DISCUSSION PAPER NO. 1614 Urban Safety Nets – Experiences from Three Countries: Benin, Republic of Congo, and Mali Vanessa Moreira and Ugo Gentilini December 2016 Urban Safety Nets – Experiences from Three Countries: Benin, Republic of Congo, and Mali Vanessa Moreira and Ugo Gentilini December 2016 Abstract: As countries implement social safety nets programs, a range of technical hurdles affects their implementation differently in rural and urban areas. In urban areas, the focus of this study, living in is expensive and more vulnerable at economy slowdowns. Poverty can be more severe than in rural areas and accompanied by high malnutrition rates. Challenges faced by poor populations in most urban areas related to the lack of proper identification, outreach, intake and registration of potential beneficiaries in part due to the lack of social cohesion and the existence of multiple channels of communication, challenging the delivery of any messages quickly and efficiently. Therefore social workers have a fundamental part during program implementation and M&E process. JEL Classification: Keywords: Social pension, non- contributory cash transfer program Acronyms and Abbreviations CBT: Community-based targeting CTC: Community targeting committee M&E: Monitoring and evaluation MIS: Management information system NGO: Non-governmental organizations NSO: National statistical office PMT: Proxy Means Test PMU: Project Management Unit SAM: Severe acute malnutrition SSN: Social Safety Nets UNICEF: United Nations children’s fund BENIN CPS: Centre de Promotion Social INSAE: National Institute of Statistics and Economic Analysis of Benin MDGLAAT: The Ministry of Decentralization, Local Government, and Administration and Development of the Territory PSDCC: Projet de Services Décentralisés Conduits par les Communautés SSDCC: Secretariat for Decentralized Community Driven Services - Benin CONGO, Republic MASAHS: Ministry of Social Affairs of Humanitarian Action and Solidarity CAS: Social Welfare Districts CCS: Targeting community committee GAC: Governance and anticorruption INS: National Statistics Office i MALI OM: Orange Money BNDA: National Bank for Agricultural Development INSTAT: National Statistical Office SAM index: Severe Acute Malnutrition UTGFS: SSN UGTFS Technical Management Unit ii Table of Contents ABSTRACT:...................................................................................................................................I I. INTRODUCTION .................................................................................................................. 1 II. NATURE OF URBAN POVERTY ............................................................................................ 3 III. PROGRAM OBJECTIVES ...................................................................................................... 7 IV. KEY PROGRAM DESIGN FEATURES ..................................................................................... 7 4.1. PROGRAMS DESIGN SPECIFICITIES ..................................................................................... 8 V. KEY STEPS IN THE DELIVERY PROCESS ............................................................................. 11 5.1 OUTREACH & INTAKE ....................................................................................................... 11 5.2 SELECTION ........................................................................................................................ 13 5.3 TARGETING / REGISTRATION ........................................................................................... 16 5.4 PAYMENT / TRANSACTIONS ............................................................................................. 18 5.5 MONITORING &EVALUATION (M&E) ............................................................................... 20 5.6 GRIEVANCE & REDRESS .................................................................................................... 21 VI. KEY ISSUES EMERGED DURING PROGRAM IMPLEMENTATION IN URBAN AREAS .......... 22 VII. URBAN AREA CHALLENGES .............................................................................................. 23 VIII. EMERGING POLICY & OPERATIONAL OPPORTUNITIES .................................................... 25 IX. REFERENCES ..................................................................................................................... 27 ANNEX 1: BENEFIT LEVELS ...................................................................................................... 28 ANNEX 2: 3 COUNTRY COMPARISON ..................................................................................... 29 Figures Figure 2. 1: GDP, PPP and Urbanization trends ........................................................................ 6 Tables Table 2. 1: Key indicators .......................................................................................................... 6 Table A1. 1: Distribution of benefits in Benin ......................................................................... 28 Table A1. 2: Distribution of benefits in Mali ........................................................................... 28 Table A1. 3: Transfer amount by household composition ..................................................... 28 iii I. Introduction As countries implement social safety nets programs, a range of technical hurdles affects their implementation. Given the fluid expansion and contraction of urban informal settlements over time and their specific characteristics, the development of an effective outreach strategy for informing and attract urban poor for safety nets programs or to effectively communicate with prospective beneficiaries about available programs is challenging. One challenge faced by poor populations in most urban areas is related to the lack of proper identification (ID), therefore being deprived of access to public services and programs that demand ID for verifying identity. Obtaining IDs can be costly, incurring private and administration costs and, a lot of times, travels back to their villages of origin. Interestingly enough the three countries in this note have not described problems in this regard. In Benin, the program created its own identification as the national ID card has limited coverage. In Congo, newborn children leave the hospital with a birth certificate (provided guardians have paid the fee of FCFA 5,000). Lastly, Mali benefited from the recent elections to assure individuals had proper IDs. One key process that is somehow weak in the urban areas presented in this study is related to the outreach, intake and registration of potential beneficiaries. The challenge of communicating with potential beneficiaries and reaching pre-identified families is problematic in urban areas because of the larger catchment areas and the multiple communication channels available, minimizing effectiveness of communication campaigns. The same occurs regarding payments messages. In contrast, in rural areas, the existence of social cohesion together with a single community radio station/ church/ community leaders enables to convey the message quickly and efficiently. Social workers are a fundamental part of program implementation and M&E process. Due to larger catchment areas and the lack of social cohesion in urban areas with high density of 1 people, families’ identification and monitoring can be challenging and community committees are not too effective without participation of social workers. A strong and transparent system for identification and selection of beneficiaries is key to ascertain credibility to the program. In urban areas, this can only be achieved with strong participation of social workers who know the population of interest. Community committees composed of youth associations, churches, community leaders, etc. tend to identify participant families that are not necessarily the poorest therefore increasing inclusion error. The poorest are often marginalized and excluded from such committees. In urban areas, data collection and monitoring takes longer. Enumerators and M&E agents have difficulties returning to pre-selected households due to the bad quality of addresses recorded (household disorderly numbered) and the differences in recorded names (household members known by a nickname) causing enumerators to perform multiple visits to households in order to correctly identify families. GSM payments are relatively uncomplicated and inexpensive. However, electronic payments transactions that are not fully functional increase accountability time. Poverty can be more severe in urban areas. On top of the scarcity of basic services as sanitation, urban poor have difficulty to have access to good schools and health centers because families’ income are constrained. Urban areas have high malnutrition rates. Individuals are subject to precarious and temporary jobs, as a way of generating income for meeting the end needs. Also, differently from rural areas, families cannot produce for subsistence. Therefore, food intake can be compromised. Living in urban areas is expensive and more vulnerable at economy slowdown. Apart from having to pay for rent and some basic services otherwise free in rural areas (for example, health and education), urban poor are subjected to higher commodities prices, and face 2 higher unemployment rates. At any economy slowdown, social unrest and violence increase mainly in the poorest urban areas. In this study we will look into urban and rural social safety nets programs specificities from three countries: Benin, Republic of Congo, and Mali. Which kinds of strategies were used to make the social safety nets programs adapted to the urban context and more effective to fight urban poverty. In the next sessions, we will present: (ii) nature of urban poverty; (iii) programs objectives; (iv) key program design features; (v) key steps in the delivery process; (vi) key issues emerged during program implementation in urban areas; (vii) emerging policy & operational opportunities; and (viii) challenges. II. Nature of Urban Poverty Rural areas in all three countries are largely agriculture dependent and commonly affected by nature shocks such as drought and floods. Therefore, poor people from rural areas move to urban areas to escape such difficulties but instead of finding relief, they are subject to worse conditions. For example, some public services, which are available for free in rural areas, are unavailable or demand payment in urban areas, such as schools and hospital, which represents additional costs for families. Poverty can be more severe in urban than in rural areas. On top of the scarcity of basic services as water, electricity and sanitation, the poor in urban areas have greater difficult in accessing good schools and health centers as families’ income are constrained. Urban poor are subjected to higher commodities prices, higher unemployment rates, and at any economy slowdown, social unrest and violence increase mainly in poorest urban areas. Moreover, in urban areas, individuals are subject to precarious and temporary jobs as a way of generating income for subsistence. Food intake can be compromised. In contrast, in rural areas, most families are less exposed to economic slowdown because many can produce or have subsistence farming so that they are able to provide food for the family. 