates are higher among female headed households. 42 percent of female headed households in urban Mozambique live in poverty as compared to 36.3 percent of male headed households below the poverty line. There exist regional variations in the share of poor households headed by a female with Maputo having a higher share (38 percent) compared to the overall poor urban households (31 percent) or the share in Tete (only 24.5 percent). Figure 1.10: Gender of Household Head in Mozambique 80 70 60 50 40 30 20 10 0 Overall Urban Non Poor Poor Urban Maputo Poor Tete Poor Urban Female Headed Male Headed Source: IOF 2008-09 About 39 percent of poor urban households in Mozambique are married, while among non-poor people in urban areas the proportion drops to 34 percent. A relatively high percentage of the poor urban youth (29.3 percent) are married, supporting the idea that early marriage can potentially contribute to poverty status. In urban households, widowhood is more common among the poor (5 percent), than among the non- poor (3.8 percent). Civil unions are particularly common among people in urban areas (9 percent), where poor households (10.5percent) are only slightly more likely to be engaged in civil unions as compared to non- poor (9 percent). Figure 1.11: Marital Status of Urban Population in Mozambique 29 50 45 40 35 30 25 20 15 10 5 0 Overall Urban Non Poor Poor Urban Maputo Poor Tete Poor Urban Single Married Civil Union Divorced Widowed Source: IOF 2008-09 The lack of household or productive assets is also a proxy for poverty and vulnerability in urban settings. Non-poor households are more likely to have accumulated household and productive assets because they have more purchasing capacity than poor households and because, unlike poor households, they don’t need to sell their assets to cope with protracted underemployment, low income, food price inflation, and other financial shocks. As consumption figures collected through household survey data may sometimes be biased, a look at the ownership of assets and household characteristics gives a snapshot of household wealth, where households that are wealth poor can be considered chronic poor. For the case of urban Mozambique, the following variables were considered:  Water source  Electricity  Television  Wall materials  Vehicle  Cattle  Floor materials  Radio  Sheep or goats  Roof materials  Computer  Chicken  Toilet  Bicycle There are significant differences in asset ownership among poor versus non-poor households in urban Mozambique. Not surprisingly, non-poor urban households have better housing conditions than poor households. 78 percent of non-poor households have good quality walls against 62 percent of poor urban households. The same pattern repeats for access to good quality walls with 75 percent of non-poor households having cement, tile or marble floor against 35 percent of poor households. Access to electricity is also a very clear proxy for poverty with only 22 percent of poor households having access to it against 62.5 percent of non-poor households. Surprisingly there are no big differences between poor an non-poor in the possession of medium livestock or poultry. The only asset that poor households are more likely to have are bicycles, but not with a very big difference compared to the non-poor. It is worth highlighting that household conditions and asset possession are on average better in Maputo than in other places. 96 percent of poor households in Maputo have access to protracted water compared to the national urban average of 66 percent. Same pattern is observed for housing conditions with poor households in Maputo having better housing conditions (walls and floors) than the national average of urban households. On the other side, poor households in Tete tend to have larger numbers of livestock than the national average of urban households, which supports the finding that in secondary cities a large number of the urban poor are engaged in farming. 30 Table 1.4: Asset ownership of urban households (%) Overall Non Poor Poor Urban Maputo Tete Poor Urban Urban Poor Brick, cement or concrete walls 72.3 78.2 62.7 76.1 52.0 Cement, tile or marble floor 59.5 74.6 35.1 85.2 30.2 Access to protected water 66.7 75.4 52.5 96.1 89.9 Concrete, cement, brick or stone roof 5.0 7.5 1.1 0.8 0.0 Access to flush toilet 13.9 20.5 3.3 5.9 4.3 Access to electricity 47.0 62.5 21.8 52.8 31.5 TV ownership 41.3 56.7 16.4 37.7 13.6 Radio ownership 50.1 53.9 44.1 34.1 30.7 Ownership of car or motorcycle 11.3 16.4 3.0 0.0 2.5 Bicycle ownership 26.7 25.5 28.6 2.4 13.8 Computer ownership 4.6 7.3 0.3 1.7 0.0 Large livestock 8.6 11.8 4.4 6.8 26.9 Medium livestock 27.6 27.5 27.9 9.8 52.2 Poultry 87.7 87.1 88.5 73.7 74.3 Source: Author’s calculations using IOF 2008-09 data A more reliable analysis of a household wealth is done by constructing a wealth index, which can be done solely from the census data using the variables given in Table 1.4. One of these indexes is the Morris Asset Index which is a weighted sum of possession of certain household assets (e.g. radio, tv, refrigerator, bicycle, car, toilet) where the weights are simply calculated as the inverse of the frequency of that asset. The logic is that since a car for instance is rarer than a bicyper centcle, it obtains a higher weight in the index. The Morris Index is then calculated as the sum of these variables (X) multiplied by a weight (W) equal to the inverse of the mean of the variable (i.e. the frequency with which the dummy variable reports a 1): 31 � 1 𝑀𝑜𝑟𝑟𝑖𝑠 𝑖𝑛𝑑𝑒𝑥 = 𝑋𝑖 ∗ 𝑊𝑖 , 𝑤ℎ𝑒𝑟𝑒 𝑊𝑖 = 𝑋𝑖 𝑖 =1 The Morris Index was used to develop urban poverty maps for the main cities of Mozambique to try to identify specific patterns in the distribution of poor populations in the urban areas. The poverty maps developed for the five biggest cities of Mozambique show that poor urban households concentrate in specific neighborhoods in the main cities. The poverty maps developed based on the 2007 census and average PMT scores identified the areas where poor urban households concentrate. The concentration can be very easily observed for the cities of Beira where poor households concentrate in the North-western neighborhoods or Quelimane, where the highest concentration happens in the Southern neighborhoods. This shows that geographical targeting through poverty maps could be a very effective way of targeting poor households for safety nets and activation programs. Figure 1.12: Baseline maps based on 2007 census average PMT scores Left: Maputo city including Catembe. Right (clockwise from top left): Beira, Nampula, Quelimane, Tete Source: Author’s elaboration There are large differences in urban poverty across provinces with the Northern provinces generally showing higher rates of urban poverty than the Southern, Gaza being the only exception to this pattern. Maputo city is the province with lowest urban poverty rate. When looking at the distribution of daily per capita consumption, the graph for Maputo lies substantially to the right of the national distribution. It should of course be noted that these consumption figures are deflated both temporally and spatially in order to reflect the differences in purchasing power in different locations. Figure 1.13: Urban headcount rates by province Figure 1.14: Daily consumption, deflated 32 Source: IOF 2008/09 Profile of poor urban population in Mozambique There are 3.4 million poor individuals in urban Mozambique. Poor urban populations, different to rural, face specific risks related to low income and protracted unemployment that make them very vulnerable to food price increases. Labor force participation and employment rates of urban poor in Mozambique are high except for the youth, but most of the urban poor are engaged in non-wage employment related to household entreprises, most of these in the agriculture sector and overall with very low levels of productivity. Despite most of them being engaged in the informal sector (non-contributory), access to non- contributory safety nets programs is very low for the urban poor, with only 15 percent of them befefitting from any type of social assistance. Poor urban individuals spend on average 4.1 years at school and only 31 percent of them have completed secondary school or more. Poor urban households are large (6.5 members) and many of them have elderly (21 percent) or children under five (75 percent) translating into high dependency ratios. Urban children (43 percent) and the urban elderly (42 percent) are more likely to be poor than the other cohorts. Female headed households in urban areas are more likely to be poor than male-headed households. Poor urban households concentrate in Northern provinces of the country and in specific neighnourhoods in the cities, where access to electricity (22 percent), protracted water (52 percent) and good quality housing (62 percent) is low. 33 3 THE PRODUCTIVE SOCIAL ACTION PROGRAM (PASP) Using household survey and administrative data, this subchapter defines the profile of urban beneficiaries of the Productive Social Action Program (PASP), which is the only national program combining safety nets and activation activities. gaps and limitations for them to access and make use of current services and programs. By using data from the Single Registry of Beneficiaries and the 2008/09 household survey, this subchapter also describes how the profile of current safety nets beneficiaries in urban settings compare to the national distribution and to the national distribution of urban poor. In 2011, INAS began implementing the PASP with the objective of promoting socio-economic inclusion of poor households with members with capacity to work. PASP has an urban component and in its design it combines cash transfers through participation in labor-intensive public works schemes with activation activities related to on-the-job training, skills provision, basic financial literacy, savings and referral to other productive initiatives. PASP has a double function: on the one hand it has an income stabilization and smoothing consumption component, which is achieved through predictable cash transfers to extreme poor households participating in labor-intensive public works schemes, and, on the other hand, PASP has a propelling, activating function, by providing skills to beneficiaries to increase their productivity and employability. The combination of cash transfers with activation activities tries to respond and mitigate some of the following characteristics of the labor market in Mozambique: - Low levels of productivity. The primary sector which is dominated by agriculture and which remains the main sector for poor urban households, and which has the most share of jobs, has the lowest labor productivity as compared to other sectors of the economy. Not only is labor productivity very low but it also has remained fairly stagnant in the past decade because of rudimentary technologies and lack of access to credit, information, infrastructure and markets. - High levels of informality. The small proportion of wage employment in urban settings shows that in spite of economic growth over time, the economy has not created good formal sector wage jobs. - Youth unemployment. Youth unemployment is also more pronounced in the urban areas - about 39 percent of youth in urban areas are unemployed. PASP in urban areas is implemented for six months every year and the selected beneficiaries receive a transfer of MZM 650 for three consecutive years that provides a predictable and seasonal source of income to poor and vulnerable households. The subsidy has been re-adjusted to MZM 1,000 for 2017 and will be increased by MZM 200 every year. In this subchapter, the PASP data is restricted to households in Maputo and Tete. The reason behind disaggregating the analysis for Maputo and Tete is twofold. First, Maputo and Tete are the two first municipalities where PASP has been implemented. Second, given the differences between the two municipalities, it is likely that they will exhibit different trendsas Maputo is considered to be more developed and with a more diverse labor market. Tete is showcased in some specific parts of the chapter to show that there are variations in trends even within urban districts that the program design should take into account. In Maputo and Tete, PASP covers a very small portion of the population. Only 4,399 households benefit from PASP in Maputo and Tete, and therefore, by extension, only about 22,000 individuals benefit from it. This is a very small proportion of the poor workable people who are estimated to live in these two cities. 34 The low coverage of PASP is not due to lack of outreach activities or financial constraints, but to weak operational capacity from INAS. Focus group respondents report that PASP is relatively well publicized and as a result, residents are informed about the program. Signing up for the program is also reported to be relatively easy, especially if beneficiaries are signing up for the second time. Low coverage of PASP is more related to operational challenges linked to weak capacity of the implementing agency, INAS, that has not been able to scale up the program in the last 4 years even though a budget was allocated to enrol over 30,000 households in urban areas. From those eligible beneficiaries registered in the Program, some of them decided not to participate due to design features of the program, namely, the low level of subsidies and the type of public works designed. 3.1 TARGETING PERFORMANCE OF PASP In urban areas, PASP targets extreme poor households with adults who are able to work so that they can participate in labor-intensive public works and other income activation activities. By design, PASP uses a three steps approach towards targeting: (i) geographical targeting through poverty maps; (ii) community based targeting to prepare the lists of potential beneficiaries; and (iii) a Proxy Means Test to minimize inclusion errors. PASP Targeting methodology Step one: Geographical targeting of cities and neighbourhoods (bairros). The Government of Mozambique produced poverty maps for all districts and municipalities in Mozambique that allow to classify these based on poverty incidence and severity. For the first phase of implementation of PASP, the government selected Maputo, Nampula, Beira, Quelimane and Tete for implementation of the program due to the high number of poor populations in these cities and due to risk of social unrest. After selecting these cities, more disaggregated geographical targeting helped to distribute resources to specific neighbourhoods. Step two: Community Based Targeting (CBT) relying on the community’s knowledge of poverty and vulnerability. CBT uses a two-step process where information comes from two different sources. The first step is based on a top-down process where the neighbourhood level imparts information to the community regarding the potential number of beneficiaries and eligibility criteria. They rely on formal local structures at neighbourhood and block level (quarterão). The second step is where the neighbourhoods’ Advisory Councils, based on the census and community knowledge, start compiling a list of potential beneficiaries. The lists are then submitted to the higher authorities for approval and once approved, INAS starts reviewing them using PMT scores. The list of potential beneficiaries reflects the community’s judgement about the poor or the vulnerable and categorical criteria are used to build the lists. Step three: A Proxy Means Test (PMT) is used to verify the eligibility of participants from the list provided by the community. Data is collected on various indicators from the potential beneficiary households that feed into the PMT score calculation. The cut offs are decided based on consumption levels derived from the household survey data (IOF 2008/09). 35 Eighty percent of urban PASP beneficiaries in Maputo are from households that are in quintile one or two, implying that the poverty targeting performance of the program is very good. Around 46 percent of PASP beneficiaries are from quintile 1 and 34 percent are from quintile 2. Around 14 percent of PASP beneficiaries belong to quintile 3, 5 percent to quintile 4 and no PASP beneficiaries belong to quintile 5. Figure 1.15: PASP benefit incidence [VALUE] 34.16% 14.37% 5.01% 0.35% Bottom 2nd 3rd 4th Top Quintile Quintile Source: ASPIRE Database Comparing targeting performance for PASP in Maputo with public works programs in urban areas in other countries, we note that PASP targets the poor relatively well. 46 percent of benefits accrue to people in the bottom quintile, compared to only 19 percent in Malawi, 11 percent in Ethiopia, 10 percent in Lesotho and 1 percent in Tanzania. Figure 1.16: Beneficiary Incidence of Public Works Programs in Urban Areas 46.11 18.58 10.44 11.08 0.97 Tanzania Lesotho Ethiopia Malawi Maputo Source: ASPIRE Database The main factor contributing to the good targeting performance is the application of a verification system (Proxy Means Test) to the list of potential beneficiaries resulting from the community based targeting process. For the application of the PMT, INAS collected socio-economic and demographic data from all potential beneficiaries through household visits. The data were entered in the registry of beneficiaries and PMT scores were calculated determining eligibility thresholds. Despite the good final results (80 percent of retained participants belong to the poorest two quintiles), the data on Table 1.5 show that the community 36 based targeting was not a very efficient system to identify the poor. When looking at the data, only 23 percent of the potential beneficiaries were actually eligible (living below the poverty line), meaning that the remaining 77 percent were not eligible. This means that a lot of operational costs were dedicated to register and verify households that were finally not be eligible for the program. If increasing the threshold to the poverty line plus 100 percent, still 17 percent of the households registered were over the threshold, implying that the community based targeting was facilitating elite capture of households not complying with the eligibility criteria. This was likely to be a combination of actual elite capture and lack of knowledge of eligibility criteria provoking targeting errors in the community based targeting. In any case, it looks clear that the community based targeting methodology needs to be revisited. Table 1.5: Targeting of urban beneficiaries (avg. PMT scores and percent eligible) Province PMT scores National National National USD PL* PL PL 1.25 +50perc +100per /day ent cent Nampula 24.82 24% 60% 82% 98% Zambézia 27.05 13% 46% 76% 98% Tete 20.93 32% 74% 93% 99% Maputo 59.60 0% 0% 25% 50% Total 24.54 23% 59% 83% 98% *Poverty line Source: Author’s calculation based on Registry and IOF 2008/09 data PASP targeting could be improved through a better use of graphical targeting through poverty maps. Figure 1.17 shows that in Maputo the distribution of beneficiaries does not exactly correspond to the variation in Morris Index Score as represented in the poverty maps. One reason for the geographic mismatch is that the PASP is not implemented in a number of bairros. In fact, several of the poorest bairros in the north of Maputo are not included in the program. With coverage rates as low as they are, it is not surprising that a lot of people who should benefit from the PASP are excluded from it. However, if the Program did a better effort in prioritizing neigbourhoods with higher prevalence of poverty, inclusion errors could be minimized. Figure 1.17. Geographical distribution of PASP beneficiaries in Maputo 37 Source: Author’s elaboration using PASP registry data Another key factor which partially explains the good poverty targeting performance for the actual beneficiaries is the low subsidy and the type of public works designed, which lead to self-selection into the program by poor people and people with few options for work. Qualitative work shows that beneficiaries find the subsidy for participating at PASP very low and the type of public works design not interesting and not fit as a skills transfer activity. The subsidy is so low that beneficiaries complained that it could only be used for food and other basic items, usually on the day that they received the funds and nothing could be saved. Furthermore, the qualitative work revealed that beneficiaries sign up for PASP because they have no other choice and this partially explains the good targeting performance. 3.2 SOCIO-ECONOMIC PROFILE OF URBAN PASP BENEFICIARIES Information used to build PASP beneficiaries’ profile comes from PASP registry data that has information on all potential beneficiaries that were shortlisted after undergoing the community based targeting for Maputo and Tete. This is complemented by qualitative data collected for the PASP in Maputo particularly focusing on main barriers for poor households to access safety nets. PASP beneficiaries have a high “workable�21 population proportion in both Maputo and Tete. In Maputo, 82 percent of PASP beneficiaries are workable and in Tete this proportion increases to 94 percent. However, in both regions, there is an important proportion of households that are not part of PASP target group (18 percent in Maputo and 6 percent in Tete), because they are either under 15, over 60 or disabled. These beneficiaries should be enrolled in other existing safety net programs for households with no members with capacity to work such as the social pension Basic Social Subsidy Program22. Figure 1.18: Proportion of population that is workable versus a dependent or disabled Maputo PASP Tete PASP Maputo Poor Tete Poor Maputo Overall Tete Overall 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Workable Elderly Under 15 Workable Elderly Under 15 Student Disabled Student Disabled Source: PASP Registry Data 2014 21We define workable as people who are not elderly, under 15, students, or disabled 22The Basic Social Subsidy Program is a social pension targeting poor households with no members with capacity to work and providing cash transfers in an amount of MZM330 to MZM550 depending on the size of the household. 38 PASP beneficiaries are on average less educated than the poor population. 57 percent of PASP beneficiaries in Maputo have no education and only 35 percent of PASP beneficiaries in Maputo have attained primary education compared to 59.5 percent of urban poor with primary education. Data for Tete are a bit worse than for Maputo, as in Tete only 39 percent of PASP beneficiaries have attained primary educ ation Figure 1.19: Educational Attainment among PASP Beneficiaries and 63 70 perc 60 ent 50 of 40 the 30 m 20 have 10 no 0 educ No education Primary Secondary Post secondary & Higher ation . Maputo Tete Source: PASP Registry Data 2014 PAS P targeting does not prioritize the poor urban youth which is the group more suitable for activation activities. The majority of PASP beneficiaries are between 45-54 years of age. 62 percent of PASP beneficiary household heads are over the age of 45 in Maputo. 16 percent of PASP beneficiary household heads in Maputo and 5 percent of PASP beneficiary household heads in Tete are over 60 years old and hence, should not be eligible for the program. On the other hand, almost no household heads are below the age of 15 in either Maputo or Tete. Only a quarter of beneficiaries in both Maputo and Tete are youth, between the ages of 15-24, which is the group better suited to participate in activation activities. However, it is worth highlighting that almost 60 percent of the households participating at PASP have at least one member between the ages of 15-24, which implies that even though a small share of the household heads participating at PASP belong to the urban youth, an important share of households has at least one member among the urban youth, which is the intended target group for activation activities under PASP. Figure 1.20: Age of Household Heads among PASP Beneficiaries PA SP 35 30 ben 25 efic 20 iari 15 es 10 5 in 0 Ma Between 15 Between 25 Between 35 Between 45 Between 55 Above 60 put & 24 and 34 and 44 and 54 and 59 Maputo Tete o live Source: PASP Registry Data 2014 39 in larger households than non-poor in Maputo, but in smaller households compared to the average poor. The average household size of a PASP beneficiary is 4.5 members whereas a poor household in Maputo has on average 7.5 members. In terms of household composition, only 31 percent of PASP beneficiaries have children under 5 in their households whereas 71 percent of poor households in Maputo have children under 5. Similarly, only 25 percent of PASP beneficiary households have an elderly (over 60 years) member, indicating that PASP beneficiaries have a lower dependency ratio than the average urban poor. Figure 1.21: Share of PASP beneficiary Households with dependents 48% 31% 25% 10% Maputo Tete Children under 5 Elderly Source: PASP Registry data PASP beneficiaries are more likely to be from female headed households. 76 percent of PASP beneficiaries in Maputo are from female headed households compared to only 38 percent of poor people in Maputo who are from female headed households. It is widely believed that women are disproportionately represented in the program because (i) men are not willing to work for low pay; (ii) the kind of work offered, namely cleaning streets, cleaning sewerage canals, cleaning cemeteries, and cleaning other public spaces, is perceived by people to be more suitable for women; and (iii) community perceives female headed households as more vulnerable and therefore prioritize their eligibility for PASP. Youth interviewed in Maputo revealed that they could not be able to do such work due to social coercion. A community social worker suggested that in order to include men, there should be more work that would appeal to men and youth, such as construction projects.23 23Focus Group Discussions with beneficiaries and non-beneficiaries in selected bairros of Maputo conducted in May 2016 40 Figur e 1. 22 76 69 : G en 31 de 24 r of H Maputo Tete o us Female Male eh ol Source: PASP Registry Data 2014 d H eads of PASP Beneficiaries PASP beneficiaries are more likely to be single or widowed rather than married, than non- beneficiaries. 42 percent of PASP beneficiaries in Maputo are single compared to only 2 percent of the poor people in Maputo that are single. On the other hand, 20 percent of PASP beneficiaries in Maputo are widowed compared to 17 percent of poor people in Maputo in the same situation. Finally, only 5 percent of PASP beneficiaries in Maputo are married, which is very low compared to 66 percent of poor people that are married in Maputo. 41 Figure 1.23: Marital Status of PASP Beneficiaries 60 50 40 30 20 10 0 Maputo Tete Source: PASP Registry Data 2014 PASP beneficiaries have fewer assets than non-beneficiaries, except for Maputo. PASP beneficiaries in other urban areas than Maputo have bad houseing conditions and are less likely to have good quality walls (8.4 percent), good quality floors (40.2 percent), access to protracted water (40.2 percent) and access to electricity (16.4 percent) than the average urban poor. PASP beneficiaries in Maputo are better off in terms of asset ownership and they are more likely to have good shelter conditions and access to water and electricity than the average poor. Table 1.6: Asset ownership for PASP beneficiaries Overall Poor Urban PASP PASP Urban Urban Maputo Brick, cement or concrete walls 72.3 62.7 8.4 72.7 Cement, tile or marble floor 59.5 35.1 8.7 79.1 Access to protected water 66.7 52.5 40.2 84.2 Concrete, cement, brick or stone roof 5.0 1.1 28.0 98.6 Access to flush toilet 13.9 3.3 0.0 7.6 Access to electricity 47.0 21.8 16.4 75.6 TV ownership 41.3 16.4 6.5 56.5 Radio ownership 50.1 44.1 7.7 31.0 Ownership of car or motorcycle 11.3 3.0 0.4 1.2 Bicycle ownership 26.7 28.6 6.5 1.5 Computer ownership 4.6 0.3 0.0 3.7 Livestock 8.6 4.4 0.8 0.5 Source: PASP Registry Data 42 Profile of PASP beneficiaries PASP beneficiaries are very poor households (80 percent belonging to the two poorest quintiles). Almost all PASP beneficiaries are workable individuals (not elderly, under 5, students or disable) having a lower level of education than the average urban poor with 57 percent of them not having any formal education at all. The majority of PASP beneficiaries are between the age of 45 and 54 and only 25 percent of them between 15 and 24. However, a very large share of PASP beneficiary households (60 percent) have at least one member between the ages of 15 and 24. PASP beneficiaries live in smaller households than the average urban poor and have a lower dependency ratio (i.e less hildren and less elderly). PASP beneficiary households are very likely to be female-headed (76 percent) and more likely to be widowed or single than married. In terms of access to good quality housing and household assets, PASP beneficiaries live in houses with bad quality shelter and sanitation and have less access to produtive assets (livestock) and household assets (radio, bycicle or TV) than average poor urban households. 43 CHAPTER 2: SOCIAL SAFETY NETS AND ACTIVATION PROGRAMS ASSESSMENT IN URBAN MOZAMBIQUE 44 4 SOCIAL SAFETY NETS AND ACTIVATION PROGRAMS IN URBAN MOZAMBIQUE 4.1 SCOPE AND CONTEXT This chapter provides a review of safety nets and activation programs, and of the legal and institutional framework around them, in Mozambique, with a focus on urban areas. The main aim of this chapter is to review programs, functions, and practices in the area of safety nets and activation interventions in urban settings, with a focus towards identifying existing gaps in coverage. The information presented in this analysis was collected from primary sources (i.e. structured interviews with Government officials), and secondary sources (i.e. reviewing existing programs through published literature and administrative data where available). The chapter begins with an anlysis the legal and institutional frameworks surrounding the delivery of safety nets and activation programs in urban areas, and then it maps the existing safety nets and activation programs in urban settings. This chapter focuses on the core urban safety net and activation initiatives. “Safety nets� refers to non- contributory transfers in cash or in-kind, including unconditional and conditional cash transfers, school feeding, and public works programs targeted to the poor to help them meet their needs in times of crisis24. On the other hand, “activation programs� will refer to non-contributory programs favoring the inclusion of targeted vulnerable population into the labor market. These programs include skills development, entrepreneurship, livelihood support programs, employment services, etc. Other aspects of social protection such as pensions or labor regulations, are not addressed in this chapter as such areas are not part of the forthcoming Government Social Protection Strategy. The review covers national programs with urban coverage and programs implemented at the municipal level. At national level, but specifically for the urban areas, this chapter covers the main public and private interventions under the National Social Protection Strategy (Estratégia Nacional de Segurança Social Básica, ENSSB), the Strategic Urban Poverty Reduction Program (Programa Estratégico de Reducção da Pobreza Urbana, PERPU) and programs from the Ministry of Labor, including the National Professional Training Institute (INEFP) interventions. At the municipal level, the chapter looks at existing Government and donor funded programs. To identify gaps in the current availability of safety nets and activation programs, it is useful to adopt a lifecycle approach to risks and vulnerabilities. Social protection programs should be available to citizens “from the cradle to the grave�25. To achieve this aspirational goal, policymakers should bear in mind that throughout a lifetime, individuals go through different age-stages, each of them associated with different needs and, most importantly, risks and vulnerabilities. For instance, the risk of becoming a widow/er or that of falling ill may increase later in life, whereas the risk of being unemployed is, by definition, borne by citizens during the working age years. Children’s’ malnutrition, on the other hand, has significant and long term 24In Mozambique, safety nets and social assistance more generally are often referred to with the terms social security, even though such expression is normally associated to contributory schemes. Throughout the note, we will follow the country’s official documents and use the term social security to refer to social assistance. 25 Beveridge, W. (1942), “The Beveridge Report�, Inter-Departmental Committee on Social Insurance and Allied Services, The National Archives, Cabinet Papers 1915-1978, www.nationalarchives.gov.uk/cabinetpapers 45 effects on individuals’ development and as such is a risk that needs to be neutralized during pregnancy and early childhood. Figure 2.1: Safety Nets and Activation Programs and Services Through a Life Cycle Lens Source: Adapted from Lindert and George, A presentation on SPL Delivery Systems, November 2016 Rapidly changing risks and vulnerabilities throughout the different stages of citizens’ lives often result in a number of safety nets and activation benefits and services as mitigation measures. Figure 2.1 gives an idea of the many (noncontributory) programs and services that can be made available to ensure that individuals are protected from the various shocks arising through time. Governments should not aim to set the complete universe of program listed in Figure 2.1; rather, they should take a systems’ approach when developing programs and services so that such programs and services are connected to one another, and that all needs specific to the country are covered at different stages. That said, Figure 2.1 provides a useful reference to assess whether substantial gaps exist in the coverage of safety nets and activation programs, and should be adapted based on the local circumstances, capacity, and vulnerabilities. 