3 BENIN Benin’s economy relies mostly on the cotton trade, and agriculture is the main source of income for 70 percent of the country’s workforce making the country very susceptible to nature shocks. In 2010, Benin experienced awful losses due to floods: thousands of people were forced to leave their houses, many schools and public buildings were damaged or destroyed and most crops were lost, worsening the food insecurity situation. Difficulties in rural life underline continuous migration to urban areas, where migrants often settle in underdeveloped areas with high concentration of people. Poor urban household are often poorly built, rented accommodations without access to water, electricity, and sewage. Urban areas of Benin have numerous pockets of poverty. People leaving in those areas lack access to basic services. Food insecurity is high as a consequence of irregular sources of income and the lack of a family production whereas in rural areas families can produce something to eat. Violence is widespread; women are subject to prostitution and various sorts of sexual abuse. High indices of child labor are observed as children work to help increase household income. Social unrest is observed as a reflection of diminish jobs in the formal sector. CONGO Since oil’s discovery in the early 1980s, oil revenues and considerable external assistance have motivated the economy in Congo. The formal economy was centralized by the state, and as a result, public spending and relatively high wages in the public sector attracted migration to Brazzaville and Pointe Noire, important urban areas. Nonetheless, oil wealth was not shared properly and poverty is still persistent with rates standing at 40.9 percent in 2011. However, almost 50 percent of the poor live in urban areas due to the high urbanization rates of the country. Gini inequality is high at 0.46. The sustained growth over 2000-2014 did not translate much into poverty and inequality reduction despite a slight decline in urban poverty. This lack of inclusiveness results from an economy driven by offshore oil production ruled by an elite with a negative home bias for investment which has installed patronage through a non-transparent job market, weak health, education, and 4 social protection services delivery. Notice also that urban mobility is high in certain areas (pockets of poverty) due to violence, lack of services and the cost of leaving (rental/ food prices). Formal jobs, which are a leading factor of economic inclusion in Congo, are concentrated in the hands of a few. The Congolese labor market has a very small formal sector, dominated by the public sector and state-owned enterprises, leading the poor to informal activities that are not sustainable, mainly in urban areas. Urban poverty rate in Congo is rather small. But the high concentration of people leaving in poor urban areas has negative implications regarding people’s access to the already weak health and education services. For example, reflected by higher rates of children per teacher in public schools compromising education quality. MALI Political and social situation in Mali remained fairly stable until March 2012, when a military coup ousted the previously elected president. The presence of armed separatists and jihadists groups in the north caused massive internal displacements, mostly to urban area in the south and neighboring countries. However, poverty is still predominantly rural where people depend on subsistence farming and are therefore vulnerable to natural disasters. Conflict in Mali nowadays is exacerbated by climate change and population growth. In urban areas, people’s livelihoods are under stress mainly due to limited opportunities (jobs and services), and persistently high levels of youth unemployment, an easy feeder into the informal economy. Therefore profound challenges remain due to economic fragility: growing insecurity, rapid demographic growth and the impact of climate change on agricultural productivity. The later, leading to food insecurity and migration to urban areas, with direct consequences in the population’s access to health, education and social protection services delivery. 5 Figure 2. 1: GDP, PPP and Urbanization trends Source: WDI, World Bank Table 2. 1: Key indicators Indicator Country 2010 2011 2012 Benin 2.12 2.96 4.64 GDP growth (annual %) Congo, Rep. 8.75 3.42 3.80 Mali 5.81 2.72 0.03 Benin 732.95 799.04 807.69 GDP per capita (current US$) Congo, Rep. 2,953.19 3,453.22 3,191.16 Mali 610.79 680.83 641.79 Benin 1,710 1,750 1,810 GNI per capita, PPP Congo, Rep. 4,120 4,180 4,470 (current international $) Mali 1,440 1,460 1,430 Poverty headcount ratio at Benin .. 36.20 .. national poverty lines (% of Congo, Rep. .. 46.50 .. population) Mali .. .. .. Benin 9,509,798 9,779,391 10,049,792 Population, total Congo, Rep. 4,066,078 4,177,435 4,286,188 Mali 15,167,286 15,639,115 16,112,333 Benin 16.97 16.37 15.81 Population in the largest city Congo, Rep. 61.23 61.38 61.54 (% of urban population) Mali 35.41 35.42 35.47 Source: WDI, World Bank (last Updated: 05/02/2016) 6 III. Program Objectives Although main objectives are similar when implementing a social safety nets program in the three countries considered, additional objectives are specific to the countries’ needs. While all three countries are interested in establishing an efficient social safety nets system to protect the poor and/or vulnerable while reducing current poverty and smoothing household consumption, in Benin the program Projet de Services Décentralisés Conduits par les Communautés (PSDCC) aims to also improve household’s ability to cope with shocks, while in Congo, the Lisungi program also aims to promote children’s access to education and health services and in Mali, through the Jigiséméjiri, the effort is also for protection of food insecurity households, due to particular high rates of food insecurity in the country. Programs vary in size and implementation areas. Benin and Mali programs are relative large in scale (especially when compared to previous programs existing in the countries) but only a small part of the benefits is destined to poor household in urban areas (main urban areas of both countries are included in respective programs, namely, Cotonou, Benin’s capital city and Bamako, Republic of Congo’s capital city). On the other hand, Lisungi, which is still in its pilot phase, is mostly implemented in urban areas due to the country’s urban population density and to program’s desired conditionalities. To guarantee that households can comply with school attendance and regular health care visits, minimum infrastructure was required for implementation, which proved to be hard in rural areas. IV. Key Program Design Features Programs’ outcomes are directly related to implementation. Hence, design issues become extremely relevant when implementing a determined program in any country where regions / areas have distinct characteristics. If such characteristics are not considered, the entire project and its relevance may be compromised (for not producing good enough results). 7 In general, community committees are set to facilitate the process but they are not working as expected in urban areas. Due to larger catchment areas and the lack of social cohesion in areas with high density of people, identification of households and individuals in urban areas can be challenging. Therefore, social workers become a fundamental part of the identification process, as they are familiar with the families that ask for regular help. Ideally, every community committees would have local social worker within its members, as their knowledge of the families can help minimize both errors of inclusion and exclusion. But this is not always the case. Community committees are often composed of youth associations, churches, community leaders, etc. This structure works well in rural areas where families are close together, but not so much in urban areas. Identification of poor families in urban areas takes longer. In response to urban areas specificities social workers revise the preliminary lists formulated by community committees. The limited number of social workers prevents their participation in urban community committees. This is an adaptation of the identification process, which only occurs in urban areas. In rural areas where social cohesion do exist, families are well known in the communities they belong to. 4.1. Programs Design Specificities In Benin’s PSDCC, about 10 percent of the selected household are in urban areas (1,187 out of the total of 13,000). The program makes use of community targeting committees (CTC) composed of members with good knowledge of the neighborhood and background in social assistance, charity work or identification of those in need. Members are elected democratically during neighborhood assemblies and committees. In rural areas, community targeting is facilitated by social