46 4.2 LEGAL BACKGROUND The 2007 Social Protection Law provides the legal base for designing and implementing basic social protection policies. Such law envisions the delivery of social protection programs and services around three pillars. The first pillar includes basic social assistance, aiming at providing coverage for citizens who are unable to work and/or without means to meet their basic needs. The second pillar envisions an employment- based mandatory social insurance system and the third pillar focuses on a complementary private social insurance scheme. On the basis of the 2007 Law, in 2009 the government issued a social security decree establishing the principles and norms for a basic social security strategy. In 2011, a further decree was issued laying the legal basis for the establishment of basic social protection programs, such as a basic social subsidy program providing regular and indefinite cash transfers to people without the means necessary to meet their most basic needs. Following the issuing of the basic social security decree, a social security strategy was adopted in 2010 and for the 2010-2014 period. The first National Strategy for Basic Social Security (ENSSB I) laid out the implementation plan of the social security law, and was to be the first step towards creating a general and nationwide basic social security system. According to the Strategy, the country’s social security system was to be composed by four pillars: Basic Social Action, Educational Social Action, Health Social Action, and Productive Social Action. The National Social Action Institute (INAS), under the Ministry of Gender, Children and Social Action (MGCAS), was appointed the responsible agency for the implementation of the Basic Social Action pillar. The first strategy was followed by a second one (ENSSB II) for the period between 2016 and 2024 . Following a close review of the implementation of the first strategy, the second strategy set new goals in terms of priorities, coverage, results and expected impacts. Four key objectives were set: (i) strengthening consumption and purchasing power of the poor; (ii) fighting malnutrition and improved access to basic health and education; (iii) preventing and mitigating all forms of abuse; and (iv) creating the necessary institutional capacities for implementing programs. What is commendable about the second strategy is that, unlike the strategies prepared in many countries, the document passed by the Mozambican Government has a clear multi-sectoral approach, and it covers not only social protection themes, but interrelated sectors such as health and education. Unlike social protection policies, activation policies are not organized under a single framework law. Rather, activation policies are embedded in the country’s poverty reduction strategy and social assistance programs, in the country’s rural development strategy and in the recently approved national employment strategy. Such strategy identified several priority areas including: human capital promotion through skills development, job creation (including in the informal economy), employment creation in the agriculture and fishery, manufacturing, mining and energy, and hotel and tourism sectors, implementing a decent work agenda, and the creation of a labor market information system. 47 4.3 INSTITUTIONAL AND FISCAL BACKGROUND Safety nets are sector wide programs, aimed at providing support to a targeted group of citizens, the implementation, coordination, and management of which at the central level is the responsibility of MGCAS. The implementation of most safety net programs is delegated to INAS, the Ministry’s implementing agency. INAS is tasked by MGCAS with the implementation of the country’s programs extending financial support to the poor. INAS, with a total of only 30 delegations across the country, is still under-represented and understaffed considering the vastness of the country and given that they are tasked with the implementation of the majority of safety nets at the national level. In fact, the second social security strategy envisions the expansion of INAS’ presence on the ground and aims to have one INAS delegation in every district at the end of the ENSSB II period. In line with the first social protection strategy, INAS is responsible for implementing four main types of social assistance programs with coverage in urban areas: - The Basic Social Subsidy Program (Programa de Subsídio Social Básico - PSSB), which provides unconditional and potentially life-long monetary transfers to the elderly poor and disabled; - The Direct Social Action Program (Programa de Apoio Social Directo - PASD) providing time-bound monetary transfers to the poor as a shock response mechanism; - The Productive Social Action Program (Programa de Acção Social Produtiva – PASP), a public work program complemented by skills development and income generating activities for poor households with capacity to work. - Social assistance services for vulnerable and abandoned children, elderly, victims of violence and the homeless, as well as managing public nurseries, support centers for elderly and disabled, and the Family Orientation and Reunification Program. The Ministry of Education (MINED) and the Ministry of Health (MISAU) complement the policies and programs implemented by MGCAS. MINED, thorough its Directorate of Special Programs implements for instance school feeding programs for children in overwhelmingly poor regions and areas. On the other hand, MISAU implements programs at the local level aimed at improving nutrition outcomes, administering vaccines and primary health care, and improving access to basic health services. The professional education sector (formal and non-formal), the responsibility for which is shared by the Ministry of Labor and MINED, represents a cornerstone of activation programs. However, despite the reform efforts undertaken through the Professional Education Reform Project (PIREP), the public vocational training system is still relatively small and does not have the capacity to cope with the large number of new entrants into the labour market, estimated at about one million per year. The Labor Ministry, through its National Professional Training Institute (INEFP), manages most of the public skills trainings and many activation measures, especially those strictly related to vocational training and skills development. INEFP, which has the legal obligation to offer free initial training to Mozambican school leavers, operates 16 Vocational Training Centers (CFPs) in urban areas spread throughout the country, and 8 mobile centers mainly providing services in the rural areas of the country, with a total annual capacity of 15,000 trainees. The technical education string, on the other hand, is part of the school education system, and is therefore managed by MINED through the National Directorate of 48 Technical Education (DINET). There is also a much larger, private vocational training network, concentrated in urban areas, and focusing mostly on white-collar professions such as management, accounting, and IT. The international donor community is very much involved in supporting basic social protection and activation programs’ design, funding, and implementation. The work of the international community is coordinated through a social protection coordination group chaired by DFID and covering mostly the work of the MGCAS and INAS, but also extends to programs and policies implemented in the education, health, and labor sectors. According to the annual budget for 2016, the Governments of Canada, Sweden, Ireland, as well as USAID, DFID, Embassy of the Netherlands, and the World Bank, have all contributed to the funding of social protection and activation programs in Mozambique. Finally, the non-Governmental sector, through organizations such as Save the Children or ESSOR, is also actively involved in the social protection and activation areas in the country. Mozambique currently spend less than 0.5 percent of GDP on social protection programs, a figure fairly lower than that of neighboring countries. Figure 2.2 presents the evolution of the expenditures in social programs by the Government of Mozambique (GoM). In this regard, it is worth highlighting that the GoM forecast for 2019 expects to achieve 2.6 percent of the GDP; however this figure is no longer realistic given the recent deterioration of the country’s fiscal situation. GoM has in fact frozen the budget allocated to social protection program. The goal of allocating 2.3 percent of GDP to social action and safety nets expenditure by 2024, putting Mozambique at similar levels with other countries in the region, therefore seems largely unachievable Figure 2.2: Basic Social Security Budget 2011-2015, as a percent of GDP 0.6 0.5 0.4 0.3 0.2 0.1 0 2011 2012 2013 2014 2015 Source: ILO 49 4.4 MAPPING OF URBAN SAFETY NETS AND ACTIVATION PROGRAMS Safety net and activation policies in urban environments fall under four different categories: 1. Policies and programs designed, funded, and implemented by national level agencies, with national (and hence urban) coverage. INAS implemented social assistance programs mentioned above fall under this category, as all programs related to technical education and skills development run by the National Directorate for Education and Training and by DINET and the National Professional Training and Employment Institute (INEFP); 2. Policies and programs designed and financed at the national level, the implementation of which is delegated to the municipal level. The main example of this kind of programs is the Urban Poverty Reduction Program (PERPU), a program supporting locally implemented safety nets and activation programs, whose budget and targets are decided by the central Government through the Ministry of State Administration and Finance, but whose implementation is delegated to the municipal departments of social action; 3. Policies and programs designed by municipal government, financed by municipal budget, and implemented by the local administration. 4. Programs implemented by civil society (NGOs) and private sector (mostly in the activation area). 4.5 NATIONAL LEVEL SAFETY NET PROGRAMS Among social assistance programs implemented at the central level, those run by INAS are most certainly the most important ones. These include the Basic Social Subsidy Program (PSSB-Programa Subsidio Social Básico), the Direct Social Action Program (PASD-Programa Acção Social Directa), and the Productive Social Action Program (PASP - Programa Acção Social Productiva). 50 Table 2.1: Targets and Actual Beneficiaries, Allocated and Executed Budgets of Main National Social Programs, 2012-2016 PSSB PASD PSSAS PASP Executed Executed Executed Executed Year Executed Budget in Executed Budget in Executed Budget in Executed Budget in Benef. Benef. Total Budget urban Total Benef. Budget urban Benef. Budget urban Budget urban urban Total benef. Total benef. urban benef. (MTN areas benef. urban areas (MZM areas urban areas (MZM areas (MZM areas areas areas ‘000) (MZM ‘000) (MTN ‘000) (MZM ‘000) (MZM ‘000) ‘000) ‘000) ‘000) NA NA NA 2012 274,025 NA 979,533 NA 36,831 NA 228,846 4,674 92,807 28,528 1,592 36,661 2,045 NA NA NA 2013 304,576 NA 1,248,118 NA 51,414 NA 330,006 8,971 117,609 30,420 2,000 48,615 3,196 NA NA NA 2014 343,484 NA 1,573,722 NA 52,168 NA 663,433 7,201 153,100 46,885 4,326 125,945 11,620 NA NA NA 2015 359,859 91,017 1,876,916 483,049 57,373 NA 694,186 9,485 102,703 12,499 1,000 43,547 3,484 NA NA 31,693 2016 372,122 NA 1,670,870 NA 49,734 2,235 705,252 8,911 91,564 46,114 7,011 221,422 33,664 Source: INAS 51 4.5.1 The Basic Social Subsidy Program (PSSB) The PSSB, a close to nation-wide social pension currently covering geographically 80 percent of the country. It is the largest safety net in Mozambique, with significant coverage in urban areas. As Figure 2.3 shows, out of the four main national programs, PSSB accounted for almost two-thirds of the budget spent in total, and for almost 80 percent of the beneficiaries of the four programs combined. PSSB provides monthly transfers in the form of old age pensions to poor uninsured people over 60 years of age, and disability pensions to adults below 60 years of age and to youth between 0 to 17 years of age suffering from disabilities and chronic illness. PSSB targeting is community based, with INAS representatives visiting villages and cities to identify those potentially eligible, based on targets, general severe poverty criteria, and local knowledge. The benefit amount is currently set at MZM 310 for a single-member household (approximately 40 percent of the 2014 poverty line of MZM 750), increases with the number of dependents, and can reach MZM 610 for an eligible household composed of five members or more. As Table 2.1 shows, the total program costs (executed budget) rose from about MZM 980 million in 2012 to about MZM 1.8 billion in 2015, and slightly decreased to MZM 1.7 billion in 2016 including operating costs. The budget increase is based on both the expansion of the program but, mainly, is related to the need to compensate beneficiaries for the rapidly increasing inflation rates. Beneficiaries of PSSB have been steadily increasing over the last few years. The program reached just over 372,000 people in 2016, up from 274,000 in 2012. Considering urban areas only, the PSSB reaches 91,017 households (2015), which represents 25.7 percent of the total Program beneficiaries, 14 percent of the urban poor and 56 percent of the Program’s target population in the urban settings (poor urban households with at least one member over 60 or disabled). The budget allocated for the implementation of the Program is urban settings was MZM 483 million in 2016. 52 Figure 2.3: Beneficiaries and Executed Budget of Main National Social Assistance Programs, 2012-2016 100% 100% 4% 8% 8% 8% 10% 10% 7% 7% 6% 3% 13% 11% 13% 10% 17% 19% 26% 12% 80% 80% 26% 26% 60% 60% 40% 40% 80% 82% 77% 76% 78% 73% 72% 69% 63% 62% 20% 20% 0% 0% 2012 2013 2014 2015 2016 2012 2013 2014 2015 2016 PSSB PASD PSSAS PASP PSSB PASD PSSAS PASP Executed Budget Beneficiaries Source: INAS Note: The shares are computed over the total number of beneficiaries and total budget spent across the four programs 4.5.2 The Direct Social Action Program PASD implemented by INAS, is the second largest program implemented nationally, absorbing about one-fourth of the budget allocated to the four main programs, and accounting for about 10 percent of beneficiaries of the programs combined. PASD covers approximately 47,000 households in 2016, up from just below 37,000 in 2012. These may include poor households facing extraordinary expenditures, households in need of food aid, and/or school material, women giving birth to twins or more, women with new born children incapable of breast feeding, etcetera. The targeting of this program is somewhat unclear, being mostly carried out by INAS staff, but lacking a clear definition of poverty status and being often driven by budget availability. Beneficiary households receive a food basket equivalent to MZM 1,200 per month, delivered directly to the beneficiary by INAS personal. PASD support can stretch from 6 months and up to two years, with an option of renewal. 53 Total PASD expenditures rose from MZM 241 million in 2012 to over MZM 705 million in 2016, including the program’s operational costs accounting for around 8 percent. PASD measures are supposed to reach those not covered by PSSB benefits; however, the law allows several benefits to be extended to the same household. At the same time, the generally unclear and often subjective targeting rules of many of these programs (combined with an overall lack of systematized information and poor monitoring around these programs), mean that, de facto, many of these benefits are probably reaching the same households, leaving others vulnerable. 4.5.3 The Productive Social Action Program PASP, implemented by INAS, and supported by the World Bank, which contributes over 90 percent of the total budget, is a small but growing program, now absorbing about 8 percent budget allocated to the four main programs, and accounting for about 10 percent of beneficiaries of the programs combined. The package includes a labor intensive public work program (providing 4 and 6 months per year of work to beneficiaries in rural and urban areas, respectively, for 3 years), complemented by activation measures (skills development and income generating activities) targeted to the work able who are structurally or seasonally unemployed. Total PASP expenditures increased almost tenfold in the last few years, rising from MZM 41 million in 2012 to over MZM 372 million in 2016 (including the program’s operational costs). While the program aimed to employ almost 80,000 people in 2016 (see Table 2.1), PASP currently covers 46,000 beneficiary households (up from 28,000 in 2012); only about 7,000 of which are in 3 urban areas -13.4percent of the total. Within PASP, beneficiaries are preselected at neighborhood (bairro) level through a community- based targeting system, and the lists are later validated by INAS staff through a PMT methodology. As of 2016, selected beneficiaries were paid MZM 650 per month for 16 hours workweeks, an amount that has not been changed since 2011. It is worth highlighting that the PASP subsidy is going to increase in 2017 to MZM 1,000, and will be annually by MZM 200. The new subsidy level (MZM 1,000) represents 28 percent of the average consumption of a poor household. Beneficiaries of activation measures complementary to the cash transfers in urban settings are set to reach 1,500 households over the next 3 years according to the program’s targets. 4.5.4 Social Services and In-care Placement Programs (PSSAS) PSSAS are a set of services implemented by INAS and the Gender and Social action Ministry, the Justice Ministry, the Ministry of Interior, and law enforcement. The package of measures, which reached about 9,000 people in 2016, include measures against all kind of abuses and against social exclusion and social abuse, and permanent or temporary in-care placement of abandoned children, elderly, disabled, etcetera. 4.6 NATIONAL LEVEL ACTIVATION PROGRAMS On the activation side, four main types of programs are currently implemented at national level. The programs are: the Strategic Program for Urban Poverty Reduction (PERPU); Skills Development Programs Implemented by INEFP; Graduation emasures from PASP; and other skills development programs implemented by the public sector other than INEFP and the private sector. 