cohesion whereas in urban areas, CTCs coordinate with local services providers (like women/youth associations) to help in the identification of households. Exclusion errors are likely because the poorest are often marginalized and inclusion errors tend to be higher because of a tendency to select people more connected to the local structures. Program administration needs to work closely with social workers to 8 revise initial listing to minimize leakage and under-coverage. Districts and villages were selected according to an index that combines poverty and access to services while poor households are selected through the implementation of a proxy means test (PMT) to identify the poorest households. Selected households receives FCFA 3,500 monthly as an unconditional transfers, and some households with at least one adult that can work can be selected for labor intensive public works. The public works is available for 12 days a month during 4 months at a daily rate of FCFA 800. Therefore transfers are either FCFA 42,000 or 80,400 a year (US$ 84 or US$ 140), which represents in per capita terms about 1.5 or 2.4 percent of the per capita GDP. In Congo, six of the seven localities selected for the pilot of the program Lisungi are urban. As the program aims to promote access of children to education and health, it was a requirement that selected areas had infrastructure facilities in which the supply of health care and primary schools was adequate for implementation. Moreover, the program is implemented in areas with well-established Social Welfare Centers (CAS). At the CAS, social workers are organized into small catchment areas (social sector) and work with the community committees to identify potential beneficiaries and organize local information campaigns. The CAS are also present in rural areas. Starting in 2011, Government invested in modernization of centers and CAS in urban areas had priority and are now fully functional. As of today about eight CAS are in place with the infrastructure needed for implementing the program nationwide. Due to the conditionalities imposed by the program, regions were selected according to the existing health and education infrastructure in order to guarantee implementation of the program. Extreme poor households, with at least one child 0-14 years old or one elderly aged 60 years old or more, are identified through a combination of community-based target and PMT. Selected households receives FCFA 10,000 monthly as an unconditional transfers, and each child within the household (for a cap of 3 per household) is eligible for FCFA 5,000 conditional of visiting health centers and attending school, depending on the age. Elderly defined as those aged 60 or above are eligible for an unconditional monthly transfer of FCFA 10,000. 9 Therefore transfers per household range between FCFA 120,000 and 540,000 a year (US$ 240 - 1,080), which represents in per capita terms 1.5 – 6.7 percent of per capita GDP. In Mali, just one urban community of the capital city Bamako, Bamako III, is eligible for the program. The program is decentralized and operated by different levels with different attributions: National (PMU), regional (focal point and M&E responsible), circles (M&E), communities committees, and villages (rural areas) or neighborhood (urban areas) committees. Although CTC’s members are supposed to work as volunteers in both urban and rural areas, they are often demanding some sort of payment as a result of the intense work to identify household. In urban areas the pressure for remunerate the CTC was more intense and often the CTC were not supporting program implementation activities such as following payment activities. Therefore, as a correction to the original design, Mali is introducing a unique payment scheme destined to support the work of CTCs. Notice that as one of the objectives of the program is to improve children’s outcomes in human capital (health, nutrition, hygiene and family savings a series of information sessions delivered by NGOs are organized and all household become beneficiaries of the program and are expected to participate in the sessions. Regions were classified according to an index combining poverty 1 and severe malnutrition 2 levels. Within each region, districts were ranked based on a weighted severe acute malnutrition index that combines poverty and infrastructure indicators and communes are selected based on geographical targeting using a combination of poverty maps and infrastructure indicators indices, which identify the potential availability of agencies for implementation of the accompanying measures and payments. All villages within each commune have a quota of eligible households, based on population size. Selected households receives FCFA 10,000 monthly as an unconditional transfers. The total annual transfer is FCFA 120,000 (US$ 240), which represents in per capita terms about 6 percent of per capita GDP. 1 Provided by the National Statistical Office 2 Unicef 10 V. Key Steps in the Delivery Process 5.1 Outreach & Intake In the urban areas of all three countries enumerators have difficulties returning to the households identified by the preliminary lists of households. Household information transmitted to the firms responsible for additional data collection (NSO in Congo and Mali and a private company in Benin) have issues due to errors in address recording (a lot of times, house numbering when available is done in a disorderly manner) and also because names recorded do not match household members’ names (it is common that people are known by their nicknames). Another problem in urban areas is the fact that household heads and/or reference adults are not found during the time of the survey for the reason that they are working or looking for a job. Therefore, enumerators need to visit households multiple times in order to collect needed data, increasing the amount of time needed for data collection. To minimize this issue, social workers are once more fundamental for the correct identification of households and subsequent interview scheduling (to minimize the number of visits to each household). In Benin, as part of the outreach, sensitization of national and local authorities is done through the use of training modules explaining the approach and defining concepts of poverty, extreme poverty, and household. The program administrators at the central level organize the training sessions, and they do take place in the morning followed immediately by the community committees’ field activities to identify potential beneficiaries. The lists of poor households produced by the committees are validated in public during neighborhood assemblies open to the public. Households in the final validated list are subjected to the PMT survey to verify their poverty status. As it is common in urban areas, families in Cotonou are not strongly associated as in smaller towns/ rural areas. Therefore, more time is required to identify poor households by the committees, and social workers from the centers for social promotion (CPS) are needed to revise the initial list to mitigate potential targeting errors, mainly in urban areas. Otherwise, no differences were observed between 11 urban and rural areas. Data collection uses smartphones without the need of connection to the Internet and data transfer to the network happens when Wi-Fi network was available. In Congo, 3 community committees are responsible for identification of poor and vulnerable households, which are further selected through household social survey and data collection. Identification is made by CAS and targeting community committee (CCS) on the basis of the existent indigence card (carte d’indigence), which contains information of population that often asks assistance at CAS. The outreach process is similar in both urban and rural areas but the challenge of communicating with potential beneficiaries and reaching the families pre-identified is problematic in urban areas because of the multiple communication channels available that minimize effectiveness of communication campaigns, and of lack of good address to locate the household in case social workers does not accompany surveyors and low social cohesion. In Mali, identification of households is performed by the members of the village committees based on the poverty criteria explained during technical workshop organized by the SSN Technical Management Unit (UTGFS). Identification methods varied according to the size of villages: (i) in the districts of Bamako, and in some rural towns where the population is sort of a census approach was used to review the measureable conditions of dwellings and make sure that all families would be identified. This work is done by the members of the village/ neighborhood committee where district/ village areas were divided in sectors and assigned to each member/ team of two members; and (ii) in small villages where people are known to each other and have daily interactions the village committee held meetings for exchange of information among its members, reviewed poorest households and established the preliminary beneficiaries list taking into account the quota allocated to the village/ neighborhood by the UTGFS. In small villages, the work of 3 The share of beneficiaries in each zone is defined by quotas given the population and poverty rates in each zone by the PMU that later organizes departmental workshops to promote cash transfers awareness. Those sessions are designed to city councils, sub-prefects and departmental directors. On a later stage, the PMU organizes training sessions to the leaders of Social Welfare District (CAS), CCS chiefs, and basic education inspectors. 