54 4.6.1 The Strategic Program for Urban Poverty Reduction (PERPU) PERPU provides seed capital to startups and employment creating projects for young people living in urban areas. PERPU’s interventions, accounting for just over MZM 115 million over the period 2011- 2015, focus around three pillars. The first one aims at furthering economic activities by funding income generating activities and self-employment projects, mostly in manufacturing and artisanship professions. The second aims at creating a favorable environment for economic activities (e.g. by creating a simplified business licensing procedures, introducing simplified tax regimes as an incentive for informal businesses to enter the formal economy). The third pillar focuses on safety net programs, somewhat duplicating the work envisioned under the country’s basic social security strategy. Table 2.2: PERPU’s Expenditure Urban District # of projects # of jobs created 2011 PERPU Expenditure approved 2011 – – 06/2015 2011 – 12/2015 06/2015 KaMpfumu 200 533 / Nhlamankulu 139 564 / KaMaxakeni 279 918 / KaMavota 235 838 / KaMubukwana 385 912 / TOTAL 1,238 3,765 115,470,000 MZM Source: Ministry of Finance 4.6.2 Skill development programs and access to employment programs offered by INEFP INEFP offers skill development programs through its 16 vocational training centers located in urban areas. The main target groups are young people between 18 and 25 years of age, with a particular focus on school drop outs, those in need for a skill training and job placement, and, to a lesser degree, adults in need of retraining26. There are three different approaches for the skills development programs: (i) formal activation programs that include 3 months of lectures and 3 months of internship; (ii) the so-called alternative activation programs lasting 6 months in total and alternating two weeks of lectures and two weeks internship for the duration of the programs; and (ii) the formação em aprendizem courses, consisting in 6 months of learning by doing in a firm. All INEFP trainees in urban areas, approximately 15,000 in 2015, are automatically enrolled in one of the 11 INEFP run employment centers, although the employment services offered at these centers are still very basic. Bearing in mind the large number of unemployed youth in the country, INEFP’s capacity to provide training is obviously only a fraction of what would be needed in Mozambique. 4.6.3 PASP activation measures (ADGIR) INAS’ implemented PASP is scheduled to offer a bundle of activation measures to labor intensive public works beneficiaries. 8,400 beneficiaries in urban areas are expected to be supported through these measures that aime at reintegrating PASP beneficiaries into the labor market. Such measures will range from skill development programs and vocational training courses run by INEFP to extending start up kits, to granting access to funding. In terms of eligibility of PASP beneficiaries to participate in vocational training 26 In the province of Sofala, INEF has two main urban training centers (in Beira and Dondo). Currently these two training centers offer a total of 30 skills development programs (23 in Beira and 7 in Dondo). The courses targeted young population that have completed at least EP2 degree (second stint of the primary education). The courses have a duration of 3 months and in average a total cost of MZM2,015 (MZM80 month). 55 courses offered by INEFP or by private training providers, there are currently some serious obstacles: on one hand, INEFP lacks the capacity to provide school dropouts with basic professional skills. On the other hand, enrollment in INEFP courses requires students to have completed the 10th grade, thereby in practice excluding many PASP beneficiaries from applying. Table 2.3: ADGIR program targets in urban PASP Municipality Planned # Expenditure (000 Beneficiaries MZM) Nampula 2,013 5,175 Quelimane 1,033 2,655 Tete 893 2,295 Beira 1,873 4,815 Maputo 2,950 6,660 TOTAL 8,400 21,600 Source: INAS Figure 2.4: PASP activation measures (ADGIR) Source: INAS 4.6.4 Other skills development programs. There exist other fragmented skills development programs and access to employment programs offered by the public sector other than INEFP and the private sector. Other than INEFP, the Ministry of Education through the national directorate of technical education (DINET) and, to a lesser extent, other line Ministries, provide skills development activities to about 12,000 people in urban Mozambique. As for the private sector, it is estimated that about 75,000 individuals receive formal or informal training oriented towards increasing employability. 56 Table 2.4: Urban Public and Private Training Provision, 2014 Nr of trainees Men Women Public Providers 27.339 16.382 10.957 Among which INEFP 15.194 9.674 5.520 (2015) Private providers 75.020 51.244 23.776 Total 102.359 67.626 34.733 Source: ESSOR 4.7 MUNICIPALLY IMPLEMENTED NATIONAL PROGRAMS Municipalities and their respective departments of social action do not directly implement safety nets and programs involving regular cash transfer to eligible beneficiaries such as PSSB, but they do play an important role in their delivery. With respect to the public works program (PASP) for instance, while INAS is responsible for the funding, beneficiary selection, and remuneration, municipalities are in practice responsible for the management (execution and control) of the program. PASP, which currently reaches 7,011 beneficiaries in urban areas throughout the country, is being implemented in 5 urban settings, namely: Maputo, Nampula, Beira, Quelimane and Tete, and it is scheduled to be extended to one additional small municipality - Mandlakaze; and envisions a significant role for local administrations. For instance, in 2015, INAS, in cooperation with its Maputo delegation, started implementing activation measures aimed at channeling PASP leavers into income generating initiatives such as skill training measures in blacksmithing, hairdressing, tailoring, and basic informatics and basic computer repair. Participants in training courses are also provided with material and basic tools to start an income generating activity. However, and despite the rather precarious living conditions of PASP participants, this venture did not really match expectations, lacking appeal among the chosen target group. Activation measures at the local level are also envisioned through PERPU, at least on paper. As mentioned above, the PERPU provides seed capital to startups and employment creating projects for young people living in urban areas. Despite being national in nature, PERPU envisions a strong role on the side of local administrations. For the purpose of implementation, PERPU’s strategy document recommends in fact that selected municipalities should, among other things, establish income generating programs focusing primarily on promoting self-employment through professional training, promoting the setting up of associations for peri-urban farmers, service providers and urban craftsmen, promoting access to finance and the creation of small manufacturing business in underserved urban zones. Municipalities should also carry out poverty profiling to identify the most in need and promote the creation of saving groups, in particular among those affected by drug and alcohol abuse. In Matola, PERPU financed activation programs reached a total of 1,800 households and an estimated 9,000 persons. In Maputo, on the other hand, have 1,238 projects received funding, creating 3,765 jobs, and about 30percent of projects financed were small manufacturing ventures. Overall, however, PERPU-based activation programs are currently not very well developed. The strategy paper is, so far, more of a wish list, lacking details, eligibility criteria, without a clear indication of the 57 funds available and the municipalities selected, accountability mechanisms and results indicators. Furthermore, looking at the age profile of PERPU beneficiaries, with an average of 46 and a median of 54, shows that the program, despite targeting all age groups, is actually reaching mostly adults. While this may be in part the consequence of the stringent eligibility requirements (according to which PERPU beneficiaries need to have a convincing business plan and a market competitive skills’ level), it does point to a gap in activation measures for the youth. 4.8 MUNICIPAL LEVEL PROGRAMS Generally speaking, there are no municipal safety net programs other than those programs run and implemented by INAS; however municipalities do have social action departments directly implementing social action measures. Municipal social action can best be described as charity, i.e. in-kind support such as food to selected households, with low coverage, often subjective selection criteria, and questionable long term poverty reduction potential. Municipal social action does typically not have any target in terms of coverage or generosity, but in some municipalities there is (encouragingly) a rather informal coordination with INAS in order to avoid overlaps among measures and programs. Municipal social action departments are often understaffed and suffer from weak capacity. They often carry out work more of administrative nature rather than socially-oriented, and they rarely interact with local associations and NGOs involved in social work at the local level. At best, on the ground cooperation practices consist of informal exchanges of information between municipal and local government agencies. At times, municipal departments channel beneficiaries to INAS and into cash transfer programs. Municipal social action departments are understaffed when it comes to social workers and assistants. This is, at least in part, a consequence of the incomplete process of decentralization in Mozambique. Municipal involvement in activation policies in urban areas is limited to offering skill training opportunities through municipal training and adult education centers, such as in Maputo’s case. The other two big activation activities with municipal involvement are already described above and are channeled through PERPU and through the public works program PASP. Besides these national or nationally funded activation policies there are only little genuine municipal activation policies. In Maputo, the municipality organizes adult education courses in a municipality run adult education center. Furthermore, the municipality of Maputo has begun hiring about half of PASP beneficiaries each year, offering renewable 3 months employment contracts, similar to the original PASP working and payment regime. Municipal work contracts do not include or foresee any capacity building measures, neither in the field of vocational training, nor in that of preparing or channeling municipal PASP workers towards income generating activities. In urban areas such as Maputo and Matola, many social services are implemented directly by municipal administrations. Such services, depending on the municipality, may include: - Providing food aid to vulnerable households; - Promoting gender equality by organizing information campaigns and supporting women associations; - Offering assistance to pre-school children and kindergarten children; - Offering support to street children; - Offering psychological support to women and children and persons who suffered abuses, as well as to drug addicts, ex-convicts, refugees, and so on; - Linking orphans with foster families and running orphanages; 58 - Providing legal support to women and children suffering from abuses and or financial neglect; - Care giving to the elderly; - Granting funeral subsidies to poor households. Overall, however, due to their size (in terms of coverage and budget), the municipal basic social services listed above are, in practice, relief-type interventions. The main goal of these interventions is to provide support to households against the immediate impacts of natural or economic shocks. Municipal social action often includes small-scale food aid packages to vulnerable groups, but it does not include regular safety net measures. The only known cash transfers realized are through a funeral subsidies’ fund covering funeral costs for financially incapacitated households. 4.9 NGO AND PRIVATE SECTOR RUN PROGRAMS Several international partners have been and are involved in the professional training sector in urban areas, contributing about US$ 231 million in total. Table 2.5 below provides a summary of the actors involved and the resources that they contribute to the professional training sector. A more detailed table providing an overview of non-state activation programs is included in Annex 1. - The World Bank is the main financier of the sector, supporting the TVET sector through the PIREP project with USD 67 million aimed at supporting the development of a National Authority of Professional Education, the development of qualification standards, training professors, and supporting innovative investments. - The Italian Government is the second most important supporter of the sector, financing the PIREP and the PIREP+ projects, which cover urban areas and support labor demand in the tourism and agricultural sectors. - DFID is also a significant financier (about USD 24 million) of the professional training sector, mainly through DFI1F resources, and through two main projects. The first project, Skills 4 Employment, aims to support the creation of a network of private professional training centers. The second project, Programa Ligada, aims to identify, test, and support the adoption of measures (such as incentives, facilitating market access, etc.) for the economic empowerment of vulnerable women in urban areas and outside of the formal sector. - The German Government, mostly through GIZ and KFW, provides technical assistance to DINET aimed at strengthening the training of trainers and improving coordination within the sector. It also supports a project promoting technical training among the youth to better prepare them for the labor market. - Canada, through its partnership with the Association of Canadian Community Colleges (ACCC) is implementing the USD 18 million Skills Training for Employment project, which aims to create Technical Training Centers of Excellence in the provinces of Tete and Cabo Delgado. 59 Table 2.5: Sources of Foreign Government Support to Activation Sector Institution / Country Contribution (USD Millions) Details World Bank 67 TVET Italy 52.3 USD 41.5 million for PREDEP and PREDEP + UK 24 USD 15 million from DFID Germany 18.9 USD 15 million from KFW 3.8 GIZ Pro Educacao Project Canada 18.9 Mostly to STEM The Netherlands 6.5 TVET/PIREM 2.5 NUFFIC/FIPAG Spain 8.9 ISDB/PIREP EU 1.3 Support to NGOs ESSOR and SNV 2.5 Support to NGOs PLAN and OIKOS 3.8 Support to NGOs Young Africa and SNV Portugal 5.4 Escolas Profissionais Norway 1.5 PIREP, OIT, IICP Australia 1.6 Escola Carbalo Tete France 0.9 AFD Agha Kahn 11 Total 230.8 Source: Information provided by Mozambican Government Among the projects financed by international donors, many are implemented at the municipal level by international or local NGOs. These are mostly small-scale projects aimed at improving access to training and inclusion in labor markets of young people from vulnerable backgrounds. Among these, a few stand out: 60 a. Activation programs offered by French NGO ESSOR in collaboration with local NGO UPA . Both agencies target disadvantaged youth through decentralized drop-in centers (4 in Beira, 1 in Dondo, and 4 to 5 in Maputo) in which staff offer information and advice to young people looking for skill training opportunities and jobs, free-of-charge pre-training life skill courses. Scholarships to the most needy, depending on the amount of funds available, are also offered. Data provided by ESSOR indicates that 400 youngsters are enrolled in ESSOR/UPA supported three months skill training courses, including three-month long internships. In Maputo and Sofala, ESSOR/UPA offer professional training to people without the necessary educational levels to qualify for vocational training and for those who reside far away from formally established INEFP training centers (under the umbrella of Formação em Aprendizagem programs). This approach offers a window of opportunity to the poorest and most vulnerable with low/no educational background. Trainees can choose from a range of training options (including graphics, hairdressing, construction, etc.) and are issued an INEFP certificated upon the completion of such courses. About MZM 10 million have been spent on these program in Maputo and Sofala, and 1,500 youth trained in the two areas. ESSOR also offers two alternative skill development programs providing firm-based trainings. It is worth mentioning that 35percent of the persons who enrolled in the Formação Em Aprendizagem were hired by private firms after they completed the courses. The individual cost of each student attending Formação Em Aprendizagem is around MZM 4,000 (MTN 666 month) excluding administrative cost. b. Local NGO-implemented projects, supported chiefly by DFID, aimed at granting vulnerable youth access to trainings and jobs. i. The JOBA Program, providing grants to a selected list of private training providers for the organization of training programs for vulnerable youngsters, and envisioning an employment fund for the youth. The courses would consist in long term courses ranging from 1 to 3 years, and short term courses ranging from 3 months to 1 year. It also includes organizing entrepreneurship and so called “second chance� courses. The program, which is still in its initial stage and currently selecting the 10 private training providers, will be implemented in Maputo and in Matola by early 2017 and for 6 years with a total budget of about GBP1 million. ii. The MUVA Program, with a budget of GBP 11 million, which since 2010 aims at empowering disadvantaged youth in urban environments through a variety of measures such as life skill courses, attitude change courses, and by providing beneficiaries with the opportunity of gaining some public sector work experience. c. Local projects supported by the International Youth Foundation (IYF – supported by USAID). These include the Passport to Success Program, preparing youth between the age of 14 and 24 for the labor market by teaching them skills necessary for their future, and the PREPA (Labor Market Preparation for the Youth) Project, which trains 2,000 HIV-positive youth between the age of 15 and 18 in the city of Maputo in view of their participation to the labor market. d. An EU financed project, implemented by Italian NGO AVSI, aimed at increasing the employment potential of up to 3,000 vulnerable youth in Maputo’s Chamanculo neighborhood by providing them with professional training. Many adult non-formal education institutions, predominantly managed by local and international NGOs, and bilateral agencies, provide (free and fee-based) skills trainings and adult literacy education for marginalized youth. In the city of Beira alone, according to a document prepared by the French Development Agency (AFD) and ESSOR, there are 19 professional training centers (other than 61 INEFP centers) supported by ESSOR and AFD offering fee-based courses on a wide range of topics and skills, including hoteling, languages, management, accounting, marketing, etcetera. Unfortunately, the non- formal professional training sector remains under-monitored and understudied, although it could constitute a vital player in organizing activation programs for the marginalized, especially in rural areas. 