12 identification and selection of poor households took between two and three days. In rural towns, it took two to three weeks. In Bamako, effort was made to achieve results after one month (deadline set by the program). It is important to note that field workers signaled this amount of weeks inadequate for collection of data in large areas where census could not be exhaustive. Despite aiming for having a census approach to map all households in urban areas, the activity did not reached all households in large neighborhoods. Therefore, it is likely that not all poor households have been identified and pre-registered. In order to fix this problem, community committees had to revise the analysis of the preliminary lists to make sure that the selection was not be solely done by the neighborhood committees. Communities’ involvement is then required to improve outreach but at an additional costs not considered by the program (communications, travel, meeting arrangements). 5.2 Selection Geo-targeting and PMT used for selection. In all three countries, a combination of geographical, poverty and social characteristic measurement is used to identify beneficiaries. Benin Selection follows a three-level selection process: (i) municipalities, (ii) villages and, (iii) households. Municipality selection: the program covers 12 municipalities with highest extreme poverty levels, 4 about 15 percent of the country's municipalities. Selection was performed during a national workshop and based on data provided by the National Institute of Statistics and Economic Analysis of Benin (INSAE), according to an index on access to basic infrastructure. Village selection is done in two stages: 4 Share of the number of people living below the food poverty line in a Municipality, relative to the country’s extreme poor population 13 a) Used of poverty maps (based on census and household survey data) 5 to estimate the distribution of poverty in small geographic areas (such as counties and villages or city neighborhoods). The targeting criteria will also be based on the villages’ share of extreme poverty to the municipalities. Each commune will establish a broad list of eligible villages. b) From the eligible list of villages, villages/ communities will be selected by the Municipal Councils. Following the selection of municipalities and poverty map analysis, a total of 120 villages were selected among the established lists of villages. Ultimately, the final number of participating villages varies by municipalities with regards to the resources allocated by the program. Household selection: in October 2013 Benin adopted a unique and rigorous method for the identification of the poor for social protection programs. This approach recommends a pre identification of the poorest by the communities, and further validation through the implementation of PMT to estimate households’ poverty levels. The PMT takes in consideration access to health, education and other services and was prepared by the national statistical office using 2011 household survey data. To account for differences between urban and rural areas, different models are in place and the eligibility cut-offs were defined differently, being higher in urban areas (adapted to purchase power parity for urban areas). The index developed is to be updated every two years. Congo Selection follows a three-level process: (i) municipalities, (ii) villages, and (iii) households. Regions selection: regions were selected according to the adequacy of infrastructure facilities to provide access to education and health for children, as those are conditionalities 5 2002 census data and 2011 Household Living Conditions survey (Enquête Modulaire Intégrée sur les Conditions de Vie des ménages - EMICoV). 14 imposed by the program. Furthermore, the program is implemented in areas with well- established CAS. Community Household: during project preparation in Congo, the National Statistics Office (INS) has worked with the PMU to assess the characteristics of poor and vulnerable households in the country on the basis of data from the Congolese household economic survey (ECOM). Based on this analysis, the INS has validated a PMT model that allows the PMU to estimate households’ per capita consumption for each household registered in the Unified Registry. The management information system (MIS) uses this linear combination to generate a list of potential beneficiaries per district by comparing the estimated welfare and food poverty line estimated by the proxy, which by the way have different models for urban and rural areas. Mali Selection follows a four-level selection process: (i) regions, (ii) circles and towns, (iii) municipalities, and (iv) households. Regions, circles and towns, and municipalities are classified according to synthetic indexes based on estimated indicators at the regional, circles and town and municipalities level. Three different indexes are used (one for each selection level) and, among other things, they comprise measures of poverty (provided by INSTAT), child severe acute malnutrition (SAM) 6 rates and access to services (water, electricity). Index range from 0 (rich) to 100 (very poor). All villages from selected municipalities are eligible to receive the program. The distribution of benefits is done at the community level. Each community will get a fixed budget to cover program expenses, based on the percentage of extreme poor provided by INSTAT. Quotas 6 UNICEF - normalized between 0 and 100 15 are calculated by the UTGFS team and evenly distributed in all villages, defined as the number of households divided by number of villages Household selection: Community/ village assemblies identify the poorest households at the village level. Each identified household will receive a score to allow household classification from the poorest to the least poor in an economic criterion (transparent and direct). Validation of beneficiaries list is done during neighborhood/ village general assemblies under the supervision members of the municipal committee and the M&E program officer. The preliminary list is read publicly and, in some cases, heads of households are introduced. Then the assembly is asked to rule. If there are no objections, the list is approved by acclamation and is a definitive list. If there are objections, complaints are registered and followed by immediate debate regarding eligibility. Households that are considered less poor than others are excluded from the final list and replaced by households considered poor. But public protests during general assemblies are rare, mainly because people are reluctant to speak in public on a sensitive issue. 5.3 Targeting/ Registration Generally, in urban areas registration was facilitated by the fact that people have identification documents. When it happened that registration was lower in urban than in rural areas, it was due to the higher urban mobility. The electronic data collection 7 in the urban areas of Benin increased data quality preventing data entry mistakes (for example, out-of-range values are) and facilitated the process of identifying beneficiary though pictures of household heads. Pictures were taken by the same device used for data collection and were used to issue beneficiary ID cards. The country has an ID card but coverage is limited. Therefore, the program has developed an alternate ID in coordination with the National Statistics and the National Health Insurance 7 GPS location of the household, roster, assets and standard living measures used by the PMT. 16 program. Once the household selections trough the PMT is finalized, the social worked delivers ID cards and compare with documents to guarantee delivery to the correct person for final registration in the program. Each Community Committee in Congo organizes an information campaign amongst pre- identified households. In Brazaville this process was more intense due to the lack of social cohesion which affects outreach. In rural areas, local radios, community leaders, social assistance centers reach a larger number of people at once. Records team, installed at registration centers request households to register the vouchers which were handed in during intake phase. At this stage, households have to present identity documents, birth certificates or any other documents required by the census strategy. Then, enumerators collect the necessary additional information, take photos and collect fingerprints of at least two members in the household. Field data is transferred to local and central servers by connecting laptops, used during registration, to local servers in each CAS. The processes are identical in both urban and rural areas but take more time in urban areas as a result of the lack of strong connections within families from the same community. Mali’s National Statistical Office (INSTAT) performed additional data collection on selected households if needed and subsequent program registration. INSTAT was responsible for instructing enumerators, training and fieldwork. During this step, household heads and a second adult member were photographed for purposes of identification and for provision of beneficiaries’ card to conclude the registration process. In Bamako, the percentage of households in the final beneficiaries list that were found to receive the registration card was slightly below than in other regions (97.7 percent against 99.8 percent and 99.3 percent in other regions). Replacement of households were then done with the help of village/ neighborhood committees. Household heads that were not found or absent were replaced by other members of household. 17 5.4 Payment/ Transactions Communication with families regarding payments in urban area is disadvantaged by the absence of an efficient communication method while in rural areas, radio/ church/ community leaders convey the message quickly and efficiently. Therefore, social workers and community committees need to spend some days in the field before payment to spread payment dates. Benin and Mali opted to deliver payments in urban areas through the use of SIM cards. But in Mali, the process is not entirely automated so beneficiaries have to go, in person, to respective payment centers, which incur some personal cost. As a developed structure for mobile payments is not yet available in Mali, payments accountability are entirely manual and can take up to three months. In the near future, with transactions changing to electronic, beneficiaries will not need to show up in order to receive cash transfer benefits at payment centers and payments can be paid monthly as in Benin. In contrast, Congo uses the same payment agency in both urban and rural areas, the Banque Postale du Congo, but beneficiaries must still go every three months to an agency instead of cashing the transfer’s monthly using ATMs cards. Benin uses distinct methods for payments in urban and rural areas. In Cotonou, cash benefits are transferred through SIM card electronic payment from MTN Bénin Telecommunications Service Provider. SIM cards are provided by the GSM operator while beneficiaries’ ID cards by program administrators. The GSM operator has a copy of the beneficiary database allowing for the monthly payment of all beneficiaries. Payments are done at any MTN operator in the region to receive the benefits, and beneficiaries knows that between 24 and 31 of the month they can demand payment at any MTN operator agency. As the transaction is electronic, accountability is fast, within seven days program administrators do have the accountability ready to be archived. In rural areas, payment is through branches of the Caisse Locale Credit Mutuel Agricol (CLCAM). Transactions are manual, therefore, accountability can last up to three months. In all areas of Benin, heads of 18 Social Promotion Centers have a copy of the beneficiaries’ database in the center’s catchment area and handles any complaints about missed payments. Congo pays all beneficiaries through the same bank system. One to two weeks before payment date, the PMU provides interested stakeholders (internal auditor, World Bank, payment agencies, etc.) and the payment agency, Banque Postal do Congo (BPC)., the nominal lists of beneficiaries together with the transfer amount to each household. Beneficiary households are informed by social workers, community committees…to be physically present in the associated BPC agency. Notice that BPC has created a virtual bank account and in fact beneficiaries can cash all or partial benefits by keeping it virtual account as a savings account. In urban areas there are more branches available therefore facilitating payment. In rural areas, payment centers by BPC are created to reduce the time and distance to payment points for beneficiaries. Once more, in rural areas accountability are manual, taking more time than in urban areas. Mali also uses distinct methods for payments in different areas. Following a study that was conducted to identify all agencies with capability to implement money transfers in all regions targeted by the program (banks and microfinance institutions to which include mobile operators and post offices), operators were pre-selected based on three criteria: (i) national coverage; (ii) capacity to do large-scale operations; and (iii) experience in cash transfers. Selected agencies differed regarding to their areas of operation. In the urban area (Bamako) the Orange Money mobile operator was selected whereas in all other areas, the tender winner was the National Bank for Agricultural Development (BNDA). In urban areas, Orange money create an account to all beneficiaries and provided free SIM cards. Even though SIM cards are used, payments are not electronically made yet. Consequently beneficiaries (or their substitutes) have to present themselves at payment centers every three months during payment time, which is informed by community committees, to withdraw the benefits upon presentation of their beneficiary card and identification card. The main difference between urban and rural areas is that the payment agency in rural areas, BNDA has more experience with payment transactions since it is a bank. As a result, 19 payment cards are developed by BNDA and include the recipient's name, identification code, date of birth and a column for observations, and payments are made following presentation of beneficiaries card and national ID cart (of beneficiary or his replacement. Given that BNDA has the expertise in providing payments, accountability is done on a timely manner whereas in the urban area, OM is inexperienced in such transactions increasing the time needed for accountability to the PMU. 5.5 Monitoring &Evaluation (M&E) M&E allows government and all partners to regularly monitor the project’s progress and to estimate program’s impact. To ensure the program is operational and implemented in compliance its objectives M&E is adopt various ways. In Benin and Congo, M&E agents visit households regularly while in Mali, agents work closely to community committees making sure program procedures are and evaluating problems and solving the issues when they appear. M&E agents in Benin visit households regularly in both urban and rural areas but the dispersion in urbans areas makes the process longer. GPS data collected during identification phase aides in the process since a lot of households do not have addresses or are not well documented. In Congo, M&E agents and social workers are in charge of following M&E activities and visiting households. The cooperation of health and education agents with social assistance is difficult due to the limited number of agents in each circumscription and the high number of people to be monitored. As a result a partnership with UNICEF is being formulated to support M&E activities. This issue is more relevant in urban areas due to the large catchment areas. M&E agents in Mali work closely with the municipal and village (rural areas) or neighborhood (urban areas) committees, the social development service and the payment agency. The M&E agent helps committees to understand program procedures through 20 regular monitoring work, evaluating problems and solving the issues. The M&E agents also help during village/ neighborhood sensitization sessions to inform beneficiaries about payment dates and place and are present during payment to monitor payment transactions, track problems if and when they appear and above all to make sure beneficiaries have all documents needed for payment. Moreover, in urban areas, monitoring is easily assisted by M&E agent and benefits from a greater presence of the civil society and partner NGOs physically located in the urban areas. For rural areas M&E agents are mostly alone to monitor activities. 5.6 Grievance & Redress The three countries have distinctive approaches to address grievance and redress. While in Congo information is collected electronically, in Benin and Mali social agents play an important part in the collection and transmission of the information to program administrators. In Benin, CPS have computers that are not connected to the network but have info of all beneficiaries but social workers are present to receive those who come to the CPS for demands and/or complaints. Following the personalized service provided by social workers, demands and complaints are delivered to program administrators via e-mail or telephone. In rural areas, as CPS are not present to centralize the collection of complaints, field agents collect complaints and demands and coordinate with program supervisors. Therefore, grievance and redress is less pronounced in rural areas. In Congo, grievance and redress is done through social worker and M&E agents located in each CAS. Information is digitalized in the system and processed automatically. There is no difference regarding the processes for urban and rural areas. In Mali, grievance and redress is not a well-established service. But the fact that NGOs are more present in urban areas, facilitates collection of demands and complaints and also 21 transmission of information to program administrators. Anyone can communicate with M&E agents, community committees and community leaders. VI. Key Issues Emerged during Program Implementation in Urban Areas Program implementation needs to be adapted to the difficult and challenging environment in urban areas. When program administrators organize general assemblies to present the program or to discuss a new feature or even to talk about results and monitoring activities, participation of authorities and the public is modest when compared to similar assemblies organized in rural areas. Possible explanations are the larger catchment area and weaker social cohesion meaning program administrators need to put more efforts and resources, both human and financial, to reach poor households for program participation. Inability to find and monitor families. Common characteristics of urban poverty are the lack of a proper household address and higher mobility of people. Given the uncoordinated growth in certain areas, households are often referred to/ identified by certain characteristics or “landmarks” instead of an address. Therefore, program workers and statistical office staff without good knowledge of the area may not be able to find the families. To escape or mitigate poverty, families are more likely to move (sometimes to areas not covered by the program) but they don’t have the habit to inform social workers or program administrators about their plans. When this happens before registration to the program is finalized administrators have difficulties in monitoring the family and offering the desired services because. Lack of interaction between community committees and social workers. When defining the plans for identifying program beneficiaries or potential beneficiaries, program administrators make active use of community committees. However, in urban areas, if social workers are not part of the identification/ selection process the outcomes of such committees could be biased, excluding potential beneficiaries from the program as the poorest are often marginalized within the community. Due to the nature of their work, 22 social workers have better knowledge of the population that often requires assistance. Therefore, enforcing participation of social workers in committees is essential but costs are high since in the studied countries there is limited availability of social workers to join committees. Communication is harder in urban areas. In rural areas, catchment areas are smaller and often communication channels are limited to one radio station or one meeting place (a religious site like a church or mosque). In contrast the number of communication channels available