62 CHAPTER 3: EVALUATING THE READINESS OF SAFETY NETS AND ACTIVATION INITIATIVES IN URBAN MOZAMBIQUE. CONCLUSIONS AND RECOMMENDATIONS 63 5 CONCLUSIONS a. Recent events are resulting in an increased vulnerability in urban areas, which makes a stronger case for safety nets and activation activities in urban areas. Around 3.4 million poor individuals live in urban areas in Mozambique. Moreover, a large share of the urban population is close to the poverty line, making these groups vulnerable to small variation in income, and transient poverty very significant. Political and economic developments in 2016 drastically worsened Mozambique’s near and medium-term prospects. On the fiscal front, large previously unreported external borrowing came to light recently which led to a rapid and dramatic deterioration in Mozambique’s macroeconomic framework. These external account pressures have contributed to the depreciation of the metical by 42 percent against the dollar in 201627, accentuating inflation at a time when rising food costs as a result of the El Niño caused drought already amplified prices. Inflation is already at 20 percent in 2017, and food prices are reported to have increased 40 percent. Urban households are more susceptible to these economic shocks and increases in food prices. For instance, data from 2009 indicate that in Mozambique, it was urban households who actually bore the brunt of the food price and financial crises, partially because they rely more heavily on food and fuel subsidies and thus face higher financial burdens when they are reduced or eliminated than the rural poor28. Urbanization in Mozambique has been a steady process over the last years. In 2005, Mozambique was among the least urbanized countries in Southern Africa, but by 2025, it is projected to be the fourth most urbanized country in the region, after Botswana, South Africa and Angola29 with an annual urban population growth of 3.3 percent30. This urbanization dynamic is likely to continue in the next coming years due to the attraction created by the mining and gas sectors in cities like Nampula and Tete and due to the lack of employment opportunities in the rural areas. This process will potentially exacerbate the unemployment situation in the urban settings, enhancing the vulnerability of the urban poor and specially the urban youth. Youth unemployment in urban settings is particularly worrisome with about 20.4 percent31 of youth in urban areas being unemployed32. The existence of a youth bulge with large numbers of unemployed urban youth is of special concern because if the country fails to provide productive employment, it could adversely affect social cohesion and stability in the country. The combination of the above-mentioned urbanization process and the potential impact of the fiscal measures that the Government of Mozambique will put in place over the next years, which includes dismantling food and fuel subsidies, are likely to result in an increased vulnerability in urban settings. The fuel price increase will impact household welfare and specially welfare in low income households, especially in urban areas. Fuel consumption represents 2.7 percent of the expenditures of the poorest 40 percent of the population. The share of income used on fuel consumption rises to 4.2 percent when including the consumption of fuel-intensive products, such as transportation. 27 World Bank. Mozambique Economic Update. December 2016. 28 World Bank. Higher Fuel and Food Prices: Impact and Responses for Mozambique. World Bank and Ministry of Planning and Development of Mozambique. 2009. 29 Cities Alliance, World Bank. 2014 30 ASPIRE. 2014. 31 INCAF, 2012 64 The expected increase in vulnerability makes a stronger case for the scale up of safety nets and activation initiatives in urban areas. b. Despite the high vulnerability context, coverage of safety nets programs in urban areas is very low. Poor urban households have very limited access to safety nets and activation programs. Despite an increase in coverage over the last years, coverage of safety net programs in urban settings remains very low with programs only benefitting 92,017 households, which represents around 15 percent of the target population in 201533. Moreover, the way current programs are designed seems to exclude one of the main target groups in urban settings, the unemployed youth. Low coverage is likely to be the result of a combination of weak technical capacity to rapidly scale up programs in an efficient manner and low overall budget allocation to social assistance. Total social welfare expenditure was set to reach a total of 0.95 percent of GDP in 2016 according to the second social protection strategy (ENSSB). This forecast, which will most likely not be met because of the recent budgetary crises in Mozambique, puts Mozambique well below other countries in the region. In fact, the current goal of allocating for social action and safety nets expenditure 2.3 percent of GDP by 2024, thereby putting the country at comparable levels relative to its neighbors, seems largely unachievable. In fact, the executed budget for the four main social welfare programs at the national level absorbed only about 0.45 percent of GDP in 2016, thereby pointing to a significant issue of underfunding of social welfare programs. On the other hand, in the last five years, the government had managed to secure donor funding to scale up the programs to additional 30,000 households in urban settings, representing 150,000 individuals. Due to operational challenges related to lack of delivery systems (payment, system, registry of beneficiaries, Management Information System, etc.) and to lack of capable human resources at decentralized level, INAS has only been able to reach 23 percent of the target. The absence of a pure poverty targeted cash transfer program, along the lines of those prevalent in many developing countries, is striking. Many countries are moving towards unconditional cash transfer programs, and this approach has many merits and drawbacks. However, in Mozambique’s case, the absence of a conditional poverty targeted cash transfer to protect households from the risk of falling into poverty is noteworthy, especially if we take into consideration the country’s low rank in the Human Development Index. Mozambique seems to be missing a good opportunity to invest in the activation of the next generation. c. Design and operationalization of safety net programs do not allow to set the path of beneficiaries for activation. Less than 2 percent of all safety nets beneficiaries are working age population. The largest social assistance program, the Basic Social Support Program targets the elderly and the disabled with no capacity to work. The only safety net program targeting the working age population is the Productive Social Action Program. However, after the first 5 years of implementation, the Program is not effectively reaching its objectives due to different reasons: (i) coverage of the program is still very low (only 7,000 beneficiaries); (ii) design features (low benefits and typology of public works) of the program do not attract the participation of the intended target groups; (iii) complementary activities to the cash transfers with the objective of activating beneficiaries 33 INAS. 2015 65 have not been developed nor implemented; and (iv) the program is not establishing links with other activation initiatives as initially planned. The typology of public works implemented under the Program activities which are mainly linked to cleaning the streets are not attractive for a large proportion of beneficiaries who do not see any skill transfer opportunity in their participation. Moreover, the generosity of the program is too low to attract beneficiaries other than destitute households. It is worth to highlight that the program in urban areas was designed to target poor urban youth, however the program due to the aforementioned operational and design constraints has been unable to attract this target group and only 25 percent of PASP beneficiaries are in the age between 15 and 24. This has a collateral effect in the activation activities because the actual group of beneficiaries seems less suited for activation activities. d. Lack of interinstitutional coordination for safety nets and activation programs. Safety nets and activation initiatives in Mozambique are both fragmented and do not work in an integrated manner. MGCAS is in charge of designing and implementing safety nets programs, while the Ministry of Labor and the Ministry of Education are in charge of activation initiatives. Even though coordination bodies such as the National Council for Social Action (CNAS) have been created to promote the coordination between these three Ministries, the truth is that operationally this is not happening. One of the reasons could be related to the lack financial incentives for the sectors and the lack of a financial framework to cover for the costs of coordination and referral from one Ministry to the other. In addition to this, the lack of operational tools such as beneficiary registries makes this coordination and referral even more difficult as information sharing between Ministries does not happen. e. Activation initiatives are small and fragmented. Activation policies are not particularly developed at the national or local level. The only program directly linked to a safety net and favoring the inclusion of a targeted vulnerable population into the labor market is the activation component of the PASP, the implementation of which is delayed and still in its initial stages. Professional trainings offered by INEFP’s 16 urban training centers, and the training programs offered by the Ministry of Education are the only sizeable and central level activation measures offered throughout the country. The private professional training sector, which recently has attracted the attention of donors in urban areas, remains small and its ability to absorb and train young and vulnerable school leavers is still a question mark. Administrative data shows that beneficiaries of activation policies are mostly adults above 40 years of age, even though a large number of youth between 15 and 24 are currently unemployed or underemployed. Overall, the size of activation programs on offer struggles to be an effective response to the steadily growing demographic challenge facing Mozambique, where an estimated one million youth enter the working age population every year. The biggest challenge consists in expanding financially affordable professional training opportunities to all provinces, districts and localities, and adapt the curriculum to cater to the uneducated youth. f. There are significant differences in labor participation patterns among different age-groups and geographical locations that should be taken into consideration for the design of activation initiatives. Labor force participation in urban Mozambique is very high. However, there are significant differences in behaviors and interests of different age cohorts. Poor urban youth is more likely to be engaged in household enterprises and in activities related to commerce, while the older cohort is more likely to access wage employment. Surprisingly agriculture remain the main employment sector for the urban poor in Mozambique. 66 The capital city, Maputo, has also a different pattern compared to other smaller cities in terms of how poor people approach the labor market. Cities like Tete have a very large proportion of the population engaged in agriculture and fisheries, whereby in Maputo the predominant sectors are commerce and services. This age and spatial difference are important to be considered when designing safety nets and activation programs and the skill sets to be provided to poor households to increase their productivity and employability. g. The lack of deconcentration and decentralization of social assistance and activation programs decreases the efficiency and effectiveness of safety nets and activation initiatives. Municipalities in Mozambique have competencies and resources to be involved in the implementation of safety nets and activation policies. However, this has not been translated into action at the grassroots level. Despite existing asocial assistance department at municipal level with a specific budget allocation, municipalities are not actually involved in the implementation of national social assistance and activation programs, which represents a missed opportunity to invest in local know-how and deconcentration of responsibilities. Paradoxically, in some municipalities (i.e Maputo or Tete), INAS provincial delegations and municipal departments for social assistance coincide, but no coordination or complementarity exists between their interventions, which represents another missed opportunity to combine efforts and enhance effectiveness. 67 6 RECOMMENDATIONS To effectively mitigate vulnerability and foster opportunity, social assistance and activation programs need to be able to work together as a portfolio of complementary initiatives, under a systems approach. Systems are “portfolios� of coherent programs that can communicate with each other, share common administrative subsystems, and work together to respond to risks and to deliver resilience, equity, and opportunity to the population34. These systems can be a mix of public and private or informal institutions, ranging from national social assistance programs to NGOs or from active labor market programs to firm-sponsored skills training. The government plays an important role in providing social protection when private and informal mechanisms are insufficient in providing socially desirable levels of resilience, equity, and opportunity. The government also has a clear role in supporting private provision (and, in the case of some large formal actors, regulating them) to achieve these goals. A systems approach involves three levels of engagement: (i) the policy level, the administration level and the program level. 6.1 RECOMMENDATIONS AT THE POLICY LEVEL The policy level ensures policy coherence attuning the entire portfolio of social protection and activation programs to act coherently in delivering national goals and social contracts. In many cases, the policy level driver is a general strategy for the social protection and labor system, outlining a medium-term vision to improve integration and coordination across programs and functions. This vision could map the existing programs and gauge their effectiveness in delivering resilience, equity and opportunity for the population. Mozambique has a good policy framework for social protection and activation, but its operationalization remains a challenge. The government has defined proper institutional arrangements for social protection, including a Social Protection Law (2007) and a second National Social Protection Strategy 2015-2024. This strategy is to be implemented with the oversight of the National Social Action Council (CNAS), which has representatives from all Ministries involved in social protection and activation programs, including the Ministry of Gender, Children and Social Action, main actor for Social Assistance, and the Ministry of Labor, Employment and Social Security and Ministry of Education, which are responsible for activation programs. a. Strengthen the role of CNAS to coordinate social Assistance and Activation Programs. Despite the existence of proper policy documents and institutional arrangements, the coordination between safety nets and activation programs is not effective. The CNAS should play a more active role in facilitating coordination of programs and policies. This coordination would be strengthened by promoting additional synergies and complementarities at administration and program level. In fact, the CNAS should act as a more explicit mechanism to promote coordination not only at central level, but with the municipal governments as well, considering that municipalities are a key role players in the ENSSB implementation. b. Invest in the next generations through Cash Transfers. 34 World Bank Social Protection Strategy for Africa. 68 As mentioned, despite the adequate policy framework, coverage of social protection and activation programs remains very low. Moreover, coverage of poor urban households with children is even smaller with less than 1 percent of this group being supported by social assistance programs. A scaled-up cash transfer program may prove to be a more effective intervention for reducing poverty compared with other policy measures. A growing body of evidence from other countries demonstrates that cash transfer programs can reduce poverty and improve welfare. Cash transfer programs have successfully contributed to reducing the number of people living below the poverty line, and the depth and severity of poverty.35 While distributing cash with no conditions can reduce poverty and improve broader development outcomes, conditions help guarantee that the children in the household will have better human capital when they become earners in their own right, and thus be more likely to escape the intergenerational transmission of poverty36. In this respect, there is a growing body of evidence showing that imposing conditions or co- responsibilities on beneficiaries in exchange for receiving cash transfers can help promote specific positive behaviors. In Tanzania an impact evaluation of the Community-Based Conditional Cash Transfer program showed that grant recipients were healthier and more likely to complete primary school. The cash grant was also used to invest in livestock, children’s shoes, insurance, and for the very poorest it was also possible to realize some modest savings37. An impact evaluation of a cash transfer program in Malawi showed that imposing conditions contributed to better school enrollment, attendance, and test scores among beneficiaries that had to fulfil conditionalities, relative to the group that had not. Other positive effects such as delayed marriage and child-bearing were also recorded38. Conditional cash transfer (CCT) programs have rapidly become one of the most widely adopted anti-poverty initiatives in the world and have the capacity to “activate� the younger cohorts and break the poverty cycle. CCTs are viewed as an effective way to provide basic income support to poor families while strengthening children’s health, education, and nutrition, a cornerstone for breaking the intergenerational cycle of poverty. CCTs provide money to poor families, usually through payments directly to mothers, conditional upon investing in their children. Typical family ‘co-responsibilities’ include prenatal care for pregnant women; regular check-ups, vaccinations, growth monitoring for children under five years old, and school enrollment and attendance for school-aged children. Where conditions or co-responsibilities are imposed, individually and socially beneficial actions can be incentivized, with positive spillovers for the overall economy39. Annex 5 presents different coverage and costing options for scaling up cash transfers. For example, it is estimated that with a budget of 1.1 percent of GDP, 98 percent of the poor in Mozambique could be supported by a cash transfer equivalent to 50 percent of their consumption and the poverty headcount ratio would reduce to 49 percent40. 35 The depth of poverty is also known as the poverty gap, and is defined as the mean shortfall from the poverty line expressed as a percentage of the poverty line. The severity of poverty is also known as the squared poverty gap and is calculated by averaging the squares of the poverty gap relative to the poverty line 36 Grosh, 2008. 37 Evans, 2014. 38 Baird, 2010. 39 Adato and Hoddinott, 2010) 40 World Bank, 2015. From Gas to Cash: Policy Options for Transferring Resource Revenues to Citizens in Mozambique 69 c. Reallocate non-targeted expenditures on subsidies (fuel, wheat, etc.) to targeted social protection programs. In the current serious fiscal crunch, the Government of Mozambique is taking a hard look at its expenditures and a relatively large item in the budget (1.1 percent of GDP in 201441) are fuel subsidies. In Mozambique, almost two-thirds of the total subsidy benefit was received by the richest 20 percent of the population. In contrast, the poorest 20 percent of the population received less than 5 percent of the subsidy share. Thus, fuel subsidies have proven to be largely benefiting the better off. The Government of Mozambique should consider reducing the fuel subsidies and replacing them by more pro-poor, targeted interventions, including cash transfer or activation programs. d. Use of public sector as employer or catalyzer for activation. The public sector has an important capacity to generate and absorb temporary and long-term employment. Under PASP, three successful initiatives were put in place that could have potential to be scaled up to other areas. First, some of the groups of PASP beneficiaries that participated in cleaning and gardening activities in Maputo, where requested to continue providing the same services after their 6-months participation in the program. Their participation in PASP served as an internship for them to establish their own enterprise and continue with an economic activity. Secondly, in Mandjacaze, public works schemes were used to provide social and community services to poor populations and cover the gap of health and social workers existing at municipal level. Beneficiaries were trained by technical staff from the municipal departments of social assistance, education and health and gained important skills that could be used to continue supporting Government in providing these necessary services. Third, under the Ligada program, a pilot was designed to train adolescent girls as teachers’ assistants and reduce the very high teacher-student ratio. All these initiatives combine skill transfer with the provision of paid services by beneficiaries and help Government provide services that cannot be provided due to lack of public servants and the fiscal incapacity to hire additional public servants. e. Further develop activation programs at the national level. From the activation point of view, the biggest challenge consists in expanding financially affordable professional training opportunities to all provinces, districts and localities and adapt the curriculum to cater to the uneducated youth. INEFP’s network, which is greatly under strain, should be expanded, and the capacity of the agency strengthened. A stronger network of public and private training centers, a network of professional schools should be developed, and non-formal education facilities and programs, including functional literacy programs linking literacy and professional skill development, should be established. In line with this, INEFP should also adapt the skills development programs targeting young population that was unable to complete primary school. The so-called Formação em Aprendizagem model might be an option to take into consideration; this kind of approach also demands a high level of involvement on behalf of the private sector in skills development. f. Support to household enterprises through productive inclusion. The household enterprises segment is rarely recognized by governments as a source of productive employment, but data show that a large proportion of the urban poor are engaged in this activity and, 41 IMF. Aide memoire Fuel Subsidy reform. October 2015 70 therefore looking into options to increase their productivity should be a priority. Young people who lack the education for a wage job tend to invest in household enterprises, but they often struggle to amass the savings, skills, and market know-how to enter and consistently succeed with their household enterprises42. Productive inclusion activities, combining technical assistance, skills transfer, asset or cash transfers and extension services for the improvement of household enterprises have shown very positive results in countries like Brazil or Peru, such as increased income of beneficiary households, increased access to productive assets, a higher participation of women in productive activities and an important diversification in the productive activities of beneficiary households. g. Move forward in the deconcentration and decentralization process of the social assistance programs. The new social protection strategy should invest in a more decentralized implementation structure. Municipal departments of social assistance should play a more prominent role in the implementation of national programs, by receiving specific budget allocations based on performance. This step forward in the decentralization would need a proper articulation between INAS and local governments, as well as proper resource planning and financial autonomy formunicipalities which currently almost does not exist. 6.2 RECOMMENDATIONS AT THE ADMINISTRATIVE LEVEL The administration level is about building basic management, information, and service delivery tools. The focus in on developing the “nuts and bolts� subsystems that facilitate the core business processes of social protection and activation programs. These include, for instance, beneficiary identification systems and registries, targeting schemes, M&E arrangements, etc. a. Development, endorsement and operationalization of a Single Registry of Beneficiaries (SRB) for social programs including safety nets and activation. A SRB can help the government operationalize safety nets and activation initiatives as a system and can improve the efficiency of the programs by avoiding duplications, minimizing inclusion errors and facilitating referral to different programs. INAS initiated the development of the SRB, but to date it is only populated with data from beneficiaries from PASP. The largest safety nets program in the country, the PSSB, has never registered its beneficiaries and no beneficiary data bases exist for activation programs. The SRB needs to be the legally endorsed tool for selecting the beneficiaries of multiple interventions. The SRB should register all poor households identified through the targeting process, including the proxy means test verification step. This would constitute a pool of households that are potentially eligible for any of the programs for which the SRB would be used. It is urgent that INAS finalizes the development of the SRB, Government endorses the use of the SRB from a legal point of view making it mandatory for selected programs and Government creates enough technical and human capacity to roll out and manage the SRB in an effective manner. b. Implementation of a common targeting system. The government has defined a common targeting system that is expected to facilitate the identification of beneficiaries for the three core programs of the social safety net in Mozambique (the PASP, the PASD, and the PSSB), but currently it is only used for PASP. The system is intended to minimize exclusion and inclusion 42 Filmer & Fox, 2014 71 errors and increase transparency by using a combination of targeting methods. The information from Chapter 1 can help the government determine what are the most effective variables for categorical targeting. Chapter 1 also shows that poor urban households tend to be concentrated in specific areas in the cities. Therefore, urban poverty maps could be a very efficient tool for poverty targeting in urban settings. For activation programs, more than a targeting methodology, referal systems should be developed so that targeted safety nets beneficiaries could be refered to existing activation initiatives filtering specific profiles that would probably consider variables related to the level of education. c. Setting up an efficient and transparent payments mechanism for social programs. The payment system needs to guarantee that beneficiaries receive their entitlements in a secure, predictable, and timely manner, minimizing the administrative and transaction costs for both beneficiaries and the program. INAS’ current manual payment system has very high operational costs, requires the participation of a large number of INAS staff and does not put in place adequate control and reconciliation mechanisms to ensure that the transfers reach the beneficiaries. Finding a payments service provider for the urban settings in Mozambique should not be a problem as the market of potential service providers includes several options like e-payments through mobile phones, payments through banks, payments through security firms, etc. and it would greatly facilitate scaling up programs by reducing the operational burden on INAS staff and increasing the cost-efficiency of the programs that could be used to increase coverage. d. Mapping of social assistance and activation services (including non-formal education). As discussed, availability of data and information on social assistance and activation programs is weak, as is the management of said data and information. Data availability and sharing, which is key for designing and implementing sound policies, and for accountability reasons, continues to be a key challenge for basic social security programs in Mozambique. Monitoring and evaluation of programs and policies remains scarcely used, and the IT systems of the different programs are disconnected. A greater investment in the management of information systems in the social assistance and activation arena is fundamental to improving the effectiveness of safety nets in the country. As such, carefully mapping social assistance and activation programs in the country, including of the non-formal education sector, both at the national but, most importantly, at the local level, building a database with budgetary allocations, beneficiaries, beneficiaries’ profiles, applications flows, etcetera, will be crucial to determine gaps in the coverage of programs, thereby better tailoring the supply of programs. Making that information available would be a first step to improve communication with beneficiaries on access to social assistance and activation initiatives through intermediation services that would require more capacitated social workers than the so called permanentes, which are community based social workers with low level of education and working on a volunteer basis. e. Outsourcing of safety nets and activation activities. As explained in Chapter 2, INAS and INEFP’s capacity to implement safety nets and activation programs remains weak and because of the current financial condition it might take several years until these two institutions receive adequate technical, financial and human resources to properly implement the programs. In this sense and, to facilitate the scale up of the programs and an efficient implementation of these, INAS and INEFP need to outsource some of the processes and activities of the programs. For instance, the NGO ESSOR has been a very effective partner of INEFP and could support the scale up of activation activities in urban settings. As for INAS, the implementation of the activation activities that are complementary to the cash transfers for PASP should also be outsource to an NGO or to the private sector. 72 For activation initiatives, when using the private sector, there is a “market failure� problem around youth labor market insertion, because firms are chary about investing in training, as the workers can be hired away. Poor individuals on the other hand, don’t necessarily have the capital to finance their own training or the right information about what the market needs. Given the public interest linked to enhancing skill acquisition and youth labor market insertion, one possible approach could be to subsidize expanding firms to train workers they need, subject to commitment to hire them for some minimum period of time after the training is complete (so long as the worker meets the required standard). This might take the form of a grant to an employer to offset part of the cost of training a new employee, conditional on the firm demonstrating that the business is expanding and will need the new job on a permanent basis. Government offices such as INEFP are not likely to be able to manage such an approach and, therefore, an option would be to hire NGOs to identify youth with appropriate potential, deliver training services and place the trainees with private employers with expanding businesses in urban areas. 6.3 RECOMMENDATIONS AT THE PROGRAM LEVEL The program level focuses on design and implementation issues, both within a given program and through synchronization across programs or integration of similar programs. Improving design of individual programs may include the following: better targeting mechanisms for safety nets programs so that they reach the poorest households with activation potential; improved M&E in skills-building programs so that their designs can be adapted to changing labor market needs; etc. An additional step needed for a systems approach is better synchronization of programs, for example, with beneficiary databases communicating with each other to be able to identify eligibility parameters across different types of programs (public works implemented by INAS and skills transfer programs implemented by INEFP for instance). 