in urban areas requires a better outreach strategy in order to efficiently spread program messages like payment dates or information on conditionalities. Three other related issues are (i) Unawareness of program administrators regarding alternates to receive payment in the absence of the main beneficiary. The weak interaction between beneficiaries, who do not inform who the alternates are, and administrators prevent alternates to receive transfers; (ii) Lost SIM cards and/or program ID cards without which payments are not made. Beneficiaries do not inform community committees or social workers about lost documents. Therefore, on payment day, if either a social worker or a member of community committee member is not present at the payment to provide a replacement card / id, beneficiaries cannot receive the benefit; and (iii) Grievance mechanism are still under development therefore complaints can be left unanswered by program administrators. At times, social workers and community committees try to solve the issues themselves, without communicating to program administrators. VII. Urban Area Challenges Strengthen the ID system for security and to increase program’s credibility. Urban areas would benefit from better access to ID systems. Challenges arise from the limited coverage of national ID systems already in place. Countries in this study worked around this issue by 23 establishing a program ID system (like in Benin) or by making investments prior to the implementation of the programs (like in Congo and Mali). Falling motivation of local actors for outreach and follow-up. Volunteer work to help program implementation in urban areas is unappealing because of the larger catchment areas and population dispersion. Therefore, social workers engage in supporting program implementation besides their other duties. As it has proved to be a very labor-intensive task, motivation has decreased. One way to increase the willingness to help, could be the establishment of TORs, describing specific activities to be carried out in exchange for some sort of payment. Limited knowledge of household poverty state. Weaker social cohesion in urban areas, imply families known by committees and social workers are those who more frequently make demands and they might not be the poorest families. Poverty indicators as poverty headcount and poverty gap are not often available in small geographic areas as communities and most of poverty indicators used by communities can be very subjective and different because of different perceptions of poverty between community committees, in urban areas. When households have similar characteristic, it is difficult for committees to determine if one household is poorer than the other. Also, Data collection takes longer. Household intake and registration is an issue stemming from: (i) larger catchment areas; (ii) the lack of trust on agencies/agents collecting data, (iii) the bad quality of addresses recorded, and (iv) the use of nicknames to identify household members. Monitoring is hindered by higher mobility. To escape poverty, violence, lack of services and the cost of leaving, families in urban areas move around more frequently, sometimes before registration process is finished or even to an area without program coverage. Higher operation costs as enumerators perform multiple visits to households in order to correctly identify families and to follow-up. In Benin, although GPS data is collected during 24 identification part, the information is not shared with social workers for further household visits. Sharing GPS data with social would improve effectiveness of registration and M&E activities. Manual payments’ accountability takes too long. It can take up to three months for program administrators to obtain payments accountability preventing payments to be done more frequently and forcing beneficiaries to visit centers in order to receive their money. VIII. Emerging Policy & Operational Opportunities By looking at the experiences in Benin, Republic of Congo, and Mali, this study has identified five emerging policies and operational opportunities: 1. Household identification (location) can be improved through the use of GPS data. GPS location data is available in cell phones and tablets often used by programs administrators in urban areas and recorded coordinates should be used for improving accuracy, diminishing the problem of lack of a proper street address often observed in urban areas. It can also be used to facilitate during monitoring and evaluation (M&E) activities, including improved measures against fraud (as a way to ascertain the enumerator has visited the household). 2. A greater number of social workers is needed as well as setting a work program with solid terms of reference and remuneration to motivate neighborhood and community committees. It is also important to create/ strengthen partnerships with other agencies so as to train social workers and a few community members in order to improve the quality of identification and M&E capacity. 3. Create communication strategies adapted to urban areas. Therefore improving communication between program administrators and beneficiaries and reducing implementation issues such as refusal to pay alternates. 25 4. Implementation of automated/ electronic systems of payment will reduced accountability time and the possibility to provide monthly payments. Given the greater development of financial institutions in urban areas, programs should make use of the existing sizeable coverage of physical agencies and payment points to establish monthly electronic payment and open bank accounts to beneficiaries in order to increase financial literacy, therefore creating an automated payment accountability system and minimizing accountability processing time and the risk of fraud. 5. Adjustments in accordance with underlying differences and provision of multiple programs (bundles) to better protect and serve the population. Programs’ design should take in consideration the different costs of living in urban and rural areas and also differences regarding household composition. Urban poverty is more severe regarding not only monetary poverty but also other vulnerabilities (such as violence, weak health and education services, and limited job offers) that should all be taken into consideration so to better achieve programs’ objectives. 26 IX. References World Bank. 2015 (?). Mali – Program’s internal audit reports. World Bank. 2014. Evaluation de la pauvreté au Benin. Washington, DC ; World Bank Group. http://documents.worldbank.org/curated/en/2014/03/23797599/assessment -poverty-benin-evaluation-de-la-pauvrete-au-benin World Bank. 2014. Benin - Decentralized Community Driven Services Project : additional financing. Washington, DC ; World Bank Group. http://documents.worldbank.org/curated/en/2014/01/18925310/benin- decentralized-community-driven-services-project-additional-financing World Bank. 2014. Benin - Executive summary: poverty assessment. Washington, DC : World Bank Group. http://documents.worldbank.org/curated/en/2014/04/23791532/benin- executive-summary-poverty-assessment World Bank. 2013. Congo, Republic of - LISUNGI Safety Nets Project. Washington, DC ; World Bank Group. http://documents.worldbank.org/curated/en/2013/12/18794003/congo- republic-lisungi-safety-nets-project World Bank. 2013. PAD - Mali Emergency Safety Nets project. World Bank. http://wbdocs.worldbank.org/wbdocs/viewer/docViewer/indexEx.jsp?objectId =090224b081e5883c&respositoryId=WBDocs&standalone=false World Bank. 2011. Mali - Filets sociaux. Washington, DC: World Bank. http://documents.worldbank.org/curated/en/2011/01/14231097/mali-social- safety-nets-mali-filets-sociaux Programs’ websites: Benin’s PSDCC: http://www.psdcc.org/ Rep. of Congo’s Lisungi: http://www.masahs-gouv.net/lisungi.php Mali’s Jigiséméjiri: http://jigisemejiri.org/index.php/en/ 27 Annex 1: Benefit Levels Table A1. 1: Distribution of Benefits in Benin Adults Base Amount / household/ month Amount / household/ year transfer Amount (FCFA) (FCFA) Number (FCFA) … 3,500 3,500 42,000 FCFA 9,600, for 4 3,500, for 8 months 28,000 3,500 1 80,400 months 13,100, for 4 months 52,400 Table A1. 2: Distribution of Benefits in Mali Group 1 Amount / month / household Group 2 Core villages (FCFA) Interventions: Interventions: Yes CT CT 10,000 62,000 hhs AM Cash transfer AM NP beneficiaries Interventions: Interventions: No AM … AM NP Note1: AM, accompanying measures are not targeted; all households in the village are expected to participate in the sessions covering approximately 283,000 households. Note2: NP, nutritional package, is expected to cover 140,000 children in total from CT and non-CT beneficiaries Table A1. 3: Transfer Amount by Household Composition Children and/or pregnant woman Elderly Transfer amount / Base household / month transfer Amount Amount Number Number (FCFA) (FCFA) (FCFA) 1 5,000 15,000 2 10,000 20,000 3 or + 15,000 25,000 1 10,000 20,000 2 20,000 30,000 3 or + 30,000 40,000 1 5,000 1 10,000 25,000 FCFA 1 5,000 2 20,000 35,000 10,000 1 5,000 3 or + 30,000 45,000 2 10,000 1 10,000 30,000 2 10,000 2 20,000 40,000 2 10,000 3 or + 30,000 45,000 3 or + 15,000 1 10,000 35,000 3 or + 15,000 2 20,000 45,000 3 or + 15,000 3 or + 30,000 45,000 28 Annex 2: Three Countries Comparison Benin Congo, Republic Mali Program PSDCC Lisunji Jegiséméjiri Duration 24 months 2014-2018 2014-2017 Poor: 43.6% of population (2009) Poverty Poor: 36.2% of population Poor: 46.5% of population 25% of population suffers from chronic Assessment Extreme poor: 15% of population Extreme poor: 25% of population food insecurity 1. Establish an efficient SSN system to 1. Establish an efficient SSN system to protect the poor and vulnerable while V.I Establish an efficient SSN system to protect the poor and vulnerable while reducing poverty and smoothing protect the poor and vulnerable while reducing poverty and smoothing consumption Objectives reducing poverty and smoothing consumption. 2. Promote access of children to education consumption 2. Improve household’s ability to cope with and health and facilitate access of V.II Protect food insecure households shocks vulnerable population to productive inclusion measures 1,187 HHs in the main urban area: Cotonou 6,000 HHs in periurbans areas of Pointe- 916 HHs in the urban area of Bamako Size (13,000 total) Noire, Brazzaville and Cuvette) (62,000 HHs total) Occupation on the North by separatists and Lack of economic diversification (heavily Lack of economic diversification (heavily jihadists causing massive internal Nature of Urban dependent on the cotton trade, and dependent on oil) stimulating displacement, mainly to urban areas. More Poverty agriculture is the main source of income displacements to urban areas in search of recently, climate change and population for 70 percent of the country’s). jobs growth. 1. Establishment of admin pillars for a national SN program with laying the 1. Cash transfers to chronically poor and 1. Basic unconditional cash transfer to all foundations of a unified registry, MIS, and vulnerable households targeted HHs Program strengthening local capacities; 2. Accompanying measure to improve 2. Labor intensive public works for the Components same HHs to provide extra transfer during 2. Implementation of a cash transfer children’s outcomes in human capital program to improve access to health and (health, nutrition, hygiene and family agriculture lean season education services savings). 3. Project management and M&E 29 Benin Congo, Republic Mali Key program design features Households are eligible if deemed extreme HH are eligible if extreme poor, with at Households are eligible if deemed extreme Eligibility criteria poor by the community committee and least on child 0-14y.o. or an elderly aged 60 poor by the community committee below defined PMT threshold or more and below defined PMT threshold Health and education for HHs with children None, but all households from selected Conditionalities None (regular visits to health centers and 80% villages are expected to participate in the minimum attendance to school) accompaniment measures sessions The Ministry of Decentralization, Local The program has its own Project The program has its own Project Government, and Administration and Management Unit (PMU), the Lisungi team, Management Unit (PMU), the UTGFS team, Development of the Territory (MDGLAAT) under the Ministry of Social Affairs of under the Ministry of Economy and will be responsible for the overall Humanitarian Action and Solidarity finance. supervision and coordination of the (MASAHS). The UTGFS team manages all project Institutional Project. The Lisungi team manages all project activities, being responsible for all arrangements A new Secretariat for Decentralized activities, being responsible for all disbursements and procurement Community Driven Services (SSDCC) disbursements and procurement procedures supported by the project. created to take responsibility for procedures supported by the project The project coordinator oversees all implementing Components Two and Three The project coordinator oversees all activities related to the implementation of and for the fiduciary oversight of the activities related to the implementation of the project. Project. the project. Targeted communities were selected according to health and education care Targeted community of Bamako was offered (to comply with program selected according to high density that Response to implementation needs) and infrastructure PMT threshold is higher, adapted to leads to significant number of poor despite urban level (to implement accompanying purchase power parity for urban areas low poverty rates. specificities measures). Social workers revise the preliminary lists Social workers revise the preliminary lists formulated by community committees. formulated by community committees Urban poverty line used. 30 Benin Congo, Republic Mali The number of beneficiary households in each community follow a simple system of fixed quota per community defined by the Quotas for each social welfare sector, The share of beneficiaries in each zone is Beneficiary Project preparation team. The number of proportionally to poverty estimates at the defined in quotas given the population and quotas households in each community that will department level poverty rates in each zone. receive the basic cash transfer is estimated to be 100. Identification of households in the Sensitization of national and local Community committees responsible for preliminary beneficiaries list was authorities is done through the use of identification of poor and vulnerable performed by the members of the village Outreach & training modules explaining the approach households which will be selected through committees based on the poverty criteria Intake and defining concepts of poverty, extreme household social survey and data explained during technical workshop poverty, and household. collection. training organized by the SSN Technical Management Unit (UTGFS). Household selection is done through the The NSO has validated a PMT model that implementation of a proxy means test allows the PMU to estimate households’ (PMT) which takes in consideration access per capita consumption for each household Validation of beneficiaries list is always to health, education and other services. To registered in the Unified Registry. The done in neighborhood / village general account for differences between urban and Selection management information system (MIS) assemblies under the supervision members rural areas, different models are in place uses this linear combination to generate a of the municipal committee and the M&E and the eligibility cut-offs were defined list of potential beneficiaries per district by program officer. differently, being higher in urban areas. comparing the estimated welfare and food The index developed is to be updated poverty line estimated by the proxy. every two years. 31 Benin Congo, Republic Mali Mali’s National Statistical Office (INSTAT) performed additional data collection on Electronic data collection in the urban The processes are identical in both urban selected households if needed and areas increasing data quality and and rural areas but take more time in subsequent program registration. preventing data entry mistakes Pictures urban areas as a result of the lack of strong Household heads and a second adult were taken by the same device used for connections within families from the same member were photographed for purposes data collection and were used to issue community. Each Community Committee in of identification and for provision of beneficiary ID cards. Program has Congo organizes an information campaign beneficiaries’ card to conclude the Targeting / developed an alternate ID in coordination amongst pre-identified households. registration process. In Bamako, the Registration with the National Statistics and the Records team, installed at registration percentage of households in the final National Health Insurance program. Once centers request households to register the beneficiaries list that were found to receive the household selections trough the PMT is vouchers which were handed in during the registration card was slightly below finalized, the social worked delivers ID intake phase. Enumerators collect the than in other regions. Replacement of cards and compare with documents to necessary additional information, take households were then done with the help guarantee delivery to the correct person photos and collect fingerprints of at least of village / neighborhood committees. for final registration in the program. two members in the household. Household heads that were not found or absent were replaced by other members of household. In Cotonou, cash benefits are transferred All beneficiaries are paid through the same through SIM card electronic payment from bank system. Beneficiary households are In the urban area (Bamako) the Orange MTN Bénin Telecommunications Service informed by social workers, community Money mobile operator was selected Provider. SIM cards are provided by the committees to be physically present in the whereas in all other areas, the tender GSM operator while beneficiaries’ ID cards associated BPC agency. In urban areas winner was the National Bank for by program administrators. As the there are more branches available Agricultural Development (BNDA). Payment / transaction is electronic, accountability is therefore facilitating payment. In rural Given that BNDA has the expertise in Transactions fast, within 7 days program administrators areas, payment centers by BPC are created providing payments, accountability has do have the accountability ready to be to reduce the time and distance to been on time whereas in the urban area, archived. In rural areas, payment is payment points for beneficiaries. Once OM is inexperienced in such transactions through branches of the Caisse Locale more, in rural areas accountability are increasing the time needed for Credit Mutuel Agricol (CLCAM). Transaction manual, taking more time than in urban accountability to the PMU. are manual therefore accountability can areas. last up to three months. 