6.3.1 Recommendations for PASP a. Increase the benefit amount for beneficiaries participating in PASP. PASP beneficiaries receive MZM 650 per month, which covers only a quarter of the food expenses of a 2,000 calorie diet. As mentioned earlier, the subsidy has been re-adjusted to MZM 1,000 in 2017 and will have an annual increase of MZM 200. Beneficiaries complain that PASP benefit amounts are low, only cover for some basic expenses, do not allow for savings and have no potential for productive investments. In fact, it is estimated that it would cost about MZM 2,570 to purchase a 2,000 calorie diet in Maputo, and around MZM 12,850 for an average 5-members family. International experience suggests that cash transfer should represent around 25-30 percent of the average consumption of a poor household and therefore the benefit amount should be set much higher than the MZM 1,000 than are planned for 2017. b. Improve the design and planning of public works so that these can have potential for on-the- job training and skills transfer. The type of work that beneficiaries are assigned to plays a significant role in their engagement with the program and skills development. Focus group discussions from selected bairros in Maputo show that all respondents cleaned the streets and roads of the neighboring area for at least four days a week. Other jobs included cleaning the sewage canals, cleaning the beachfront, cleaning the cemetery and cleaning the local police station. Beneficiaries revealed that the type of work offered was not interesting for many poor individuals and this combined with the low benefit amounts made some eligible households refuse to participate in the program. In addition, the on-the-job training component of PASP is completely missing. 73 Almost all PASP who were interviewed were unaware of the capacity building element in PASP, or when they were aware of capacity building elements, they were unaware of how to access the training. c. Revise the seasonality for the implementation of public works in urban areas. PASP is implemented for 6 months in urban areas and beneficiaries participate in the program for a three years’ period. The logic for this seasonality comes from the experience in rural areas, where households are engaged in agriculture for longer periods of the year and labor-intensive public works are only implemented during the lean season in order not to overlap with households’ productive activities. However, this seasonality does not exist in urban areas (at least in some cities like Maputo). Experiences from other programs in Brazil or Peru, have shown that it is more efficient to provide a more intensive and continuous support during a shorter period of time. d. Provide PASP in urban areas a stronger productive inclusion approach. A productive inclusion approach would include providing skills and grants to poorer population groups, to promote self-employment, entrepreneurship and increase local productivity and enhancing linkages with other existing Government initiatives. The productive inclusion sub-component would be implemented through three main pillars. The first pillar would be a mandatory training/capacity building package to make sure that all productive inclusion participants receive training on soft skills, basic financial management, financial literacy (including savings) and entrepreneurship, including how to prepare a simplified business plan. This pillar is already considered under PASP design. At the end of the training, beneficiaries would develop a simplified business plan for a productive activity based on their capacity and potential returns. The second pillar would be a cash transfer based on performance. Beneficiaries of productive inclusion would receive two asset transfers. The first one would be delivered after approval of the simplified business plan defining the type of productive activity they want to invest in and the type of assets or technical assistance that they would use the asset transfer for. The second asset transfer would be delivered after completion of agreed milestones defined in the business plan that would be related to the purchase of the planned productive assets or services and to other performance indicators more related to project management, such as a proper financial books keeping, etc. The third pillar of the sub-component would be related to technical assistance and extension services. The PASP would need to ensure that all productive inclusion beneficiaries have proper access to extension services or technical assistance for at least two years. Similar experiences in Brazil or Peru, combining technical assistance, skills transfer, cash transfers and extension services have shown very positive results such as increased income of beneficiary households, increased access to productive assets, a higher participation of women in productive activities and an important diversification in the productive activities of beneficiary households. e. Improve targeting processes for PASP. Despite PASP showing very positive results in terms of Project beneficiaries in the first two quintiles of the consumption distribution, the operational assessment of PASP showed that a large share of potential beneficiaries from the lists provided by the local structures where not eligible. This translated into a waste of resources as data were collected for many households that were not eligible (money and time was spent in these processes), delaying the implementation of the program as in order to reach a sufficient number of beneficiaries, a second targeting campaign had to be carried out in some locations. Geographical targeting and community based targeting for PASP could be improved. Related to the geographical targeting, PASP in its first phase of implementation did not prioritize the poorest areas of the municipalities benefitting from the Program. As for the community based targeting, the methodology was not 74 efficient because (i) different to the rural areas, in the urban areas community members do not necessarly know each other due to high mobility of people in the cities; and (ii) because communities’ perception of poverty can be very different to the criteria defined in the program design and therefore communities prioritized the elderly and the female-headed households to participate in PASP and they excluded the unemployed youth. Looking forward, PASP targeting could be improved by implementing a strong geographical targeting through the development of urban poverty maps. The combination household survey data and satellite images are a very strong instrument to develop accurate poverty maps that would allow identifying the areas with higher poverty concentration, with no need of having updated census data. The geographical targeting could be complemented by a categorical targeting based on proxies that can be obtained from the profiles described in Chapter 1 of this report. In this sense, some categorical criteria should include families with three or more children or female headed households. Last, the targeting methodology should continue using a household verification mechanism to be applied to the potential list of beneficiaries as it is currently done. The formula for the verification system should be updated using the latest household survey data (2014/2015). f. Provide regular and predictable cash transfers. Irregular beneficiaries’ payments are common in PASP implementation. This is due to several factors: (i) INAS doesn’t has enough resources to compile beneficiaries attendance to public works in a timely manner in the district to prepare the payment sheet; (ii) INAS doesn’t has an effective coordination link between delegation and central level which derives in continuous delays in the information flow; (iii) lack of payment plan in a systemic way. Ensuring regular, timely and predictable cash transfers would allow beneficiary households to better plan their productive investments. As mentioned earlier, for this to happen, adequate delivery systems (payments systems, registry, identification system, MIS, etc.) need to be developed. 75 REFERENCES Adato and Hoddinott (2010). Conditional Cash Trasnsfers in Latin America. IFPRIl Washington DC. ASPIRE. 2014. The Atlas of Social Protection Indicators of Resilience and Equity. The World bank, Washington DC, Ardnt, Benfica, Maximiano, Nucifora and Thurlow (2008). Higher fuel and food prices: impacts and responses for Mozambique. Agricultural Economics, 39(S1): 497-511. Baird, garfein, McIntosh and Ozler (2012). Effect of a cash transfer programme for schooling on prevalence of HIV and herpes simplex type 2 in Malawi: a cluster randomised trial. Lancet, 379(9823): 1320-1329. Beck, P. (2016), Safety net and activation policies and programs with particular focus on programs implemented in Matola and Maputo. Beveridge, W. (1942), “The Beveridge Report� , Inter-Departmental Committee on Social Insurance and Allied Services, The National Archives, Cabinet Papers 1915-1978, www.nationalarchives.gov.uk/cabinetpapers Evans and Popova (2014). Cash transfers and temptation goods: a revew of global evidence. Policy research working paper 6886. The World Bank, Wahsington DC. Filmer D & Fox L (2014). Overview: Youth Employment in Sub-Saharan Africa. World Bank, Washington DC. Grosh, del Ninno, Tesliuc and Ouerghi (2008). For Protection and Promotion: The design and implementation of effective safety nets. World Bank, Washington DC. INAS (2014). Balanço anual dos programas de Segurança Social Básica. Instituto Nacional de Accao Social, Maputo. INAS (several years). Informe annual. Instituto Nacional de Accao Social, Maputo. INE (2010). Informe Orcamento Familiar 08/09. Instituto Nacional de Estatistica, Maputo. Jones and Tarp (2013). Jobs and Welfare in Mozambique, 2013. Background paper to World Development Report. World Bank (2009). Higher Fuel and Food Prices: Impact and Responses for Mozambique. World Bank and Ministry of Planning and Development of Mozambique. World Bank (2012). Social Protection Strategy for Africa. The World Bank, Washington DC. World Bank (2014). Cities Alliances report. The World Bank, Wahsington DC. World Bank (2015). From Gas to Cash: Policy Options for Transferring Resource Revenues to Citizens in Mozambique. World Bank (2016). “Investing in the Early Years: Country Planning Summary, Mozambique�, unpublished. 76 World Bank (2016). Mozambique Economic Update. The World Bank, Maputo. 77 Annex 1: Data collection tool for PASP Single registry of Beneficiaries 78 79 80 81 Annex 2. Data collection tool for the qualitative analysis in Maputo urban areas Banco Mundial, Programa de Protecção Social, Inquérito aos Beneficiários Grupo focal dos beneficiários de PASP Bairro: Pessoas no grupo focal: Data, hora de início: Entrevistado por: 1. Vocês todos/todas são participantes do programa PASP? Qual é a sua experiência do programa? Como funciona? Em que consiste? Sempre vai trabalhar a mesma pessoa, ou pode ir outro membro da família? Inclui alguma capacitação? 2. Como é que vocês souberam do PASP? 3. Podem contar como foi juntar-se ao programa? Foi fácil ou tiveram algumas dificuldades? Se sim, quais? 4. Sabem explicar como são seleccionados os participantes ao programa PASP? Quem são as pessoas envolvidas na selecção? Qual é o seu papel? Vocês os residentes estão envolvidos no processo de selecção? De que maneira? 5. Sabem quem é aqui a pessoa chamada ‘Permanente’? Qual é o seu papel? Vocês participaram na escolha desta pessoa? 6. Depois de terem sido seleccionados para o PASP, todos vocês foram visitados por alguém do INAS para uma entrevista sobre a sua família, sua casa e outras posses? 7. Qual é a sua opinião sobre o processo de selecção? É justo? Os mais necessitados são incluídos aqui no vosso bairro? Quem são os mais necessitados? 8. Uma pessoa pode fazer queixa se não for seleccionada para o PASP? Como? Conhecem alguém que teria feito queixa? 9. Qual é a vossa experiência sobre a colaboração com o Secretário do bairro e com o Permanente? Funciona bem a organização do trabalho, o pagamento, etc? 10. Como poderia ser melhorado o programa PASP? Ainda há mais algo para acrescentar? 82 Banco Mundial, Programa de Protecção Social, Inquérito aos Beneficiários Beneficiário individual de PASP Bairro: Mulher/Homem, Idade: Nivél de escolaridade: Profissão/ocupação: Data e hora: Entrevistado por: 1. A Senhora/o Senhor é participante do programa PASP? Qual é a sua experiência do programa? Como funciona? Em que consiste? Sempre vai trabalhar a mesma pessoa, ou pode ir outro membro da família? Inclui alguma capacitação? 2. Como é que você soube do PASP? 3. Pode contar como foi juntar-se ao programa? Foi fácil ou teve algumas dificuldades? Se sim, quais? 4. Sabe explicar como são seleccionados os participantes ao programa PASP? Quem são as pessoas envolvidas na selecção? Qual é o seu papel? Vocês os residentes estão envolvidos no processo de selecção? De que maneira? 5. Conhece alguém que participa mas que não deveria participar? Conhece alguém que se candidatou mas não foi incluído? 6. Há pessoas que têm outro rendimento, para além de participar no PASP? Por que é que algumas pessoas não participam? Conhece alguém que foi incluído mas saiou do programa. Sabe porque? 7. Sabe quem é aqui a pessoa chamada ‘Permanente’? Qual é o seu pap el? Vocês, a comunidade, participaram na escolha desta pessoa? 8. Depois de ter sidoselecionado/a para o PASP, foi visitado/a por alguém do INAS para uma entrevista sobre a sua família, sua casa e outras posses? 9. Qual é a sua opinião sobre o processo de selecção? É justo? Os mais necessitados são incluídos aqui no vosso bairro? Quem são os mais necessitados? 10. Uma pessoa pode fazer queixa se não for seleccionada para o PASP? Como? Conhece alguém que teria feito queixa? 11. Qual é a sua experiência sobre a colaboração com o Secretário do bairro e com o Permanente? Funciona bem a organização do trabalho, o pagamento, etc? 12. Quando e como recebe o pagamento? Onde vai receber? Tem alguma dificuldade nisso? Qual? 13. Vamos falar sobre o uso deste rendimento do PASP. Que diferença faz, em comparação com antes de participar no programa? Em que usa este montante quando receber? Consegue poupar? Para que fins? Quem decide sobre como é usado este dinheiro? Quem decide no seu lar? 14. Como poderia ser melhorado o programa PASP? 15. Ainda vamos falar sobre a situação de emprego: dos residentes daqui, quem consegue emprego? Que tipo de trabalho têm? Os que estão desempregados, como é que podem melhorar as suas possibilidades de encontrar emprego?Aqui existe alguém para lhes ajudar? (tipo agencia de emprego, escola ou curso de formação profissional, ou...?) 83 16. A Senhora/o Senhor tentou encontrar outro emprego? O que fez para tentar encontrar emprego? Por que acha que não conseguiu? O que fez para melhorar as suas possibilidades de ter emprego? 84 Banco Mundial, Programa de Protecção Social, Inquérito aos Beneficiários Secretário do bairro/Permanente Bairro: Nome e posição: Data, hora de início: Entrevistado por: Selecção dos beneficiários/participantes do PASP 1. Como é que vocês seleccionam as pessoas que podem participar no programa? Quais são os critérios que utilizam? 2. Neste processo de selecção, a comunidade esta envolvida, ou seja, que as comunidades indicam quem são as famílias mais vulneráveis?De que forma?Ou somente é uma responsabilidade dos líderes locais? 3. Porque é que há tantas mulheres nas pessoas que são participantes do PASP? Estratégias de comunicação 4.Como é que as pessoas ao nível da comunidade recebem informação sobre a existênciado programa? 5. Como é que são informadas sobre as decisões de quem ficou, ou não ficou,aprovado/incluído no programa? 6.Como é que as pessoas recebem informação mais detalhada sobre o programa uma vez incluídas? Por exemplo, quem é que informa os participantes sobre quando irão começar a receber os subsídios do programa, ou sobre quais os trabalhos que vão realizar, ou sobreo local e datas de pagamento, etc.)? Mecanismos de reclamação 7.Caso os participantes tenham alguma duvida ou queixa em relação ao programa, como é que podem resolver as suas inquietações? Houve alguma vez alguém que fez muito barulho para ter informação ou para se queixar do programa? De que é que eles reclamaram? Conhece alguém que foi incluído, mas saiu do programa? Sabe por que? Constrangimentos 8.No seu ponto de vista, quais são os maiores constrangimentos ou problemas actuais do programa? Como é que se poderia ajudar melhor a comunidade aqui neste bairro? Situação de emprego 9. Dos residentes daqui, quem consegue emprego? Que tipo de trabalho têm? Os que estão desempregados, como é que podem melhorar as suas possibilidades de encontrar emprego?Aqui existe alguém para lhes ajudar? (tipo agencia de emprego, escola ou curso de formação profissional, ou...?) 85 Banco Mundial, Programa de Protecção Social, Inquérito aos Beneficiários Não-beneficiário individual de PASP Bairro: Mulher/Homem, Idade: Nivél de escolaridade: Profissão/ocupação: Data e hora: Entrevistado por: 1. A Senhora/o Senhor conhece o programa PASP? Como é que soube do PASP? 2. Já se candidatou? Sabe por que não foi incluído/a? 3. Sabe explicar como são seleccionados os participantes ao programa PASP? Quem são as pessoas envolvidas na selecção? Qual é o seu papel? Vocês os residentes estão envolvidos no processo de selecção? De que maneira? 4. Conhece alguém que participa mas que não deveria participar? Conhece mais alguém que se candidatou mas também não foi incluído? 5. Há pessoas que têm outro rendimento, para além de participar no PASP? Por que é que algumas pessoas não participam? Conhece alguém que foi incluído, mas saiu do programa? Sabe por que? 6. Sabe quem é aqui a pessoa chamada ‘Permanente’? Qual é o seu papel? Vocês, a comunidade, participaram na escolha desta pessoa? 