32 Benin Congo, Republic Mali M&E agents in Mali work closely with the municipal and village (rural areas) or M&E agents and social workers are in neighborhood (urban areas) committees, charge of following M&E activities and the social development service and the visiting households. The cooperation of M&E agents visit households regularly in payment agency. The M&E agent helps health and education agents with social both urban and rural areas but the committees to understand program assistance is difficult due to the limited dispersion in urbans areas makes the procedures through regular monitoring number of agents in each circumscription process longer. GPS data collected during work, evaluating problems and solving the M&E and the high number of people to be identification phase aides in the process issues. The M&E agents also help during monitored. As a result a partnership with since a lot of households do not have village/neighborhood sensitization sessions UNICEF is being formulated to support addresses or are not well documented. to inform beneficiaries about payment M&E activities. This issue is more relevant dates and place and are present during in urban areas due to the large catchment payment to monitor payment transactions, areas. track problems if and when they appear and above all to make sure beneficiaries have all documents needed for payment. CPS have computers that are not connected to the network but have info of all beneficiaries but social workers are present to receive those who come to the Not a well stablished services. But the fact CPS for demands and/or complaints. Done through social worker and M&E that NGOs are more present in urban Following the personalized service agents located in each CAS. Information is areas, facilitates collection of demands and provided by social workers, demands and digitalized in the system and processed Grievance & complaints and also transmission of complaints are delivered to program automatically. There is no difference Redress administrators via e-mail or telephone. In regarding the processes for urban and rural information to program administrators. Anyone can communicate with M&E rural areas, as CPS are not present to areas. agents, community committees and centralize the collection of complaints, field community leaders. agents collect complaints and demands and coordinate with program supervisors. Therefore, grievance and redress I less pronounced in rural areas. 33 Social Protection & Labor Discussion Paper Series Titles 2014-2016 No. Title 1614 Urban Safety Nets – Experiences from Three Countries: Benin, Republic of Congo, and Mali by Vanessa Moreira and Ugo Gentilini, December 2016 1613 Issues for Civil Service Pension Reform in Sub-Saharan Africa by Anita M. Schwarz and Miglena Abels, November 2016 1612 How to Target Households in Adaptive Social Protection Systems? Relative Efficiency of Proxy Means Test and Household Economy Analysis in Niger by Pascale Schnitzer, October 2016 1611 Pensions for Public-Sector Employees: Lessons from OECD Countries’ Experience by Edward Whitehouse, October 2016 1610 Pension Systems in Sub-Saharan Africa: Brief Review of Design Parameters and Key Performance Indicators by Miglena Abels and Melis U. Guven, October 2016 1609 Household Enterprises in Fragile and Conflict-Affected States: Results from a Qualitative Toolkit Piloted in Liberia, Volume 2 – Annexes by Emily Weedon and Gwendolyn Heaner, August 2016 1608 Household Enterprises in Fragile and Conflict-Affected States: Results from a Qualitative Toolkit Piloted in Liberia, Volume 1 – Report by Emily Weedon and Gwendolyn Heaner, August 2016 1607 Benefits and Costs of Social Pensions in Sub-Saharan Africa by Melis U. Guven and Phillippe G. Leite, June 2016 1606 Assessing Benefit Portability for International Migrant Workers: A Review of the Germany- Turkey Bilateral Social Security Agreement by Robert Holzmann, Michael Fuchs, Seçil Paçacı Elitok and Pamela Dale, May 2016 1605 Do Bilateral Social Security Agreements Deliver on the Portability of Pensions and Health Care Benefits? A Summary Policy Paper on Four Migration Corridors Between EU and Non-EU Member States by Robert Holzmann, May 2016 1604 Assessing Benefit Portability for International Migrant Workers: A Review of the France-Morocco Bilateral Social Security Agreement by Robert Holzmann, Florence Legro and Pamela Dale, May 2016 1603 Assessing Benefit Portability for International Migrant Workers: A Review of the Belgium-Morocco Bilateral Social Security Agreement by Robert Holzmann, Jacques Wels and Pamela Dale, May 2016 1602 Assessing Benefit Portability for International Migrant Workers: A Review of the Austria-Turkey Bilateral Social Security Agreement by Robert Holzmann, Michael Fuchs, Seçil Paçaci Elitok and Pamela Dale, May 2016 1601 The Greek Pension Reform Strategy 2010-2016 by Georgios Symeonidis, July 2016 1507 Integrating Disaster Response and Climate Resilience in Social Protection Programs in the Pacific Island Countries by Cecilia Costella and Oleksiy Ivaschenko, September 2015 1506 Effectiveness of Targeting Mechanisms Utilized in Social Protection Programs in Bolivia by Ignacio Apella and Gastón Blanco, September 2015 1505 Kyrgyz Republic: Social Sectors at a Glance by João Pedro Azevedo, Paula Calvo, Minh Nguyen and Josefina Posadas, August 2015 1504 Entering the City: Emerging Evidence and Practices with Safety Nets in Urban Areas by Ugo Gentilini, July 2015 1503 Pension Patterns in Sub-Saharan Africa by Mark Dorfman, July 2015 1502 Social Protection in Fragile and Conflict-Affected Countries: Trends and Challenges by Mirey Ovadiya, Adea Kryeziu, Syeda Masood and Eric Zapatero, April 2015 1501 Defining, Measuring, and Benchmarking Administrative Expenditures of Mandatory Social Security Programs by Oleksiy Sluchynsky, February 2015 1425 Old-Age Financial Protection in Malaysia: Challenges and Options by Robert Holzmann, November 2014 1424 Profiling the Unemployed: A Review of OECD Experiences and Implications for Emerging Economies by Artan Loxha and Matteo Morgandi, August 2014 1423 Any Guarantees? An Analysis of China’s Rural Minimum Living Standard Guarantee Program by Jennifer Golan, Terry Sicular and Nithin Umapathi, August 2014 1422 World Bank Support for Social Safety Nets 2007-2013: A Review of Financing, Knowledge Services and Results by Colin Andrews, Adea Kryeziu and Dahye Seo, June 2014 1421 STEP Skills Measurement Surveys: Innovative Tools for Assessing Skills by Gaëlle Pierre, Maria Laura Sanchez Puerta, Alexandria Valerio and Tania Rajadel, July 2014 1420 Our Daily Bread: What is the Evidence on Comparing Cash versus Food Transfers? by Ugo Gentilini, July 2014 1419 Rwanda: Social Safety Net Assessment by Alex Kamurase, Emily Wylde, Stephen Hitimana and Anka Kitunzi, July 2012 1418 Niger: Food Security and Safety Nets by Jenny C. Aker, Carlo del Ninno, Paul A. Dorosh, Menno Mulder-Sibanda and Setareh Razmara, February 2009 1417 Benin: Les Filets Sociaux au Bénin Outil de Réduction de la Pauvreté par Andrea Borgarello et Damien Mededji, Mai 2011 1416 Madagascar Three Years into the Crisis: An Assessment of Vulnerability and Social Policies and Prospects for the Future by Philippe Auffret, May 2012 1415 Sudan Social Safety Net Assessment by Annika Kjellgren, Christina Jones-Pauly, Hadyiat El-Tayeb Alyn, Endashaw Tadesse and Andrea Vermehren, May 2014 1414 Tanzania Poverty, Growth, and Public Transfers: Options for a National Productive Safety Net Program by W. James Smith, September 2011 1413 Zambia: Using Social Safety Nets to Accelerate Poverty Reduction and Share Prosperity by Cornelia Tesliuc, W. James Smith and Musonda Rosemary Sunkutu, March 2013 1412 Mali Social Safety Nets by Cécile Cherrier, Carlo del Ninno and Setareh Razmara, January 2011 1411 Swaziland: Using Public Transfers to Reduce Extreme Poverty by Lorraine Blank, Emma Mistiaen and Jeanine Braithwaite, November 2012 1410 Togo: Towards a National Social Protection Policy and Strategy by Julie van Domelen, June 2012 1409 Lesotho: A Safety Net to End Extreme Poverty by W. James Smith, Emma Mistiaen, Melis Guven and Morabo Morojele, June 2013 1408 Mozambique Social Protection Assessment: Review of Social Assistance Programs and Social Protection Expenditures by Jose Silveiro Marques, October 2012 1407 Liberia: A Diagnostic of Social Protection by Andrea Borgarello, Laura Figazzolo and Emily Weedon, December 2011 1406 Sierra Leone Social Protection Assessment by José Silvério Marques, John Van Dyck, Suleiman Namara, Rita Costa and Sybil Bailor, June 2013 1405 Botswana Social Protection by Cornelia Tesliuc, José Silvério Marques, Lillian Mookodi, Jeanine Braithwaite, Siddarth Sharma and Dolly Ntseane, December 2013 1404 Cameroon Social Safety Nets by Carlo del Ninno and Kaleb Tamiru, June 2012 1403 Burkina Faso Social Safety Nets by Cécile Cherrier, Carlo del Ninno and Setareh Razmara, January 2011 1402 Social Insurance Reform in Jordan: Awareness and Perceptions of Employment Opportunities for Women by Stefanie Brodmann, Irene Jillson and Nahla Hassan, June 2014 1401 Social Assistance and Labor Market Programs in Latin America: Methodology and Key Findings from the Social Protection Database by Paula Cerutti, Anna Fruttero, Margaret Grosh, Silvana Kostenbaum, Maria Laura Oliveri, Claudia Rodriguez-Alas, Victoria Strokova, June 2014 To view Social Protection & Labor Discussion papers published prior to 2013, please visit www.worldbank.org/spl. Abstract As countries implement social safety nets programs, a range of technical hurdles affects their implementation differently in rural and urban areas. In urban areas, the focus of this study, living in is expensive and more vulnerable at economy slowdowns. Poverty can be more severe than in rural areas and accompanied by high malnutrition rates. Challenges faced by poor populations in most urban areas related to the lack of proper identification, outreach, intake and registration of potential beneficiaries in part due to the lack of social cohesion and the existence of multiple channels of communication, challenging the delivery of any messages quickly and efficiently. Therefore social workers have a fundamental part during program implementation and M&E process. About this series Social Protection & Labor Discussion Papers are published to communicate the results of The World Bank’s work to the development community with the least possible delay. This paper therefore has not been prepared in accordance with the procedures appropriate for formally edited texts. The findings, interpretations, and conclusions expressed herein are those of the author(s), and do not necessarily reflect the views of the International Bank for Reconstruction and Development/The World Bank and its affiliated organizations, or those of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. For more information, please contact the Social Protection Advisory Service, The World Bank, 1818 H Street, N.W., Room G7-803, Washington, DC 20433 USA. Telephone: (202) 458-5267, Fax: (202) 614-0471, E-mail: socialprotection@worldbank.org or visit us on-line at www.worldbank.org/spl. © 2016 International Bank for Reconstruction and Development / The World Bank