7. Qual é a sua opinião sobre o processo de selecção? É justo? Os mais necessitados são incluídos aqui no vosso bairro? Quem são os mais necessitados? 8. Uma pessoa pode fazer queixa se não for seleccionada para o PASP? Como? Conhece alguém que teria feito queixa? 9. Qual é a sua experiência sobre a colaboração com o Secretário do bairro e com o Permanente? Funciona bem a organização do trabalho, o pagamento, etc? 10. Como poderia ser melhorado o programa PASP? 11. Ainda vamos falar sobre a situação de emprego: dos residentes daqui, quem consegue emprego? Que tipo de trabalho têm? Os que estão desempregados, como é que podem melhorar as suas possibilidades de encontrar emprego?Aqui existe alguém para lhes ajudar? (tipo agencia de emprego, escola ou curso de formação profissional, ou...?) 12. A Senhora/o Senhor tentou encontrar emprego? O que fez para tentar encontrar emprego? Por que acha que não conseguiu? O que fez para melhorar as suas possibilidades de ter emprego? 86 Annex 3: Skill Training Projects from Non State Agencies Organisat Geographi Timefra ion/Agen Project Name Brief Description of Activity Budget cal Focus me cy Technical skills training at 4 medium-level Cabo 2014- CAD Canada STEM Project institutes and one INEFP center with a focus on Delgado 2020 $18M extractive industries. and Tete Providing training to school managers and Entrepreneurshi trainers on entrepreneurship and supporting Niassa and 2013- CAD Canada p in training institutions to develop and implement Nampula 2016 $700,000 Mozambique their own "empresas-escolas" Project to The expected results of the project are: i) 2016 EUR 1-3 France support TVET Creation of public and private partnerships in Inhambane onwards million in Tourism Inhambane province Working with EDM to help them on their Project to formation system. Several studies have been 2017 EUR 1-3 France support TVET National launched; we are now waiting for the first result onwards million in Energy of these studies to start complementary work. EUR F'SASEC French To contribute to vocational training in electricity, Maputo 100,000 Southern African providing relevant equipment and technical (first 2016- France for Scneider electric assistance and short training sessions to the phase). 2019 equipme center teachers. ISDB nt Primary Supporting Primary Education in Manica, Sofala Education: and Inhambane provinces with focus on Maputo, advisory services to the ministry within the Inhambane, Swap, Teacher Training, Capacity Building in Sofala, Pro-Educação Decentralized Education Administrations, HIV Manica; 2013- EUR 23 Germany / Pro-Education work place program; Supporting Technical and TVET: 2017 Million Vocational Education and Training (TVET) Maputo, reform in industrial maintenance with focus on Matola, Teacher Training, School Management and Nampula, Cooperation with the Private Sector Beira, Inhambane Skills for Oil and Gas Africa� (SOGA) is an innovative public-private partnership implemented in four East African countries (Kenya, Mozambique, Tanzania, and Uganda) by Deutsche Gesellschaft für Internationale Zusammenarbeit GmbH (GIZ) and the UK’s SOGA - Skills Germany/ Department for International Development Cabo 2014- EUR 7.9 for Oil and Gas UK (DFID) in collaboration with international oil Deldago 2019 Million in Africa and gas companies and major supplier companies and contractors. SOGA’s vision is to significantly expand the benefits for local people from new oil and gas development in East Africa, by improving access to skilled jobs and economic opportunities. Support to the Contribution to the Mozambican Technical and Mozambican Vocational Education and Training (TVET) Inhambane, 2016- EUR 15 Germany Technical and reforms through the support of selected public Manica, 2019 million Vocational and private TVET institutions in the provinces Sofala Education and of Inhambane, Manica, Sofala. At these selected 87 Training Sector institutions, the Project measures comprise the (TVET) procurement of equipment, consumables and materials needed for the implementation of the competency-based curricula, the rehabilitation, extension and construction of essential buildings. PRETEP - PLUS - Support Strengthen and expand the activities of technical to Technical and EUR assistance and institutional capacity building in 2016- Italy Professional National 36,442.4 support of the Ministry of Education 2019 Training Reform 00 (MINEDH), System in Mozambique Align training provision with the needs of the NGO OIKOS - emerging tourism sector; contribute to reducing Vocational disparities in access to education for women and training and vulnerable groups; promoting environmental educational sustainability, economic and social development programs to of the tourism sector. The project will have the Cabo 2014- EUR Italy increase the rate following specific objectives: to improve Delgado 2017 342,.905 of employment employment of local people by increasing the in the hotels and quality and opportunities of professional training tourism sector public and private sectors in the formal and in Cabo informal; respond to the growing demands of Delgado. the labor market in the tourism sector. Contribute to the improvement of the living conditions of Mozambique's population, with particular attention to young people with limited access to decent work opportunities. Expand NGO Maputo and 2016- EUR 1.8 Italy opportunities for access to the formal labor ProgettoMondo Inhambane 2018 million market, particularly for women and youth, improving the level of short and long duration professional technical training (FPLD) offered in the Maputo and Inhambane Provinces. The Project is to strengthen INEFP institutional capacity and to enhance vocational training Enhancing the institutions by establishing educational Maputo, Professional management control system, as well as Zambezia, 2017- Japan Education improving the training quality in order to Nampula, -- 2021 System Model in establish a more strategic human resource Sofala or Mozambique development that reflects the needs of the Manica private sector as well as measure to address the issue of youth unemployment Supports all Professional Schools of level II - basic level (public, private and community schools, which are around 50 in the country, by Project to 2001 - Portugal providing training to school's director and National -- Support TVET ongoing teachers, as well as to officers of the Ministry of Education and Human Development (at national and provincial level). Cluster of the Portuguese Project to contribute to the integral development Ilha de Cooperation in of Ilha de Moçambique by enhancing the Moçambiqu 2011 - Portugal -- Ilha de human, cultural and environmental potential of e- ongoing Moçambique the island. Nampula (2nd phase) 88 National JOBA Achieve higher income among targeted youth, (starting in (previously Skills especially young women and girls, acquired Maputo, 2015- GBP UK for Employment through increased access to relevant quality non- Gurue, 2021 17m (S4E)) state vocational training leading to formal or self Dondo, programme employment Tete, Chimoio) Promote decent employment for girls by creating UK Ligada linkages between skills training, employment -- opportunities and girls in urban areas More and better jobs in Cabo Strengthen INEFP systems to contribute to an Cabo Delgado and improved and more egalitarian workforce with Delgado Nampula: vocational skills and competencies, improving 2015 - US$ 1.5 UN and Harnessing the their employability opportunities in Extractive 2017 million Nampula opportunities of Industries firms and in SMEs operating in value provinces Mozambique’s chains feeding into the EI sector New Economy Improve the access of persons from local communities to training and to opportunities to Skills for turn training into productive livelihoods. It aims Employment and to reach a broader base of job creation, along the Cabo Dec 2014 US$ 1.8 UN Productivity in value chains in the extractive industry, through Delgado - Aug million Low-Income an accelerated development of human capital, Province 2017 Countries focusing on the most vulnerable groups, especially young people and women in Cabo- Delgado Province Support the ongoing reform of the Mozambican Technical and Vocational Education and Training Sector (TVET), contributing for the consolidation of the institutional reform and Higher activities related to qualifications and quality Education control of teaching and learning in TVET ( 2015- US$ 7 UN Science and (i.e., support to governance and financing national 2018 million Technology mechanisms, training and provision of Project scholarships for teachers, development of qualifications in priority areas). The current support builds on the support provided through the TVET Project that closed in September 2015. Support the Development of Establish and test a new concept of TVET the Oil and Gas targeting the Oil & Gas sector in Cabo Delgado Cabo 2013 - 7.0m Norway Technical and with participation from both Government and Delgado 2016 NOK Vocational the industry – Pilot Phase (PIREP) Training - TVET - Pilot I Upgrading the Agrarian To strengthen the agriculture sector in Cabo Cabo Institute of 2016 - Norway Delgado through increased availability of well- Delgado - -- Bilibiza in Cabo 2021 trained agriculture technicians and entrepreneurs Bilibiza Delgado, Mozambique 89 Annex 4: National Level Programs in a Life Cycle Approach Table A.1 provides a summary of the policies and programs available at the national level, matched against the various life-cycle risks; while many risks are partly covered, what is most notable from Table A.1 is the absence of a pure poverty targeted cash transfer program, along the lines of those prevalent in many developing countries. Most social assistance programs appear in fact to have one or more categorical elements beyond the requirement of being poor, which is in many case loosely defined and not assessed using objective criteria (such as, for instance, a proxy means testing like in the case of the PASP program). This gives rise to the risk of excluding many poor individuals and households from the social assistance system. Many countries are moving towards unconditional cash transfer programs, and this approach has many merits and drawbacks. However, in Mozambique’s case, the absence of a conditional poverty targeted cash transfer is especially striking, given the extremely high prevalence of stunting (43percent - the fourth highest in Sub-Saharan Africa), the country’s low rank in the Human Development Index (180th out of 188 countries), and the 4percent pre-primary school enrollment rate at the national level, based on World Bank data. Table A.1: Risk and Vulnerability Patterns by Life Cycle and Policy Response at the National Level Target Age Risk and Management Group / Policy program response vulnerability responsibility Category Plan to design and introduce three new Health and safety net programs (cash transfers, Pregnant food aid and complementary social mortality risks for women and services to young children, orphans, mothers and Health Ministry small children children in poor households, and to children. Chronic until 2 years households headed by children. and acute mal Gender and Social action of age nutrition Ministry and INAS Support to poor students enrolled in primary schools and articulate with MGCAS, INAS plus other safety net programs Health Ministry, Education Ministry, plus School fee exemptions to poor families SETSAN for implementing articulated Make provisions (plan) for special complementary measures Delayed education needs and enhance measures (nutrition) and for M&E Children a promoting their inclusion into ‘normal’ enrolment. Lack school age. schools of school Education Ministry Orphan performance and children and Prepare social services for adopting drop out. Early Justice Ministries (registry children preventive and follow up measures pregnancy and offices, law Enforcement separated against any kind of abuse directed marriage. Child from their against children, including abuse and child NGOs parents. implementing measures against labour. premature pregnancies and marriages Promote children rights and child birth registration Not included in ENSSB strategy but INAS, INEFP, private Adolescent activation programs: access to skill TVET providers and Unemployment Youth training and to employment donor organization 90 New employment policy PASP Activation Measures and other activation programs Public works program PASP in rural and urban areas. Geographic targeting focusing on poor districts with high unemployment rates Un- and (seasonal) Gender and Social action underemployment. Include disaster prevention and Ministry and INAS Low and irregular Adults management component in PASP and earnings and make necessary revisions. INEFP income. Include graduation/ activation measures for PASP beneficiaries Other activation programs Loss of income Extension of old age pension cash Ministry of Finance, generating transfers beneficiaries by 10percent per Gender and Social action capacity. Health Elderly (60 +) year reaching 62percent coverage by Ministry and INAS related 2024. expenditure Revised PSSB providing unconditional regular cash transfer to elderly, chronically ill and incapacitated through Ministry of Finance, extending eligibility by abandoning the Gender and Social action one person/household principle Ministry and INAS Chronically ill Reduced income Extend coverage through PSSB to and disabled District Service of generating 49percent until 2014 people Gender and Social Action capacity, (children and Ministry exclusion, and PASD and social services persons up to stigmatization. age 59). Law enforcement and, Anti discrimination measures against and NGOs persons and households with HIV/AIDS Ministry of Finance, Social services including preventive and Gender and Social action follow up measures against any kind of Ministry and INAS Gender related abuse directed against women and girls, exposure to Girls and including cases of domestic violence District Service of discrimination, women Gender and Social Action inequality and Review rules of attending victims and Ministry violence use social action beneficiary cards and record books Law enforcement and, and NGOs Early childhood program, PSSB, PASP Ministry of Finance, and PASD Gender and Social action Poor Ministry and INAS Households Review PASD eligibility for including Poverty and notably those individuals and groups affected by Ministry of Health vulnerability to in remote natural disaster natural and areas with no District Service of economic crises or little access Review and reinforce health service Gender and Social Action to basic fees’ exemptions for social security Ministry infrastructures beneficiaries by introducing a social beneficiary identification card and Law enforcement and, 91 record book and NGOs Organization and subsidization of poor peoples’ hospitalization and funeral costs 92 Annex 5: Coverage and Costs of Cash Transfers in Mozambique Total Cost as a % of GDP Eligibility cut- Transfer amount per person per year: % Impact on Poverty % of Pop. % of the Population targeted off point of median consumption of the poor (in (2009 National Poverty Rate: 52.08%) Targeted 2015 US dollars) US$ Probability of Below US$ US$ US$ US$ US$ US$ US$ Coverage Coverage Coverage 85.50 being poor Coverage food 171.00 256.00 342.00 85.50 171.00 256.00 342.00 Leakage of bottom of bottom of bottom (50% (greater than of poor poverty (100% (150% (200% (50% of (100% (150% (200% 20% 40% 60% of or equal to) line of p50) of p50) of p50) p50) of p50) of p50) of p50) p50) 0 100.00 47.92 100.00 100.00 100.00 100.00 1.28 2.57 3.85 5.13 48.36 44.69 40.81 36.56 100.00 5 98.86 43.80 99.03 99.27 98.95 98.54 1.12 2.25 3.37 4.49 48.48 44.95 41.18 36.98 91.61 10 96.89 41.87 97.76 98.73 97.55 95.91 1.04 2.07 3.11 4.14 48.91 45.52 41.82 37.85 86.80 15 95.07 39.83 96.05 97.60 95.70 93.42 0.96 1.91 2.87 3.83 49.02 45.84 42.34 38.48 82.29 20 92.09 37.77 93.34 95.69 92.57 89.63 0.87 1.74 2.62 3.49 49.23 46.20 42.99 39.35 77.07 25 87.67 35.84 89.11 91.93 87.97 84.75 0.78 1.56 2.34 3.12 49.53 46.87 43.87 40.50 71.15 30 82.87 33.99 84.95 88.21 83.35 79.52 0.70 1.39 2.09 2.78 49.77 47.62 44.92 42.03 65.38 35 77.90 31.98 80.04 84.58 78.52 74.11 0.61 1.23 1.84 2.46 50.04 48.04 45.66 42.98 59.64 40 71.60 30.30 73.87 78.72 72.52 67.85 0.54 1.08 1.62 2.15 50.36 48.50 46.53 44.14 53.50 45 63.93 29.27 66.71 72.13 65.18 60.16 0.46 0.92 1.39 1.85 50.75 49.16 47.54 45.52 47.07 50 57.55 27.71 60.36 65.78 58.72 53.72 0.39 0.79 1.18 1.58 50.91 49.57 48.21 46.56 41.46 55 49.45 26.21 52.33 57.33 50.29 45.65 0.32 0.64 0.96 1.28 51.14 50.05 49.09 47.83 34.90 60 41.32 24.80 44.31 47.54 41.85 38.02 0.26 0.51 0.77 1.02 51.35 50.61 49.82 48.79 28.61 65 33.28 23.13 35.59 38.99 33.24 30.30 0.19 0.39 0.58 0.77 51.63 51.00 50.37 49.60 22.55 70 25.77 22.77 27.38 30.06 25.48 23.37 0.15 0.29 0.44 0.58 51.74 51.25 50.84 50.23 17.38 75 18.15 21.17 19.34 22.25 17.99 16.25 0.10 0.19 0.29 0.38 51.89 51.61 51.33 50.93 11.99 80 12.04 19.38 13.28 15.41 12.26 10.54 0.06 0.12 0.18 0.23 51.98 51.85 51.74 51.55 7.78 85 7.04 16.52 8.03 9.99 7.47 6.15 0.03 0.06 0.09 0.13 52.07 52.01 51.98 51.86 4.39 90 2.93 …* 3.61 4.90 3.29 2.55 0.01 0.03 0.04 0.06 52.08 52.06 52.06 52.05 2.04 95 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 52.08 52.08 52.08 52.08 0.00 100 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 52.08 52.08 52.08 52.08 0.00 93 94